Building my first DRSSTC - Updated thread...
nzoomed, Tue Mar 24 2015, 04:42AMIm adding to this thread with construction photos and i will have the odd question or two since this is my first coil ive built :)
----------------------------------------------- -------------------------------
Hi everyone, im new here and i come from down under in New Zealand.
Ive got a few questions here and LoneOceans has been very much help to me, but ive got alot to read up on here.
Basically i have wanted to do a tesla coil for some time and now that ive got some spare time after completing my other projects i was inspired by highvoltagefeathers tesla coil playing "sail" on youtube.
I want to build a similar spec'd coil to his, i would probably class it as a medium sized coil, with sparks over a metre long im guessing.
His coil can be found here:
I want to work on similar specs, but if there is room for improvement i will improve on his.
Im looking at using:
Secondary
* 160mm diameter pipe
* 524mm length (according to wintesla)
* 2391 turns
* 1207m of 32AWG wire
* 101.45 KHz without toroid (according to what wintesla tells me)
* 150x600mm toroid? I may make this slightly smaller, but larger the better i believe for a DRSSTC if i can use a smaller toroid i would rather do so, but my aim is to have it running around 75 KHz
Primary
* 10 turns Quarter inch copper pipe flat Archimedes spiral
Inverter
* full bridge inverter using 2x CM300DY-12H IGBT's
Power supply
* 230v 0-270v VARIAC @8A with voltage doubler should put out up to 540v, but have been advised it could be closer to 650v.
Driver:
UD 2.7
Ok, so where do i start? lol
Basically i want to know if there is anything wrong with the above specs. My main concern is if 32 AWG is too fine for a medium sized coil, if more efficiency/power can be gained maybe i should go to the larger size of 30 AWG, as i dont think it will be much taller.
Im also unsure about the data provided in wintesla, everyone talks about JavaTC, but i cant get it to work because the values H1 and H2 for the secondary dont seem to be accepted, so i must be doing something wrong.
I want to work out if 524mm is the correct height for this much wire before cutting my pipe to the correct length.
With the primary im also wondering if 5/8" (8mm) pipe for the primary is better than quarter inch due to skin effect, it seems most people use this size. Im still unsure if i will need water cooling, but i see that this person is doing just that.
I understand that the radius of the inner primary winding should be 1 inch greater than that of the secondary, and that the spacings between each turn should be one quarter of an inch.
Are my IGBT's a good choice? Do i have to cut any wires inside these modules to disable current limiting?
With the power supply, im now uncertain of how much power such a coil will draw. Ive got a VARIAC, so i can always turn it down if it pops the circuit breakers, but basically in the USA he was running it on a 110v-200V VARIAC with a voltage doubler taking it up to 400V, his power supply was rated at 3000W.
We have 230V here so i will have a higher working voltage, so i expect that the wattage would be around 1500w.
My 8A VARIAC should handle a good 1800W, but i have been advised that my coil may be more powerful than this, if so i will look for a bigger VARIAC in the future to run full power.
My voltage doubler will be using 2x 4700uf capacitors rated at 450v, is this too high a capacitance, or is the larger the better? Im having alot of trouble sourcing these larger caps here, but i can get plenty at around 1500uf
I think this is all i can think of asking for now, but i will need more advice about choosing the right size toroids for the GDT's when ready, but i will be working on my primary and secondary coils for now before working on the rest of the electronics.
TIA for any advice :)
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Tue Mar 24 2015, 08:20AM
Your numbers look fine at a first glance, for a more in-depth analysis of your separate components like IGBTs and DC bus capacitors, you should take a look at the DRSSTC guide I am currently working on:
There is no exact numbers for primary to secondary spacing or primary winding to winding spacing, this is all a part of getting your desired coupling between the coils, use JAVATC to calculate this, the advised coupling from JAVATC works well for a DRSSTC.
I have fed my DRSSTC1, 60N60 miniblock bridge, from a 10A 230V variac and I blew the 10A fuse and the coil could easily eat more power at high break rates.
I would advise you to use at least 8 mm copper tubing, quarter inch seems too small for a CM300 coil. It is pretty much standard to use 10 mm in Europe.
Mads Barnkob, Tue Mar 24 2015, 08:20AM
Your numbers look fine at a first glance, for a more in-depth analysis of your separate components like IGBTs and DC bus capacitors, you should take a look at the DRSSTC guide I am currently working on:
There is no exact numbers for primary to secondary spacing or primary winding to winding spacing, this is all a part of getting your desired coupling between the coils, use JAVATC to calculate this, the advised coupling from JAVATC works well for a DRSSTC.
I have fed my DRSSTC1, 60N60 miniblock bridge, from a 10A 230V variac and I blew the 10A fuse and the coil could easily eat more power at high break rates.
I would advise you to use at least 8 mm copper tubing, quarter inch seems too small for a CM300 coil. It is pretty much standard to use 10 mm in Europe.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue Mar 24 2015, 09:56AM
Yes i feel that 8mm (5/16) pipe is the better option, i dont think i will need water cooling from what i gather.
I also have discovered that the IGBT modules i quoted are not suitable i need CM300DY-24h which are rated at 1200v
Im just wondering if the coil itself could be possibly too light using 32 AWG wire, would i be better of using 30 gauge?
I also thought this was a rather nifty bus layout here, ive never seen such a capacitor before, but it lends to a tidy construction.
Fullbridge inverter of cm300dy-12h IGBTs by HighVoltageFeathers, on Flickr
I wont be able to use such capacitors unless i can find one that is rated at a higher voltage of 650v or more, so i expect i will need to put 2 capacitors in series instead on a larger laminated bus.
I have seen what i did wrong in javaTC also.
I was working in mm instead of cm!
Anyway my resonant frequency is coming up to around 59.82 KHz which is actually lower than what i was expecting.
nzoomed, Tue Mar 24 2015, 09:56AM
Mads Barnkob wrote ...
Your numbers look fine at a first glance, for a more in-depth analysis of your separate components like IGBTs and DC bus capacitors, you should take a look at the DRSSTC guide I am currently working on:
There is no exact numbers for primary to secondary spacing or primary winding to winding spacing, this is all a part of getting your desired coupling between the coils, use JAVATC to calculate this, the advised coupling from JAVATC works well for a DRSSTC.
I have fed my DRSSTC1, 60N60 miniblock bridge, from a 10A 230V variac and I blew the 10A fuse and the coil could easily eat more power at high break rates.
I would advise you to use at least 8 mm copper tubing, quarter inch seems too small for a CM300 coil. It is pretty much standard to use 10 mm in Europe.
I have been flicking through your guide and its rather helpful thanks :)Your numbers look fine at a first glance, for a more in-depth analysis of your separate components like IGBTs and DC bus capacitors, you should take a look at the DRSSTC guide I am currently working on:
There is no exact numbers for primary to secondary spacing or primary winding to winding spacing, this is all a part of getting your desired coupling between the coils, use JAVATC to calculate this, the advised coupling from JAVATC works well for a DRSSTC.
I have fed my DRSSTC1, 60N60 miniblock bridge, from a 10A 230V variac and I blew the 10A fuse and the coil could easily eat more power at high break rates.
I would advise you to use at least 8 mm copper tubing, quarter inch seems too small for a CM300 coil. It is pretty much standard to use 10 mm in Europe.
Yes i feel that 8mm (5/16) pipe is the better option, i dont think i will need water cooling from what i gather.
I also have discovered that the IGBT modules i quoted are not suitable i need CM300DY-24h which are rated at 1200v
Im just wondering if the coil itself could be possibly too light using 32 AWG wire, would i be better of using 30 gauge?
I also thought this was a rather nifty bus layout here, ive never seen such a capacitor before, but it lends to a tidy construction.
Fullbridge inverter of cm300dy-12h IGBTs by HighVoltageFeathers, on FlickrI wont be able to use such capacitors unless i can find one that is rated at a higher voltage of 650v or more, so i expect i will need to put 2 capacitors in series instead on a larger laminated bus.
I have seen what i did wrong in javaTC also.
I was working in mm instead of cm!
Anyway my resonant frequency is coming up to around 59.82 KHz which is actually lower than what i was expecting.
Re: Building my first DRSSTC - Updated thread...
Uspring, Tue Mar 24 2015, 10:31AM
JavaTC seems to run fine with H1=10 and H2=63 (unit cm).
A full bridge of CM300DY-24h can easily handle more than 1500W of input power, probably twice as much.
Uspring, Tue Mar 24 2015, 10:31AM
JavaTC seems to run fine with H1=10 and H2=63 (unit cm).
A full bridge of CM300DY-24h can easily handle more than 1500W of input power, probably twice as much.
Re: Building my first DRSSTC - Updated thread...
dexter, Tue Mar 24 2015, 10:45AM
you can use CM300DY-12H's but not with a doubler
dexter, Tue Mar 24 2015, 10:45AM
you can use CM300DY-12H's but not with a doubler
Re: Building my first DRSSTC - Updated thread...
Hydron, Tue Mar 24 2015, 12:10PM
My coil has broadly similar specs (also very similar to loneocean's DRSSTC 3), and worked okay with 600V IGBTs and the primary circuit tuned for low Z, but does run better with 1200V IGBTs.
Off the top of my head specs of mine are:
160mm dia, 700mm winding length
~1900 turns 0.315mm wire
160mm x 740mm dryer duct toroid (rather fragile unfortunately)
11 turn 1/4" copper pipe primary
48x 2uF 1000V Aerovox snubbers for MMC (two parallel x 24 series strings, can adjust number in series to change tank capacitance)
CM300-24H full bridge with voltage doubler rectifier
UD2.5 driver with some mods
Have gotten strikes to nearly 2.5m without popping it, didn't get a chance to try for more before heading to europe.
Feel free to ask questions about NZ suppliers of stuff - I'd also offer you some surplus parts except that they're all packed away under my parent's house right now :(
Hydron, Tue Mar 24 2015, 12:10PM
My coil has broadly similar specs (also very similar to loneocean's DRSSTC 3), and worked okay with 600V IGBTs and the primary circuit tuned for low Z, but does run better with 1200V IGBTs.
Off the top of my head specs of mine are:
160mm dia, 700mm winding length
~1900 turns 0.315mm wire
160mm x 740mm dryer duct toroid (rather fragile unfortunately)
11 turn 1/4" copper pipe primary
48x 2uF 1000V Aerovox snubbers for MMC (two parallel x 24 series strings, can adjust number in series to change tank capacitance)
CM300-24H full bridge with voltage doubler rectifier
UD2.5 driver with some mods
Have gotten strikes to nearly 2.5m without popping it, didn't get a chance to try for more before heading to europe.
Feel free to ask questions about NZ suppliers of stuff - I'd also offer you some surplus parts except that they're all packed away under my parent's house right now :(
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue Mar 24 2015, 10:18PM
Thanks for your reply.
Yes if you know any good NZ suppliers, im keen to know, i really am after bus capacitors mainly, i dont think i will be able to get any suitable tank capacitors here in NZ unless you know of a source, i was going to get mine from eastern voltage research.
I will be using CM300DY-24H modules, i initally thought that all CM300 were rated at 1200v
My other question is really if i can get as large sparks using a finer winding on the secondary, if a heavier gauge wire will give me a higher output then i guess i may as well go this direction.
TIA
nzoomed, Tue Mar 24 2015, 10:18PM
Hydron wrote ...
My coil has broadly similar specs (also very similar to loneocean's DRSSTC 3), and worked okay with 600V IGBTs and the primary circuit tuned for low Z, but does run better with 1200V IGBTs.
Off the top of my head specs of mine are:
160mm dia, 700mm winding length
~1900 turns 0.315mm wire
160mm x 740mm dryer duct toroid (rather fragile unfortunately)
11 turn 1/4" copper pipe primary
48x 2uF 1000V Aerovox snubbers for MMC (two parallel x 24 series strings, can adjust number in series to change tank capacitance)
CM300-24H full bridge with voltage doubler rectifier
UD2.5 driver with some mods
Have gotten strikes to nearly 2.5m without popping it, didn't get a chance to try for more before heading to europe.
Feel free to ask questions about NZ suppliers of stuff - I'd also offer you some surplus parts except that they're all packed away under my parent's house right now :(
My coil has broadly similar specs (also very similar to loneocean's DRSSTC 3), and worked okay with 600V IGBTs and the primary circuit tuned for low Z, but does run better with 1200V IGBTs.
Off the top of my head specs of mine are:
160mm dia, 700mm winding length
~1900 turns 0.315mm wire
160mm x 740mm dryer duct toroid (rather fragile unfortunately)
11 turn 1/4" copper pipe primary
48x 2uF 1000V Aerovox snubbers for MMC (two parallel x 24 series strings, can adjust number in series to change tank capacitance)
CM300-24H full bridge with voltage doubler rectifier
UD2.5 driver with some mods
Have gotten strikes to nearly 2.5m without popping it, didn't get a chance to try for more before heading to europe.
Feel free to ask questions about NZ suppliers of stuff - I'd also offer you some surplus parts except that they're all packed away under my parent's house right now :(
Thanks for your reply.
Yes if you know any good NZ suppliers, im keen to know, i really am after bus capacitors mainly, i dont think i will be able to get any suitable tank capacitors here in NZ unless you know of a source, i was going to get mine from eastern voltage research.
I will be using CM300DY-24H modules, i initally thought that all CM300 were rated at 1200v
My other question is really if i can get as large sparks using a finer winding on the secondary, if a heavier gauge wire will give me a higher output then i guess i may as well go this direction.
TIA
Re: Building my first DRSSTC - Updated thread...
Hydron, Wed Mar 25 2015, 04:32PM
For tank cap I bought these:
I think I paid 120 USD for them (including shipping within the states) plus ~110 NZD them forwarded to NZ via youshop (big box made the forwarding pricey). Lisa at CTR surplus was helpful giving me an estimate of shipping size/weight before I ordered them.
For the DC bus cap I scored a bunch of RIFA 4700uF/450V electrolytics which were going to be scrapped (along with many IGBTs), so can't really help there.
Your secondary wire does sound a little finer than required, but I'm sure it would work. My secondary ended up about the same frequency as you're aiming at despite using thicker wire (due to the increased ~70cm winding length). There is apparently no magic number, just don't get the DC resistance or the frequency too high to reduce copper/switching losses respectively.
If you haven't already bought it, I think my wire came from CMG Electric Motors in Avondale, Auckland - went in there in person and they tried their best to find a part reel (only needed <1kg, full reels are ~3kg) and when they couldn't find one they gave me a good price on the full reel ($65ish). Again if I were home I could sell you what I have left, but it's too well packed away to tell others where to retrieve it from :(.
Hydron, Wed Mar 25 2015, 04:32PM
For tank cap I bought these:
I think I paid 120 USD for them (including shipping within the states) plus ~110 NZD them forwarded to NZ via youshop (big box made the forwarding pricey). Lisa at CTR surplus was helpful giving me an estimate of shipping size/weight before I ordered them.
For the DC bus cap I scored a bunch of RIFA 4700uF/450V electrolytics which were going to be scrapped (along with many IGBTs), so can't really help there.
Your secondary wire does sound a little finer than required, but I'm sure it would work. My secondary ended up about the same frequency as you're aiming at despite using thicker wire (due to the increased ~70cm winding length). There is apparently no magic number, just don't get the DC resistance or the frequency too high to reduce copper/switching losses respectively.
If you haven't already bought it, I think my wire came from CMG Electric Motors in Avondale, Auckland - went in there in person and they tried their best to find a part reel (only needed <1kg, full reels are ~3kg) and when they couldn't find one they gave me a good price on the full reel ($65ish). Again if I were home I could sell you what I have left, but it's too well packed away to tell others where to retrieve it from :(.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed Mar 25 2015, 09:07PM
1.5m sparks are still pretty decent, but im now thinking its better for me to take advantage of our 230v mains, use 1200v rated IGBT's and make the coil a bit taller to allow longer sparks. The price difference between 12H and 24H IGBT's is not much, so its silly not to make the coil more powerful since im spending this money.
I will be getting my tank capacitors from eastern voltage research, they are the cheapest place to get the famous CDE 942C20P15K-F series caps.
nzoomed, Wed Mar 25 2015, 09:07PM
Hydron wrote ...
For tank cap I bought these:
I think I paid 120 USD for them (including shipping within the states) plus ~110 NZD them forwarded to NZ via youshop (big box made the forwarding pricey). Lisa at CTR surplus was helpful giving me an estimate of shipping size/weight before I ordered them.
For the DC bus cap I scored a bunch of RIFA 4700uF/450V electrolytics which were going to be scrapped (along with many IGBTs), so can't really help there.
Your secondary wire does sound a little finer than required, but I'm sure it would work. My secondary ended up about the same frequency as you're aiming at despite using thicker wire (due to the increased ~70cm winding length). There is apparently no magic number, just don't get the DC resistance or the frequency too high to reduce copper/switching losses respectively.
If you haven't already bought it, I think my wire came from CMG Electric Motors in Avondale, Auckland - went in there in person and they tried their best to find a part reel (only needed <1kg, full reels are ~3kg) and when they couldn't find one they gave me a good price on the full reel ($65ish). Again if I were home I could sell you what I have left, but it's too well packed away to tell others where to retrieve it from :(.
Wire shouldnt be a problem for me, the person from Coils and Transformers here in Tauranga, has all the wire i need. My main issue ive discovered with the finer wire is mainly the short coil height i will have, since this coil was originally going to be run at around 400V bus voltage, it means that its output is much lower with sparks probably only around 1.5m, but running a higher bus voltage of around 650v will mean more powerful sparks, which i wont necessarily get the benefit from because the ground plane is closer to the toroid and most sparks will be striking the strike rail, so i dont think i will get as many long streamers that i want.For tank cap I bought these:
I think I paid 120 USD for them (including shipping within the states) plus ~110 NZD them forwarded to NZ via youshop (big box made the forwarding pricey). Lisa at CTR surplus was helpful giving me an estimate of shipping size/weight before I ordered them.
For the DC bus cap I scored a bunch of RIFA 4700uF/450V electrolytics which were going to be scrapped (along with many IGBTs), so can't really help there.
Your secondary wire does sound a little finer than required, but I'm sure it would work. My secondary ended up about the same frequency as you're aiming at despite using thicker wire (due to the increased ~70cm winding length). There is apparently no magic number, just don't get the DC resistance or the frequency too high to reduce copper/switching losses respectively.
If you haven't already bought it, I think my wire came from CMG Electric Motors in Avondale, Auckland - went in there in person and they tried their best to find a part reel (only needed <1kg, full reels are ~3kg) and when they couldn't find one they gave me a good price on the full reel ($65ish). Again if I were home I could sell you what I have left, but it's too well packed away to tell others where to retrieve it from :(.
1.5m sparks are still pretty decent, but im now thinking its better for me to take advantage of our 230v mains, use 1200v rated IGBT's and make the coil a bit taller to allow longer sparks. The price difference between 12H and 24H IGBT's is not much, so its silly not to make the coil more powerful since im spending this money.
I will be getting my tank capacitors from eastern voltage research, they are the cheapest place to get the famous CDE 942C20P15K-F series caps.
Re: Building my first DRSSTC - Updated thread...
Benjamin, Thu Mar 26 2015, 05:31AM
Hi, nzoomed, I just have a quick question. Where are you getting your ud 2.7 driver? Or are you building your own? Because if I ever build a big drsstc that's probably what I would want to use.
Benjamin, Thu Mar 26 2015, 05:31AM
Hi, nzoomed, I just have a quick question. Where are you getting your ud 2.7 driver? Or are you building your own? Because if I ever build a big drsstc that's probably what I would want to use.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Mar 26 2015, 09:42AM
nzoomed, Thu Mar 26 2015, 09:42AM
Benjamin wrote ...
Hi, nzoomed, I just have a quick question. Where are you getting your ud 2.7 driver? Or are you building your own? Because if I ever build a big drsstc that's probably what I would want to use.
I got mine from loneoceans, he has got a thread here on getting the boards, he has plenty available too :)Hi, nzoomed, I just have a quick question. Where are you getting your ud 2.7 driver? Or are you building your own? Because if I ever build a big drsstc that's probably what I would want to use.
Re: Building my first DRSSTC - Updated thread...
Hydron, Thu Mar 26 2015, 06:53PM
8 = Radius 1
8 = Radius 2
50 = Height 1
120 = Height 2
1920 = Turns
0.0315 = Wire Diameter
Also if the MMC cost looks too high, be aware that there are alternative to the 942C caps, especially for DRSSTCs with their lower peak currents than spark gap coils.
Hydron, Thu Mar 26 2015, 06:53PM
nzoomed wrote ...
Im also unsure about the data provided in wintesla, everyone talks about JavaTC, but i cant get it to work because the values H1 and H2 for the secondary dont seem to be accepted, so i must be doing something wrong.
JavaTC should work fine - was invaluable while designing my coil, here are the inputs I used for my secondary (metric mode, note use of cm rather than mm as a unit, and the inconsistent use of diameter vs radius)Im also unsure about the data provided in wintesla, everyone talks about JavaTC, but i cant get it to work because the values H1 and H2 for the secondary dont seem to be accepted, so i must be doing something wrong.
8 = Radius 1
8 = Radius 2
50 = Height 1
120 = Height 2
1920 = Turns
0.0315 = Wire Diameter
Also if the MMC cost looks too high, be aware that there are alternative to the 942C caps, especially for DRSSTCs with their lower peak currents than spark gap coils.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Mar 26 2015, 08:26PM
Yes i since found out i was using mm in JavaTC and thats where i went wrong! lol
If you can suggest any other caps for the MMC that are cheaper, that would be good, i did think they were alot higher spec, but your right, alot of these are higher grade for use in a SGTC.
I was trying to find some eurofarad caps, but they are probably very high spec also.
Ive been going through reworking the secondary specs and now im going with a slightly heavier wire, i will use 30 AWG and now the coil will be having a H/D aspect of 4.51 which i believe is a desired coil ratio.
The coil will now be 28.53 inches long with 2621 turns approx and an operating of 97.91 KHZ without toroid.
I should be able to keep the same sized toroid with this coil and hopefully this will significantly reduce the amount of flashover to the strike ring.
Here is the screenshot in JavaTC, if anything ive probably got alot better specs now.
nzoomed, Thu Mar 26 2015, 08:26PM
Hydron wrote ...
8 = Radius 1
8 = Radius 2
50 = Height 1
120 = Height 2
1920 = Turns
0.0315 = Wire Diameter
Also if the MMC cost looks too high, be aware that there are alternative to the 942C caps, especially for DRSSTCs with their lower peak currents than spark gap coils.
nzoomed wrote ...
Im also unsure about the data provided in wintesla, everyone talks about JavaTC, but i cant get it to work because the values H1 and H2 for the secondary dont seem to be accepted, so i must be doing something wrong.
JavaTC should work fine - was invaluable while designing my coil, here are the inputs I used for my secondary (metric mode, note use of cm rather than mm as a unit, and the inconsistent use of diameter vs radius)Im also unsure about the data provided in wintesla, everyone talks about JavaTC, but i cant get it to work because the values H1 and H2 for the secondary dont seem to be accepted, so i must be doing something wrong.
8 = Radius 1
8 = Radius 2
50 = Height 1
120 = Height 2
1920 = Turns
0.0315 = Wire Diameter
Also if the MMC cost looks too high, be aware that there are alternative to the 942C caps, especially for DRSSTCs with their lower peak currents than spark gap coils.
Yes i since found out i was using mm in JavaTC and thats where i went wrong! lol
If you can suggest any other caps for the MMC that are cheaper, that would be good, i did think they were alot higher spec, but your right, alot of these are higher grade for use in a SGTC.
I was trying to find some eurofarad caps, but they are probably very high spec also.
Ive been going through reworking the secondary specs and now im going with a slightly heavier wire, i will use 30 AWG and now the coil will be having a H/D aspect of 4.51 which i believe is a desired coil ratio.
The coil will now be 28.53 inches long with 2621 turns approx and an operating of 97.91 KHZ without toroid.
I should be able to keep the same sized toroid with this coil and hopefully this will significantly reduce the amount of flashover to the strike ring.
Here is the screenshot in JavaTC, if anything ive probably got alot better specs now.
Re: Building my first DRSSTC - Updated thread...
Hydron, Tue Mar 31 2015, 10:06AM
That's more like the numbers I had, which worked (in my opinion) rather well. Your DC resistance is about double what I had, which will increase the secondary losses a bit, but the increased turns will lower the IGBT switching speed requirements, so it's a tradeoff.
You'll probably want to raise the toroid position a bit in JavaTC too, which will probably pull the frequency down a little bit as well.
One other thing - try not to post twice in a row. It's frowned upon in this forum, so better to edit the earlier post to add stuff.
Hydron, Tue Mar 31 2015, 10:06AM
That's more like the numbers I had, which worked (in my opinion) rather well. Your DC resistance is about double what I had, which will increase the secondary losses a bit, but the increased turns will lower the IGBT switching speed requirements, so it's a tradeoff.
You'll probably want to raise the toroid position a bit in JavaTC too, which will probably pull the frequency down a little bit as well.
One other thing - try not to post twice in a row. It's frowned upon in this forum, so better to edit the earlier post to add stuff.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue Mar 31 2015, 10:34AM
Wound the coil today, phew it was a hard job to do it by hand but at least i had a couple of friends to help turn it and keep the windings even.
Wire snapped on my first attempt about 2 inches into the coil and had to start over.
Ready to coat with moisture curing polyurethane.
Hopefully i can get the drill to turn the lathe next time, we had issues with the drill adapter slipping.
nzoomed, Tue Mar 31 2015, 10:34AM
Wound the coil today, phew it was a hard job to do it by hand but at least i had a couple of friends to help turn it and keep the windings even.
Wire snapped on my first attempt about 2 inches into the coil and had to start over.
Ready to coat with moisture curing polyurethane.
Hopefully i can get the drill to turn the lathe next time, we had issues with the drill adapter slipping.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 05 2016, 10:33PM
Here are some of the latest construction photos.
The "misti" fan grills were 3D printed and are an obvious tribute to the manufacturer of the IGBTs that will power this coil!
They have LED backlit fans to light them up as shown in my avatar photo.
I added the primary coil last weekend and am still in the process of assembling the toroid, i still have to wire up everything to the UD board etc, but its almost done.
Cant wait to try it for the first time.
Couple of questions, ive been reading a bit on setting the phase lead, and it seems you dont need to run the coil at full bus voltage when scoping the bridge output.
Is this correct?
If so, this will make tuning my coil much simpler and safer and i wont need to worry about getting an expensive differential probe if i can just test it with an isolated 50VDC power supply.
I thought that the behaviour of the IGBT's may be different at higher voltages , but if this is not the case, thats good news. :)
Another question is on power connectors.
Im using an IEC C19 / C20 connector for the HT supply.
These are rated at 15A, so i though it should be sweet, however i see people using far more heavy duty power connectors on their coils of similar power to mine, so was wondering if i should use something a bit more beefy?
Thanks for your help, and please keep an eye on this thread, as i will have more questions to come thats for sure :)
nzoomed, Thu May 05 2016, 10:33PM
Here are some of the latest construction photos.
The "misti" fan grills were 3D printed and are an obvious tribute to the manufacturer of the IGBTs that will power this coil!
They have LED backlit fans to light them up as shown in my avatar photo.
I added the primary coil last weekend and am still in the process of assembling the toroid, i still have to wire up everything to the UD board etc, but its almost done.
Cant wait to try it for the first time.
Couple of questions, ive been reading a bit on setting the phase lead, and it seems you dont need to run the coil at full bus voltage when scoping the bridge output.
Is this correct?
If so, this will make tuning my coil much simpler and safer and i wont need to worry about getting an expensive differential probe if i can just test it with an isolated 50VDC power supply.
I thought that the behaviour of the IGBT's may be different at higher voltages , but if this is not the case, thats good news. :)
Another question is on power connectors.
Im using an IEC C19 / C20 connector for the HT supply.
These are rated at 15A, so i though it should be sweet, however i see people using far more heavy duty power connectors on their coils of similar power to mine, so was wondering if i should use something a bit more beefy?
Thanks for your help, and please keep an eye on this thread, as i will have more questions to come thats for sure :)
Re: Building my first DRSSTC - Updated thread...
Hydron, Fri May 06 2016, 08:42AM
IEC plug should be fine.
As for scoping the output you can just grab a cheap 100x probe - they should be rated at least 1kV and are handy to have anyway (they normally have low input capacitance, so are also useful for probing sensitive circuits).
Scope from earth to one bridge output, the waveform will be centered around 0V, but you should still be able to see what you need without investing in a differential probe, and you'll be able to run the coil up to realistic voltages as you do so.
For reference see this waveform, captured as I described:
Hydron, Fri May 06 2016, 08:42AM
IEC plug should be fine.
As for scoping the output you can just grab a cheap 100x probe - they should be rated at least 1kV and are handy to have anyway (they normally have low input capacitance, so are also useful for probing sensitive circuits).
Scope from earth to one bridge output, the waveform will be centered around 0V, but you should still be able to see what you need without investing in a differential probe, and you'll be able to run the coil up to realistic voltages as you do so.
For reference see this waveform, captured as I described:
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 06 2016, 12:11PM
OK ill look into getting one. I take it these are safe enough to protect the scope from high voltages?
Do i still need an isolation transformer? I would need a fairly high wattage unit if im tuning it at full power.
nzoomed, Fri May 06 2016, 12:11PM
Hydron wrote ...
IEC plug should be fine.
As for scoping the output you can just grab a cheap 100x probe - they should be rated at least 1kV and are handy to have anyway (they normally have low input capacitance, so are also useful for probing sensitive circuits).
Scope from earth to one bridge output, the waveform will be centered around 0V, but you should still be able to see what you need without investing in a differential probe, and you'll be able to run the coil up to realistic voltages as you do so.
For reference see this waveform, captured as I described:
IEC plug should be fine.
As for scoping the output you can just grab a cheap 100x probe - they should be rated at least 1kV and are handy to have anyway (they normally have low input capacitance, so are also useful for probing sensitive circuits).
Scope from earth to one bridge output, the waveform will be centered around 0V, but you should still be able to see what you need without investing in a differential probe, and you'll be able to run the coil up to realistic voltages as you do so.
For reference see this waveform, captured as I described:
OK ill look into getting one. I take it these are safe enough to protect the scope from high voltages?
Do i still need an isolation transformer? I would need a fairly high wattage unit if im tuning it at full power.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun May 08 2016, 12:24AM
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
nzoomed, Sun May 08 2016, 12:24AM
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
Re: Building my first DRSSTC - Updated thread...
loneoceans, Tue May 10 2016, 12:45AM
The gates of IGBTs used in coils are typically driven +15V. Datasheet maximums are +-20V, but coilers often overdrive the IGBTs. Having a greater gate voltage (e.g. +-24V) helps in a variety of ways such as forcing the IGBT into greater saturation, however at the risk of blowing out the gate, so YMMV.
If you're driving your IGBT at 19V, try the 1.5KE20CA or 22CA with a 20 or 22V clamp. These are bidirectional TVS diodes. Alternatively you can put back to back zener diodes of your desired voltage. The 1.5KE series come in 1.5KE_breakdownvolts_A/CA with the CA line being bidirectional (no polarity so they look like a diode with no white line). The breakdown voltage is the one you should be looking at. The clamping voltage is when the TVS becomes very conductive at its ~I_FSM rating (typically 00s of A), but typically just look at the breakdown voltage.
RS link example:
loneoceans, Tue May 10 2016, 12:45AM
nzoomed wrote ...
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
The gates of IGBTs used in coils are typically driven +15V. Datasheet maximums are +-20V, but coilers often overdrive the IGBTs. Having a greater gate voltage (e.g. +-24V) helps in a variety of ways such as forcing the IGBT into greater saturation, however at the risk of blowing out the gate, so YMMV.
If you're driving your IGBT at 19V, try the 1.5KE20CA or 22CA with a 20 or 22V clamp. These are bidirectional TVS diodes. Alternatively you can put back to back zener diodes of your desired voltage. The 1.5KE series come in 1.5KE_breakdownvolts_A/CA with the CA line being bidirectional (no polarity so they look like a diode with no white line). The breakdown voltage is the one you should be looking at. The clamping voltage is when the TVS becomes very conductive at its ~I_FSM rating (typically 00s of A), but typically just look at the breakdown voltage.
RS link example:
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Tue May 10 2016, 06:14AM
From my guide:
Maximum Gate-to-Emitter Voltage (VGE)
The gate voltage is limited by the thickness and characteristics of the gate oxide layer. Though the gate dielectric rupture is typically around 80 volts, the user is normally limited to 20 or 30V to limit current under fault conditions and to ensure long term reliability.
It is normal practice to drive IGBTs at +24VDC with the Steve Ward universal driver and clones of it that most use. The reason for the gate voltage limit is not so much for protecting the gate itself, it will first break down at some 80 Volt. Higher gate voltage means higher currents can be conducted through the Collector-Emitter. We take advantage of this by pushing the gate a little over its rated voltage to allow us to conduct higher currents through the IGBT, at the cost of higher switching losses. As explained by manufacturers in the following.
It is important to note however that IGBTs exhibit relatively high gain even at high gate-emitter voltage. This is because increasing the flow of electrons by increasing the gate-emitter voltage also increases the flow of holes. The gain of a high voltage power MOSFET however is very insensitive to gate voltage once fully on.
VCE(SAT) – Collector-Emitter On/Saturation Voltage
This is the collector-emitter voltage across the IGBT at a specified collector current, gate-emitter voltage, and junction temperature. Since VCE(sat) is temperature dependent, it is specified both at room temperature and hot.
From these graphs, a circuit designer can estimate conduction loss and the temperature coefficient of VCE(sat). Conduction power loss is VCE(sat) times collector current.
It is important to find a IGBT with as low a VCE(SAT) rating as possible. The conduction losses across the IGBT scales linearly with the VCE(SAT) voltage.
Collector-Emitter saturation voltage lowers dramatically at gate voltages above 12 Volt Gate-Emitter voltage, which is why a IGBT should never be driven with less than 15 Volt. Above 20 Volt the Collector-Emitter saturation voltage does not decline that much, but it is still lessened with higher gate voltage.
Mads Barnkob, Tue May 10 2016, 06:14AM
loneoceans wrote ...
The gates of IGBTs used in coils are typically driven +15V. Datasheet maximums are +-20V, but coilers often overdrive the IGBTs. Having a greater gate voltage (e.g. +-24V) helps in a variety of ways such as forcing the IGBT into greater saturation, however at the risk of blowing out the gate, so YMMV.
nzoomed wrote ...
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
Ive been working on the coil again this weekend and things are coming together nicely.
One question is regarding TVS Diodes.
Its supposed to be a good idea to put a TVS Clamp across each GDT output to protect the gates from any voltage spikes.
Im trying to find a suitable TVS diode on RS and the selection is huge.
What confuses me is that they say you need 30V TVS clamps, but i didnt think the gates of a CM300 could take much more than 20V?
I believe the voltage from the GDT is close to 19V so there is not much room for error.
When they say a 30V TVS clamp, is that supposed to be the minimum breakdown voltage or maximum breakdown voltage?
I would assume it would be the minimum.
Is bi-directional the same as bipolar? Its important that they are bipolar i believe.
TIA :)
The gates of IGBTs used in coils are typically driven +15V. Datasheet maximums are +-20V, but coilers often overdrive the IGBTs. Having a greater gate voltage (e.g. +-24V) helps in a variety of ways such as forcing the IGBT into greater saturation, however at the risk of blowing out the gate, so YMMV.
From my guide:
Maximum Gate-to-Emitter Voltage (VGE)
The gate voltage is limited by the thickness and characteristics of the gate oxide layer. Though the gate dielectric rupture is typically around 80 volts, the user is normally limited to 20 or 30V to limit current under fault conditions and to ensure long term reliability.
It is normal practice to drive IGBTs at +24VDC with the Steve Ward universal driver and clones of it that most use. The reason for the gate voltage limit is not so much for protecting the gate itself, it will first break down at some 80 Volt. Higher gate voltage means higher currents can be conducted through the Collector-Emitter. We take advantage of this by pushing the gate a little over its rated voltage to allow us to conduct higher currents through the IGBT, at the cost of higher switching losses. As explained by manufacturers in the following.
It is important to note however that IGBTs exhibit relatively high gain even at high gate-emitter voltage. This is because increasing the flow of electrons by increasing the gate-emitter voltage also increases the flow of holes. The gain of a high voltage power MOSFET however is very insensitive to gate voltage once fully on.
VCE(SAT) – Collector-Emitter On/Saturation Voltage
This is the collector-emitter voltage across the IGBT at a specified collector current, gate-emitter voltage, and junction temperature. Since VCE(sat) is temperature dependent, it is specified both at room temperature and hot.
From these graphs, a circuit designer can estimate conduction loss and the temperature coefficient of VCE(sat). Conduction power loss is VCE(sat) times collector current.
It is important to find a IGBT with as low a VCE(SAT) rating as possible. The conduction losses across the IGBT scales linearly with the VCE(SAT) voltage.
Collector-Emitter saturation voltage lowers dramatically at gate voltages above 12 Volt Gate-Emitter voltage, which is why a IGBT should never be driven with less than 15 Volt. Above 20 Volt the Collector-Emitter saturation voltage does not decline that much, but it is still lessened with higher gate voltage.
Re: Building my first DRSSTC - Updated thread...
loneoceans, Tue May 10 2016, 04:46PM
Just a quick note about the IGBT gate - it's a bit more nuanced that a '80V dielectric limit' so let me elaborate a bit more since you brought it up. The voltage across the gate delectric when above a certain threshold will start to lead to a tunneling current of carriers and creates heat. Eventually after a long period of time, this will cause damage to the oxide layer since the damage caused is cumulative. Increasing your gate drive voltage per se doesn't usually lead to this effect but it will increase the baseline of any transients in gate drive. In some datasheets you might also notice that VGE max is also specified with some duty cycle and T_pulse duration for this reason. As for increased switching losses, in a typical resonant drive for inverters used in DRSSTCs, again this is more nuanced but the conduction losses far outweigh typical switching losses, though this starts to get significant in regular non-teslacoil inverter design.
Traditionally IGBTs have been 'over-driven' well past their datasheet limits with the +-24V gate drive which increases saturation current as explained by Mads, and due to the relatively low duty cycle and cumulative run times of most coils, this has been quite reliable. However modern advances in IGBT technology have lead to dramatically increased current density (which is great for cost!), but allows less 'overhead' for over driving the transistors.
Bottom line - if you're not planning to drive your CM300DY-24H at 1500A (which has been done pretty reliably in the past!), a 24V gate drive isn't critical, and I wouldn't worry about a 18-20V gate drive which will be more than sufficient for say 800A of I_pri.
loneoceans, Tue May 10 2016, 04:46PM
Mads Barnkob wrote ...
From my guide:
Maximum Gate-to-Emitter Voltage (VGE)
The gate voltage is limited by the thickness and characteristics of the gate oxide layer. Though the gate dielectric rupture is typically around 80 volts, the user is normally limited to 20 or 30V to limit current under fault conditions and to ensure long term reliability.
It is normal practice to drive IGBTs at +24VDC with the Steve Ward universal driver and clones of it that most use. The reason for the gate voltage limit is not so much for protecting the gate itself, it will first break down at some 80 Volt. Higher gate voltage means higher currents can be conducted through the Collector-Emitter. We take advantage of this by pushing the gate a little over its rated voltage to allow us to conduct higher currents through the IGBT, at the cost of higher switching losses. As explained by manufacturers in the following.
It is important to note however that IGBTs exhibit relatively high gain even at high gate-emitter voltage. This is because increasing the flow of electrons by increasing the gate-emitter voltage also increases the flow of holes. The gain of a high voltage power MOSFET however is very insensitive to gate voltage once fully on.
VCE(SAT) – Collector-Emitter On/Saturation Voltage
This is the collector-emitter voltage across the IGBT at a specified collector current, gate-emitter voltage, and junction temperature. Since VCE(sat) is temperature dependent, it is specified both at room temperature and hot.
From these graphs, a circuit designer can estimate conduction loss and the temperature coefficient of VCE(sat). Conduction power loss is VCE(sat) times collector current.
It is important to find a IGBT with as low a VCE(SAT) rating as possible. The conduction losses across the IGBT scales linearly with the VCE(SAT) voltage.
Collector-Emitter saturation voltage lowers dramatically at gate voltages above 12 Volt Gate-Emitter voltage, which is why a IGBT should never be driven with less than 15 Volt. Above 20 Volt the Collector-Emitter saturation voltage does not decline that much, but it is still lessened with higher gate voltage.
From my guide:
Maximum Gate-to-Emitter Voltage (VGE)
The gate voltage is limited by the thickness and characteristics of the gate oxide layer. Though the gate dielectric rupture is typically around 80 volts, the user is normally limited to 20 or 30V to limit current under fault conditions and to ensure long term reliability.
It is normal practice to drive IGBTs at +24VDC with the Steve Ward universal driver and clones of it that most use. The reason for the gate voltage limit is not so much for protecting the gate itself, it will first break down at some 80 Volt. Higher gate voltage means higher currents can be conducted through the Collector-Emitter. We take advantage of this by pushing the gate a little over its rated voltage to allow us to conduct higher currents through the IGBT, at the cost of higher switching losses. As explained by manufacturers in the following.
It is important to note however that IGBTs exhibit relatively high gain even at high gate-emitter voltage. This is because increasing the flow of electrons by increasing the gate-emitter voltage also increases the flow of holes. The gain of a high voltage power MOSFET however is very insensitive to gate voltage once fully on.
VCE(SAT) – Collector-Emitter On/Saturation Voltage
This is the collector-emitter voltage across the IGBT at a specified collector current, gate-emitter voltage, and junction temperature. Since VCE(sat) is temperature dependent, it is specified both at room temperature and hot.
From these graphs, a circuit designer can estimate conduction loss and the temperature coefficient of VCE(sat). Conduction power loss is VCE(sat) times collector current.
It is important to find a IGBT with as low a VCE(SAT) rating as possible. The conduction losses across the IGBT scales linearly with the VCE(SAT) voltage.
Collector-Emitter saturation voltage lowers dramatically at gate voltages above 12 Volt Gate-Emitter voltage, which is why a IGBT should never be driven with less than 15 Volt. Above 20 Volt the Collector-Emitter saturation voltage does not decline that much, but it is still lessened with higher gate voltage.
Just a quick note about the IGBT gate - it's a bit more nuanced that a '80V dielectric limit' so let me elaborate a bit more since you brought it up. The voltage across the gate delectric when above a certain threshold will start to lead to a tunneling current of carriers and creates heat. Eventually after a long period of time, this will cause damage to the oxide layer since the damage caused is cumulative. Increasing your gate drive voltage per se doesn't usually lead to this effect but it will increase the baseline of any transients in gate drive. In some datasheets you might also notice that VGE max is also specified with some duty cycle and T_pulse duration for this reason. As for increased switching losses, in a typical resonant drive for inverters used in DRSSTCs, again this is more nuanced but the conduction losses far outweigh typical switching losses, though this starts to get significant in regular non-teslacoil inverter design.
Traditionally IGBTs have been 'over-driven' well past their datasheet limits with the +-24V gate drive which increases saturation current as explained by Mads, and due to the relatively low duty cycle and cumulative run times of most coils, this has been quite reliable. However modern advances in IGBT technology have lead to dramatically increased current density (which is great for cost!), but allows less 'overhead' for over driving the transistors.
Bottom line - if you're not planning to drive your CM300DY-24H at 1500A (which has been done pretty reliably in the past!), a 24V gate drive isn't critical, and I wouldn't worry about a 18-20V gate drive which will be more than sufficient for say 800A of I_pri.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue May 10 2016, 11:48PM
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
nzoomed, Tue May 10 2016, 11:48PM
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
Re: Building my first DRSSTC - Updated thread...
loneoceans, Wed May 11 2016, 12:28AM
Quick reply:
- You mentioned previously that you were driving at 19V. If you're driving at 24V, you should using something closer to 26V TVS. Or you can simply feed ~20V DC into the UD2 DC input which is easy if you're using a switching power supply (they usually have a small adjustable pot and many 24V supplies should go down to say 20 or 21V). If you're using the UD2.7, adjust the UVLO appropriately.
- You will get 24V drive from your GDT with 24V in assuming a 1:1 winding ratio. Alternatively you can do 15 turns on the primary and 12 on the secondaries to get around 19 without changing your PSU. Alternatively, a combination of both, or you can just use a 24V gate drive and different TVSes.
- Might be a good idea to increase your MMC in the future. Your MMC is only rated to 2.5kVAC. At 60kHz with your setup you will see 10kVpk across your MMC at 850A. You can calculate the Z of your cap and multiply by Ipk to get Vpk. I think in practice it'll work fine for now.
- There is no 'typical' peak overshoot value. It depends signficantly on your gate drive setup.
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
loneoceans, Wed May 11 2016, 12:28AM
nzoomed wrote ...
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
Quick reply:
- You mentioned previously that you were driving at 19V. If you're driving at 24V, you should using something closer to 26V TVS. Or you can simply feed ~20V DC into the UD2 DC input which is easy if you're using a switching power supply (they usually have a small adjustable pot and many 24V supplies should go down to say 20 or 21V). If you're using the UD2.7, adjust the UVLO appropriately.
- You will get 24V drive from your GDT with 24V in assuming a 1:1 winding ratio. Alternatively you can do 15 turns on the primary and 12 on the secondaries to get around 19 without changing your PSU. Alternatively, a combination of both, or you can just use a 24V gate drive and different TVSes.
- Might be a good idea to increase your MMC in the future. Your MMC is only rated to 2.5kVAC. At 60kHz with your setup you will see 10kVpk across your MMC at 850A. You can calculate the Z of your cap and multiply by Ipk to get Vpk. I think in practice it'll work fine for now.
- There is no 'typical' peak overshoot value. It depends signficantly on your gate drive setup.
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed May 11 2016, 01:19AM
Sorry, i probably confused you with my earlier post, I mentioned that i believed the GDT should put out 19V, but that was not my actual measurement, i assumed this was what i should be seeing going by the datasheet etc.
Yes my transformer is 1:1 so i must be driving at 24V, i thought i may have seen some losses through the transformer, but im forgetting that its current that i loose and not the voltage, feel so stupid right now!
Yes your right sorry, it was 850A limit for my MMC, I should be fine if i set my OCD to 800A. I will have to add a significant amount of capacitors if i want 1KA or more.
Do you have any sources for those eurofarad doorknob caps?
I cant find any on ebay, but i may just switch to those instead of upgrading, depending on how many CDE series i would need to add to my bank.
nzoomed, Wed May 11 2016, 01:19AM
loneoceans wrote ...
Quick reply:
- You mentioned previously that you were driving at 19V. If you're driving at 24V, you should using something closer to 26V TVS. Or you can simply feed ~20V DC into the UD2 DC input which is easy if you're using a switching power supply (they usually have a small adjustable pot and many 24V supplies should go down to say 20 or 21V). If you're using the UD2.7, adjust the UVLO appropriately.
- You will get 24V drive from your GDT with 24V in assuming a 1:1 winding ratio. Alternatively you can do 15 turns on the primary and 12 on the secondaries to get around 19 without changing your PSU. Alternatively, a combination of both, or you can just use a 24V gate drive and different TVSes.
- Might be a good idea to increase your MMC in the future. Your MMC is only rated to 2.5kVAC. At 60kHz with your setup you will see 10kVpk across your MMC at 850A. You can calculate the Z of your cap and multiply by Ipk to get Vpk. I think in practice it'll work fine for now.
- There is no 'typical' peak overshoot value. It depends signficantly on your gate drive setup.
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
nzoomed wrote ...
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
Ok thanks for that info.
I had ordered some P6KE22CA TVS diodes.
Clamping voltage is 30.6V and the breakdown voltage is 20.9V.
My UD is putting out about 24V at the GDT pins, so this is normal from what i understand.
I have not hooked up my GDT to the driver yet, but i expect i will see some losses anyway?
Loneoceans, what is your GDT putting out on your DRSSTC3? That coil is using CM300 series like im using on my coil.
MY GDT has 14 turns like most people have on theirs.
I take it that its not idea to be clamping the whole signal if the GDT voltage is within the TVS breakdown threshold?
I do expect to run the coil up to 850A eventually, i think from memory i calculated that my MMC could handle up to 1000ApK (CDE942 series, 224nF, 5 in series x 8 in parallel), but if i can potentially run the coil up to 1500ApK, then i may consider upgrading the MMC and replacing the bus capacitors.
Going by what mads is saying, if anywhere between 20-30V on the gates is OK, then i should be alright with the TVS ive got, if i had no TVS on the gates, what would be a typical peak transient voltage if i am running them at 24V?
On a side note, an electrical engineer told me that after looking at the UD schematic, "that there is no interlock delay in the circuit, therefore on every switching transition of the IGBT gates that will conduct simultaneously,
however designed this circuit has not appreciated what is required to safely switch IGBTs."
I told him that it probably has been deliberately designed this way, since we are pushing these things to their limits.
But perhaps it does not affect operation for DRSSTC use? IDK, but im interested to hear your thoughts on this, maybe it is indeed an improvement that can be made on a future revision?
Quick reply:
- You mentioned previously that you were driving at 19V. If you're driving at 24V, you should using something closer to 26V TVS. Or you can simply feed ~20V DC into the UD2 DC input which is easy if you're using a switching power supply (they usually have a small adjustable pot and many 24V supplies should go down to say 20 or 21V). If you're using the UD2.7, adjust the UVLO appropriately.
- You will get 24V drive from your GDT with 24V in assuming a 1:1 winding ratio. Alternatively you can do 15 turns on the primary and 12 on the secondaries to get around 19 without changing your PSU. Alternatively, a combination of both, or you can just use a 24V gate drive and different TVSes.
- Might be a good idea to increase your MMC in the future. Your MMC is only rated to 2.5kVAC. At 60kHz with your setup you will see 10kVpk across your MMC at 850A. You can calculate the Z of your cap and multiply by Ipk to get Vpk. I think in practice it'll work fine for now.
- There is no 'typical' peak overshoot value. It depends signficantly on your gate drive setup.
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
Sorry, i probably confused you with my earlier post, I mentioned that i believed the GDT should put out 19V, but that was not my actual measurement, i assumed this was what i should be seeing going by the datasheet etc.
Yes my transformer is 1:1 so i must be driving at 24V, i thought i may have seen some losses through the transformer, but im forgetting that its current that i loose and not the voltage, feel so stupid right now!
Yes your right sorry, it was 850A limit for my MMC, I should be fine if i set my OCD to 800A. I will have to add a significant amount of capacitors if i want 1KA or more.
Do you have any sources for those eurofarad doorknob caps?
I cant find any on ebay, but i may just switch to those instead of upgrading, depending on how many CDE series i would need to add to my bank.
Re: Building my first DRSSTC - Updated thread...
orac12, Wed May 11 2016, 09:04PM
What's this anti-parallel diode across the gate resistor? I don't think I have seen that in any schematics anywhere and so don't have them on mine. I feel like it's something I could benefit from!
orac12, Wed May 11 2016, 09:04PM
loneoceans wrote ...
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
- Your EE friend is absolutely correct, there is no dead-time setup in the UD2 and therefore it is essential to have fast anti-parallel diodes across the gate resistors to force the turn off to be faster than the turn-on. This is not reflected on the schematic, and in some cases, shoot-through can still be a possibility e.g. if your device is slow and your f is high. Though this is a fundamental thing to look out for in inverters. Adding dead time adjust adds quite a bit of complexity. Resonant drive also leads to a bit more nuances. Though in practice, this works ok if you know what you're doing, so it was left out in the UD2.
What's this anti-parallel diode across the gate resistor? I don't think I have seen that in any schematics anywhere and so don't have them on mine. I feel like it's something I could benefit from!
Re: Building my first DRSSTC - Updated thread...
Hydron, Wed May 11 2016, 09:23PM
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
Edit: realised I'm basically just repeating what was quoted. Another thing to remember is that the GTD type drive (all gates coupled via the transformer, pulls the gates negative as far as it goes positive) makes shoot-through less likely as it essentially forces the IGBTs half bridge gates to be driven exactly out of phase (only deviation is due to gate resistance, which is minimised with the anti-parallel diode)
Hydron, Wed May 11 2016, 09:23PM
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
Edit: realised I'm basically just repeating what was quoted. Another thing to remember is that the GTD type drive (all gates coupled via the transformer, pulls the gates negative as far as it goes positive) makes shoot-through less likely as it essentially forces the IGBTs half bridge gates to be driven exactly out of phase (only deviation is due to gate resistance, which is minimised with the anti-parallel diode)
Re: Building my first DRSSTC - Updated thread...
orac12, Wed May 11 2016, 09:26PM
Ah yep, makes sense. I want to try it, what would a recommended diode be?
orac12, Wed May 11 2016, 09:26PM
Hydron wrote ...
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
Ah yep, makes sense. I want to try it, what would a recommended diode be?
Re: Building my first DRSSTC - Updated thread...
Hydron, Wed May 11 2016, 09:29PM
schottkey or ultrafast diode, doesn't need to be very high current. 1N5819 or similar
Hydron, Wed May 11 2016, 09:29PM
schottkey or ultrafast diode, doesn't need to be very high current. 1N5819 or similar
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed May 11 2016, 10:48PM
I always thought those diodes were quite cruical along with the gate resistor.
Supposed to stop lazy switching on the IGBT's, shorts out the resistor during turn off as you mention.
Orac, here is Steve Ward's Schematic which shows the diodes mentioned
width=600
nzoomed, Wed May 11 2016, 10:48PM
Hydron wrote ...
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
Edit: realised I'm basically just repeating what was quoted. Another thing to remember is that the GTD type drive (all gates coupled via the transformer, pulls the gates negative as far as it goes positive) makes shoot-through less likely as it essentially forces the IGBTs half bridge gates to be driven exactly out of phase (only deviation is due to gate resistance, which is minimised with the anti-parallel diode)
The idea is to try and turn the IGBT off as quickly as possible (diode shorts out the gate resistor during turn-off), makes shoot-through less likely.
I'm not really sure how much difference it actually makes, has anyone got any experience running a coil with/without the diodes?
Edit: realised I'm basically just repeating what was quoted. Another thing to remember is that the GTD type drive (all gates coupled via the transformer, pulls the gates negative as far as it goes positive) makes shoot-through less likely as it essentially forces the IGBTs half bridge gates to be driven exactly out of phase (only deviation is due to gate resistance, which is minimised with the anti-parallel diode)
I always thought those diodes were quite cruical along with the gate resistor.
Supposed to stop lazy switching on the IGBT's, shorts out the resistor during turn off as you mention.
Orac, here is Steve Ward's Schematic which shows the diodes mentioned
width=600Re: Building my first DRSSTC - Updated thread...
orac12, Thu May 12 2016, 12:32AM
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
orac12, Thu May 12 2016, 12:32AM
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 12 2016, 12:51AM
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
nzoomed, Thu May 12 2016, 12:51AM
orac12 wrote ...
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
Re: Building my first DRSSTC - Updated thread...
orac12, Thu May 12 2016, 01:00AM
Have had the coil up and running decently and making 2 meter streamers so far at a reduced bus voltage. Primary to secondary flash overs are stopping me from turning it up any further at the moment.
orac12, Thu May 12 2016, 01:00AM
nzoomed wrote ...
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
orac12 wrote ...
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
Have had the coil up and running decently and making 2 meter streamers so far at a reduced bus voltage. Primary to secondary flash overs are stopping me from turning it up any further at the moment.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 12 2016, 01:15AM
Thats great news!
If you are still having flashover issues, i would try raising the coil as you mention and perhaps put some acrylic sheet over the primary.
nzoomed, Thu May 12 2016, 01:15AM
orac12 wrote ...
Have had the coil up and running decently and making 2 meter streamers so far at a reduced bus voltage. Primary to secondary flash overs are stopping me from turning it up any further at the moment.
nzoomed wrote ...
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
orac12 wrote ...
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
Well I haven't blown anything up with out them, but I'll install them and hopefully reduce some of my switching spikes.
You should get more effective switching with them, and may help to resolve your issues with your coils output possibly.
Have had the coil up and running decently and making 2 meter streamers so far at a reduced bus voltage. Primary to secondary flash overs are stopping me from turning it up any further at the moment.
Thats great news!
If you are still having flashover issues, i would try raising the coil as you mention and perhaps put some acrylic sheet over the primary.
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Thu May 12 2016, 05:38AM
Flashover from primary to secondary sound more like a extremely high coupling, or the coil could be badly out of tune.
Mads Barnkob, Thu May 12 2016, 05:38AM
Flashover from primary to secondary sound more like a extremely high coupling, or the coil could be badly out of tune.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 12 2016, 11:18PM
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
nzoomed, Thu May 12 2016, 11:18PM
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Re: Building my first DRSSTC - Updated thread...
loneoceans, Fri May 13 2016, 12:45AM
Those are fine. Just lower your 24V input to the UD2 to 22V.
loneoceans, Fri May 13 2016, 12:45AM
nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Those are fine. Just lower your 24V input to the UD2 to 22V.
Re: Building my first DRSSTC - Updated thread...
orac12, Fri May 13 2016, 02:24AM
I used some 1N5361 back to back, which is around 27V.
orac12, Fri May 13 2016, 02:24AM
nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
I used some 1N5361 back to back, which is around 27V.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 13 2016, 03:29AM
nzoomed, Fri May 13 2016, 03:29AM
loneoceans wrote ...
Those are fine. Just lower your 24V input to the UD2 to 22V.
I cant easily do that as im using a transformer, i could rectify it and use a voltage regulator i guess to feed 22VDC into the DC input.nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Those are fine. Just lower your 24V input to the UD2 to 22V.
Re: Building my first DRSSTC - Updated thread...
Justin, Fri May 13 2016, 04:12AM
You might try these.
Justin, Fri May 13 2016, 04:12AM
nzoomed wrote ...
loneoceans wrote ...
Those are fine. Just lower your 24V input to the UD2 to 22V.
I cant easily do that as im using a transformer, i could rectify it and use a voltage regulator i guess to feed 22VDC into the DC input.nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Those are fine. Just lower your 24V input to the UD2 to 22V.
You might try these.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 13 2016, 04:30AM
I see those on RS also, but their clamping voltage is rather high at 53V.
If 53V is OK, then I should be OK with mine rated at 35V
nzoomed, Fri May 13 2016, 04:30AM
Justin wrote ...
You might try these.
nzoomed wrote ...
loneoceans wrote ...
I cant easily do that as im using a transformer, i could rectify it and use a voltage regulator i guess to feed 22VDC into the DC input.nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Those are fine. Just lower your 24V input to the UD2 to 22V.Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
You might try these.
I see those on RS also, but their clamping voltage is rather high at 53V.
If 53V is OK, then I should be OK with mine rated at 35V
Re: Building my first DRSSTC - Updated thread...
Justin, Fri May 13 2016, 05:08AM
Working Voltage: 25.6 V
Breakdown Voltage: 28.5 V
Clamping Voltage: 41.4 V
I don't see 53V anywhere. It breaks down at 28.5 Volts.
Justin, Fri May 13 2016, 05:08AM
Working Voltage: 25.6 V
Breakdown Voltage: 28.5 V
Clamping Voltage: 41.4 V
I don't see 53V anywhere. It breaks down at 28.5 Volts.
Re: Building my first DRSSTC - Updated thread...
orac12, Fri May 13 2016, 05:18AM
I ran these back to back.
orac12, Fri May 13 2016, 05:18AM
nzoomed wrote ...
I see those on RS also, but their clamping voltage is rather high at 53V.
If 53V is OK, then I should be OK with mine rated at 35V
Justin wrote ...
You might try these.
nzoomed wrote ...
loneoceans wrote ...
I cant easily do that as im using a transformer, i could rectify it and use a voltage regulator i guess to feed 22VDC into the DC input.nzoomed wrote ...
Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
Those are fine. Just lower your 24V input to the UD2 to 22V.Im still having trouble finding suitable TVS diodes for 24V gate operation.
The closest ive got are P6KE22CA, these have a breakdown voltage start at 24.4v and the maximum clamping voltage is 35V.
I really need them to clamp at around 26 or 27V ideally, but all the TVS i see have a far lower breakdown voltage, that means it will breakdown well under 24V.
You might try these.
I see those on RS also, but their clamping voltage is rather high at 53V.
If 53V is OK, then I should be OK with mine rated at 35V
I ran these back to back.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 13 2016, 07:22AM
RS tells me the clamping voltage is 53.5V
I take it that clamping voltage is important, since its fully conducting even though it breaks down much lower?
If i run my gates at 24V, and the breakdown voltage on my diodes starts at 24.4V, is this OK?
You obviously dont want them clamping the whole signal.
nzoomed, Fri May 13 2016, 07:22AM
Justin wrote ...
Working Voltage: 25.6 V
Breakdown Voltage: 28.5 V
Clamping Voltage: 41.4 V
I don't see 53V anywhere. It breaks down at 28.5 Volts.
Working Voltage: 25.6 V
Breakdown Voltage: 28.5 V
Clamping Voltage: 41.4 V
I don't see 53V anywhere. It breaks down at 28.5 Volts.
RS tells me the clamping voltage is 53.5V
I take it that clamping voltage is important, since its fully conducting even though it breaks down much lower?
If i run my gates at 24V, and the breakdown voltage on my diodes starts at 24.4V, is this OK?
You obviously dont want them clamping the whole signal.
Re: Building my first DRSSTC - Updated thread...
Hydron, Fri May 13 2016, 08:56AM
from my PM on this topic:
The 35V figure is when they are passing their rated current. They'll do the job, though you should really be using ones rated at 24V rather than 22V (see link:,4294866063 )
BTW the gates won't instantly pop with excess voltage so i wouldn't worry too much.
(yes i should have qualified "excess" - I believe others have had gates up at the 35V level without instant failure. Obviously crazy voltages will kill them instantly)
Hydron, Fri May 13 2016, 08:56AM
from my PM on this topic:
The 35V figure is when they are passing their rated current. They'll do the job, though you should really be using ones rated at 24V rather than 22V (see link:
,4294866063 )BTW the gates won't instantly pop with excess voltage so i wouldn't worry too much.
(yes i should have qualified "excess" - I believe others have had gates up at the 35V level without instant failure. Obviously crazy voltages will kill them instantly)
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 13 2016, 11:23AM
Thanks for that info.
I just got your message also :)
I have not had much to do with TVS diodes before, so ive been learning alot!
Anyway, ive been testing my driver board again, and ive found that i am indeed getting a signal on all fout GDT pins, but the second pin on each output is weaker, and both are not a square wave.
I think its not helping that i dont have a 10x probe on my scope when probing the IC pins, but ive traced it back to IC5 and the scope is showing a strong signal on pin6, but very faint on the scope on pin3 with little or no signal showing (when i get hold of a digital scope this weekend i should have better readings hopefully).
The signal on these pins is similar pattern to what i posted on the screenshot earlier.
So my main concern is im not getting a square wave.
All the MOSFETS are working fine.
I thought i would test it under load with the GDT connected to see what i got on the scope, I assume this noise from the GDT and UD is completley normal from what ive read? Its quite loud, and it changes pitch when i adjust the beats on the MIDI controller.
Its also very vague what the jumper does (SV1) next to the tunable inductor.
From what i understand and going by the schematic, its designed to bypass the inductor?
I dont get any signal output from the UD if this jumper is set to pins 1-2, pins 2-3 give me a signal output.
My signal generator is injecting a 50KHz signal into the CT input.
These are the scope shots i took from the GDT connected to the UD, its far from the square wave im supposed to be getting, but hopefully this will point me into the right direction on whats wrong.
this is 10v/div
This is 5v/div
this is 0.2v/div
nzoomed, Fri May 13 2016, 11:23AM
Hydron wrote ...
from my PM on this topic:
The 35V figure is when they are passing their rated current. They'll do the job, though you should really be using ones rated at 24V rather than 22V (see link:,4294866063 )
BTW the gates won't instantly pop with excess voltage so i wouldn't worry too much.
(yes i should have qualified "excess" - I believe others have had gates up at the 35V level without instant failure. Obviously crazy voltages will kill them instantly)
from my PM on this topic:
The 35V figure is when they are passing their rated current. They'll do the job, though you should really be using ones rated at 24V rather than 22V (see link:
,4294866063 )BTW the gates won't instantly pop with excess voltage so i wouldn't worry too much.
(yes i should have qualified "excess" - I believe others have had gates up at the 35V level without instant failure. Obviously crazy voltages will kill them instantly)
Thanks for that info.
I just got your message also :)
I have not had much to do with TVS diodes before, so ive been learning alot!
Anyway, ive been testing my driver board again, and ive found that i am indeed getting a signal on all fout GDT pins, but the second pin on each output is weaker, and both are not a square wave.
I think its not helping that i dont have a 10x probe on my scope when probing the IC pins, but ive traced it back to IC5 and the scope is showing a strong signal on pin6, but very faint on the scope on pin3 with little or no signal showing (when i get hold of a digital scope this weekend i should have better readings hopefully).
The signal on these pins is similar pattern to what i posted on the screenshot earlier.
So my main concern is im not getting a square wave.
All the MOSFETS are working fine.
I thought i would test it under load with the GDT connected to see what i got on the scope, I assume this noise from the GDT and UD is completley normal from what ive read? Its quite loud, and it changes pitch when i adjust the beats on the MIDI controller.
Its also very vague what the jumper does (SV1) next to the tunable inductor.
From what i understand and going by the schematic, its designed to bypass the inductor?
I dont get any signal output from the UD if this jumper is set to pins 1-2, pins 2-3 give me a signal output.
My signal generator is injecting a 50KHz signal into the CT input.
These are the scope shots i took from the GDT connected to the UD, its far from the square wave im supposed to be getting, but hopefully this will point me into the right direction on whats wrong.
this is 10v/div
This is 5v/div
this is 0.2v/div
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue May 17 2016, 12:09PM
Just one more question.
Does it matter what mode the MIDI interrupter is set to when testing?
Ive got it set to spark gap mode so its putting out a constant signal to the UD.
My other concern is if my signal generator is not working properly.
I was also told to ground pin 6 of the optical receiver, as this should allow the driver to put out a square wave without the MIDI interrupter connected, I tried this, and I got no signal.
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
TIA
nzoomed, Tue May 17 2016, 12:09PM
Just one more question.
Does it matter what mode the MIDI interrupter is set to when testing?
Ive got it set to spark gap mode so its putting out a constant signal to the UD.
My other concern is if my signal generator is not working properly.
I was also told to ground pin 6 of the optical receiver, as this should allow the driver to put out a square wave without the MIDI interrupter connected, I tried this, and I got no signal.
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
TIA
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 26 2016, 10:40AM
Im still getting nowhere with my U.D board and am running out of ideas.
Ive got hold of a decent signal generator and scope now and I get a clean square wave on all four MOSFET outputs if i connect pin 6 to ground on the optical receiver.
With the GDT connected, i get the same clean square wave, albeit at a higher voltage for whatever reason.
The GDT used to make alot of noise, but it now no longer does also, which is some concern to me, as i expect that its not getting enough current, yet the voltage is 24VAC at the MOSFET outputs.
If i disconnect this pin from ground and hook up the MIDI interrupter, i no longer get any square wave on the second pair of pins and the GDT output is similar to the scope shots ive posted earlier. When changing the BPS etc on the interrupter, the gaps between these spikes change accordingly, but thats it.
Im suspecting that its harmonics that is giving me a reading of about 60-80v, but ive tested the transformer and its putting out a perfect 1:1 ratio when i test it on the scope with a signal generator to the input.
TIA for any suggestions, this is driving me crazy right now, im so close to getting this darn coil working!
nzoomed, Thu May 26 2016, 10:40AM
Im still getting nowhere with my U.D board and am running out of ideas.
Ive got hold of a decent signal generator and scope now and I get a clean square wave on all four MOSFET outputs if i connect pin 6 to ground on the optical receiver.
With the GDT connected, i get the same clean square wave, albeit at a higher voltage for whatever reason.
The GDT used to make alot of noise, but it now no longer does also, which is some concern to me, as i expect that its not getting enough current, yet the voltage is 24VAC at the MOSFET outputs.
If i disconnect this pin from ground and hook up the MIDI interrupter, i no longer get any square wave on the second pair of pins and the GDT output is similar to the scope shots ive posted earlier. When changing the BPS etc on the interrupter, the gaps between these spikes change accordingly, but thats it.
Im suspecting that its harmonics that is giving me a reading of about 60-80v, but ive tested the transformer and its putting out a perfect 1:1 ratio when i test it on the scope with a signal generator to the input.
TIA for any suggestions, this is driving me crazy right now, im so close to getting this darn coil working!
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Thu May 26 2016, 08:00PM
I had some problems with a recent UD2.7C build. The design and board is fine, but had some faulty components. A diode and IC had to be replaced. These things happen. Are you using an ST fiber receiver? If so did you remember to remove D13? Does the signal LED come on when you fire up the interrupter? It should. If it's not, follow the signal until you lose it and then check the components on that part of the board. This requires some debugging skills but no better way to learn how the UD driver works.
If you are getting an input signal but no U1-U4 square wave output, track backwards until you find the input signal and again find suspect components. A good sig gen and scope will be essential. 2 things that really helped me 1) Learn and understand the UD circuit, 2) Follow LoneOceans testing and debugging notes !
Graham Armitage, Thu May 26 2016, 08:00PM
nzoomed wrote ...
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
I had some problems with a recent UD2.7C build. The design and board is fine, but had some faulty components. A diode and IC had to be replaced. These things happen. Are you using an ST fiber receiver? If so did you remember to remove D13? Does the signal LED come on when you fire up the interrupter? It should. If it's not, follow the signal until you lose it and then check the components on that part of the board. This requires some debugging skills but no better way to learn how the UD driver works.
If you are getting an input signal but no U1-U4 square wave output, track backwards until you find the input signal and again find suspect components. A good sig gen and scope will be essential. 2 things that really helped me 1) Learn and understand the UD circuit, 2) Follow LoneOceans testing and debugging notes !
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu May 26 2016, 10:23PM
OK thanks for that info.
Yes the signal LED does come on and flash at the same time as the LED on my interrupter.
Ive had a friend who is an electrical engineer help me over the phone and he has explained to me in good detail how it works.
He told me what pins to check for signals on the IC's etc, and said that if you ground pin 6 of the optical receiver (yes im using the ST connector, D13 is removed) you should get a perfect square wave, which i am, but perhaps there is a problem with one of the IC's ad=s you say.
Who was your supplier? I got everything from mouser. What Parts were at fault with you?
I might check those, as i do wonder if an IC is at fault.
nzoomed, Thu May 26 2016, 10:23PM
Graham Armitage wrote ...
I had some problems with a recent UD2.7C build. The design and board is fine, but had some faulty components. A diode and IC had to be replaced. These things happen. Are you using an ST fiber receiver? If so did you remember to remove D13? Does the signal LED come on when you fire up the interrupter? It should. If it's not, follow the signal until you lose it and then check the components on that part of the board. This requires some debugging skills but no better way to learn how the UD driver works.
If you are getting an input signal but no U1-U4 square wave output, track backwards until you find the input signal and again find suspect components. A good sig gen and scope will be essential. 2 things that really helped me 1) Learn and understand the UD circuit, 2) Follow LoneOceans testing and debugging notes !
nzoomed wrote ...
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
All the IC voltages appear to be correct on the rails etc. but one thing i see a fault with is D12, im getting +5 on the anode, which is normal, but im getting nearly 5v on the cathode side (about 4.8v) does this mean the diode is breaking down and letting current pass through?
I had some problems with a recent UD2.7C build. The design and board is fine, but had some faulty components. A diode and IC had to be replaced. These things happen. Are you using an ST fiber receiver? If so did you remember to remove D13? Does the signal LED come on when you fire up the interrupter? It should. If it's not, follow the signal until you lose it and then check the components on that part of the board. This requires some debugging skills but no better way to learn how the UD driver works.
If you are getting an input signal but no U1-U4 square wave output, track backwards until you find the input signal and again find suspect components. A good sig gen and scope will be essential. 2 things that really helped me 1) Learn and understand the UD circuit, 2) Follow LoneOceans testing and debugging notes !
OK thanks for that info.
Yes the signal LED does come on and flash at the same time as the LED on my interrupter.
Ive had a friend who is an electrical engineer help me over the phone and he has explained to me in good detail how it works.
He told me what pins to check for signals on the IC's etc, and said that if you ground pin 6 of the optical receiver (yes im using the ST connector, D13 is removed) you should get a perfect square wave, which i am, but perhaps there is a problem with one of the IC's ad=s you say.
Who was your supplier? I got everything from mouser. What Parts were at fault with you?
I might check those, as i do wonder if an IC is at fault.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Thu May 26 2016, 10:41PM
Sounds like you're making progress. I also got all the parts from mouser, but it's very possible I killed the components :( I had to replace IC 9 as it was not producing any output. I cannot recall which diode, but it was toast - essentially a short circuit. I probably cooked it when soldering. By following the signal from IC to IC I was able to find the general area of where things stopped working. Then by probing all components in between I located the offender.
Graham Armitage, Thu May 26 2016, 10:41PM
Sounds like you're making progress. I also got all the parts from mouser, but it's very possible I killed the components :( I had to replace IC 9 as it was not producing any output. I cannot recall which diode, but it was toast - essentially a short circuit. I probably cooked it when soldering. By following the signal from IC to IC I was able to find the general area of where things stopped working. Then by probing all components in between I located the offender.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 27 2016, 12:13AM
It seems to be the path from the optical receiver that is the problem i wonder.
All four MOSFET outputs puts out a square wave of about 24V on the scope, but this is something ive just noticed and i will try and post some more screenshots soon.
Ive noticed that the second output on each pair is above 0V.
The first output on each pair will put out a square wave with 0V in the centre (+/-12VAC on each side of 0V) but the second pin on each output will give me a square wave above the 0V mark (between 0V and +24V)
If that makes sense to you, 0V is where the trace line will sit with no signal.
nzoomed, Fri May 27 2016, 12:13AM
Graham Armitage wrote ...
Sounds like you're making progress. I also got all the parts from mouser, but it's very possible I killed the components :( I had to replace IC 9 as it was not producing any output. I cannot recall which diode, but it was toast - essentially a short circuit. I probably cooked it when soldering. By following the signal from IC to IC I was able to find the general area of where things stopped working. Then by probing all components in between I located the offender.
Sounds like you're making progress. I also got all the parts from mouser, but it's very possible I killed the components :( I had to replace IC 9 as it was not producing any output. I cannot recall which diode, but it was toast - essentially a short circuit. I probably cooked it when soldering. By following the signal from IC to IC I was able to find the general area of where things stopped working. Then by probing all components in between I located the offender.
It seems to be the path from the optical receiver that is the problem i wonder.
All four MOSFET outputs puts out a square wave of about 24V on the scope, but this is something ive just noticed and i will try and post some more screenshots soon.
Ive noticed that the second output on each pair is above 0V.
The first output on each pair will put out a square wave with 0V in the centre (+/-12VAC on each side of 0V) but the second pin on each output will give me a square wave above the 0V mark (between 0V and +24V)
If that makes sense to you, 0V is where the trace line will sit with no signal.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Fri May 27 2016, 04:08AM
What do the outputs look like on IC 9 and 10 (Pin 5 and 7)? Is there any DC offset or does that look good? How do those outputs compare to the U1-4 outputs? Check C30 as I believe this should eliminate DC bias and maybe what you are seeing if it's not working. I am no expert on this driver - just trying to share some similar debugging experiences I have gone through in the past.
Graham Armitage, Fri May 27 2016, 04:08AM
What do the outputs look like on IC 9 and 10 (Pin 5 and 7)? Is there any DC offset or does that look good? How do those outputs compare to the U1-4 outputs? Check C30 as I believe this should eliminate DC bias and maybe what you are seeing if it's not working. I am no expert on this driver - just trying to share some similar debugging experiences I have gone through in the past.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri May 27 2016, 07:00AM
I think ive found the issue and my friend who is an electrical engineer pointed this out to me, but i was not overly concerned until he told me this, but...
Basically all four MOSFET outputs should put out the same identical square wave, except that the second MOSFET on each pair inverts it.
What im getting on the output of U2 and U4 seems to be a shift up about 12V higher than the outputs from U1 and U3.
So if you take the trace line at 0V, the square wave is perfectly balanced on U1 and U3 with +/-12V either side of the 0V mark on the grid.
With the output on U2 and U4, im getting the same amplitude square wave, only this time its 12V higher sitting between 0V and +24V peak to peak, and he tells me this is wrong and they should all be the same as U1 and U3.
Any ideas?
nzoomed, Fri May 27 2016, 07:00AM
Graham Armitage wrote ...
What do the outputs look like on IC 9 and 10 (Pin 5 and 7)? Is there any DC offset or does that look good? How do those outputs compare to the U1-4 outputs? Check C30 as I believe this should eliminate DC bias and maybe what you are seeing if it's not working. I am no expert on this driver - just trying to share some similar debugging experiences I have gone through in the past.
What do the outputs look like on IC 9 and 10 (Pin 5 and 7)? Is there any DC offset or does that look good? How do those outputs compare to the U1-4 outputs? Check C30 as I believe this should eliminate DC bias and maybe what you are seeing if it's not working. I am no expert on this driver - just trying to share some similar debugging experiences I have gone through in the past.
I think ive found the issue and my friend who is an electrical engineer pointed this out to me, but i was not overly concerned until he told me this, but...
Basically all four MOSFET outputs should put out the same identical square wave, except that the second MOSFET on each pair inverts it.
What im getting on the output of U2 and U4 seems to be a shift up about 12V higher than the outputs from U1 and U3.
So if you take the trace line at 0V, the square wave is perfectly balanced on U1 and U3 with +/-12V either side of the 0V mark on the grid.
With the output on U2 and U4, im getting the same amplitude square wave, only this time its 12V higher sitting between 0V and +24V peak to peak, and he tells me this is wrong and they should all be the same as U1 and U3.
Any ideas?
Re: Building my first DRSSTC - Updated thread...
nzoomed, Tue May 31 2016, 10:55PM
Got my UD all working properly now it appears.
However ive got alot of ringing on the GDT.
Loneoceans told me it should be much less than what ive currently got.
This should drop when connected to the gates of my IGBT's, but i believe i can get it down if i shorten the leads.
How long is too long?
Both ends for the primary and secondary are 300 long.
If i have to shorten them, I will have to find somewhere else to mount the GDT other than the heatsink.
Below is my screenshot from the scope.
TIA :)
nzoomed, Tue May 31 2016, 10:55PM
Got my UD all working properly now it appears.
However ive got alot of ringing on the GDT.
Loneoceans told me it should be much less than what ive currently got.
This should drop when connected to the gates of my IGBT's, but i believe i can get it down if i shorten the leads.
How long is too long?
Both ends for the primary and secondary are 300 long.
If i have to shorten them, I will have to find somewhere else to mount the GDT other than the heatsink.
Below is my screenshot from the scope.
TIA :)
Re: Building my first DRSSTC - Updated thread...
Justin, Wed Jun 01 2016, 05:28AM
I wouldn't worry about it until you've got it connected, if you're really concerned you can always add a clamp on ferrite bead afterwards.
Justin, Wed Jun 01 2016, 05:28AM
I wouldn't worry about it until you've got it connected, if you're really concerned you can always add a clamp on ferrite bead afterwards.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed Jun 01 2016, 08:01AM
Here is a magnified scope shot across the secondary without being connected to the IGBT.
And here is another scope shot across the secondary when its connected to the gate.
How does this compare with others scope readings?
nzoomed, Wed Jun 01 2016, 08:01AM
Justin wrote ...
I wouldn't worry about it until you've got it connected, if you're really concerned you can always add a clamp on ferrite bead afterwards.
I wouldn't worry about it until you've got it connected, if you're really concerned you can always add a clamp on ferrite bead afterwards.
Here is a magnified scope shot across the secondary without being connected to the IGBT.
And here is another scope shot across the secondary when its connected to the gate.
How does this compare with others scope readings?
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sat Jun 04 2016, 04:44AM
sorry, here it is.
nzoomed, Sat Jun 04 2016, 04:44AM
sorry, here it is.
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Sat Jun 04 2016, 06:53AM
That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Mads Barnkob, Sat Jun 04 2016, 06:53AM
That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sat Jun 04 2016, 10:51AM
Im using 5.1 ohm currently for the gate resistors.
How high can you typically go? I take it that overshoot is undesired, but i expect its difficult to remove completley?
nzoomed, Sat Jun 04 2016, 10:51AM
Mads Barnkob wrote ...
That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Yes loneoceans also confirmed to me that its normal, so i hope i can test everything out tomorrow with a UPS transformer and adjust the phase lead initially at low voltage, and hope to perhaps see some short sparks from the secondary.That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Im using 5.1 ohm currently for the gate resistors.
How high can you typically go? I take it that overshoot is undesired, but i expect its difficult to remove completley?
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Sat Jun 04 2016, 12:14PM
It is a trade-off between larger amplitude ringing with lower value gate resistors and higher switching losses with higher gate resistors. There is no optimal value, it is all up to your design criteria, most industrial designs only have one key parameter, temperature. We do not really have that as a limit, so adjust gate resistors for best possible inverter switching output, you do not want to switch too fast or too slow, just right in-between.
Mads Barnkob, Sat Jun 04 2016, 12:14PM
nzoomed wrote ...
Im using 5.1 ohm currently for the gate resistors.
How high can you typically go? I take it that overshoot is undesired, but i expect its difficult to remove completley?
Mads Barnkob wrote ...
That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Yes loneoceans also confirmed to me that its normal, so i hope i can test everything out tomorrow with a UPS transformer and adjust the phase lead initially at low voltage, and hope to perhaps see some short sparks from the secondary.That looks perfectly fine, running a transformer unloaded will ofc create overshoot and oscillations in a much higher frequency as it oscillates with the parasitic capacitance of your probes.
You could experiment with different gate resistor values to get a faster rising edge or dampen the overshoot even more.
Im using 5.1 ohm currently for the gate resistors.
How high can you typically go? I take it that overshoot is undesired, but i expect its difficult to remove completley?
It is a trade-off between larger amplitude ringing with lower value gate resistors and higher switching losses with higher gate resistors. There is no optimal value, it is all up to your design criteria, most industrial designs only have one key parameter, temperature. We do not really have that as a limit, so adjust gate resistors for best possible inverter switching output, you do not want to switch too fast or too slow, just right in-between.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 05 2016, 01:46AM
Thanks for that info.
At present im concerned about my tuning the coil.
I thought 10 turns on the primary would have been plenty, JavaTC gives me a resonant frequency of around 53.78 KHz for the primary and 58.22 for the secondary.
If i click on auto tune on Java TC, it tells me that 9.29 turns should be enough to tune it.
Im using 160mm pipe, secondary is 725 long and using 30AWG wire. Primary is 8mm copper pipe and inner radius around 95mm and outer radius is 225mm. MMC is 0.24uF
Toroid is 670 outside diameter made from 150mm semi rigid ducting.
The scope was showing the resonant frequency at just under 53KHz.
When i scoped the primary, i cant tune it any lower than 56 KHz so there is a 3KHz difference between the primary and secondary.
I could easily squash the ducting to make a smaller toroid to raise the frequency.
Im a bit lost here, as my initial calculations with javaTC were almost 10KHz higher.
Any ideas?
Will i have to add a larger primary at some point?
nzoomed, Sun Jun 05 2016, 01:46AM
Thanks for that info.
At present im concerned about my tuning the coil.
I thought 10 turns on the primary would have been plenty, JavaTC gives me a resonant frequency of around 53.78 KHz for the primary and 58.22 for the secondary.
If i click on auto tune on Java TC, it tells me that 9.29 turns should be enough to tune it.
Im using 160mm pipe, secondary is 725 long and using 30AWG wire. Primary is 8mm copper pipe and inner radius around 95mm and outer radius is 225mm. MMC is 0.24uF
Toroid is 670 outside diameter made from 150mm semi rigid ducting.
The scope was showing the resonant frequency at just under 53KHz.
When i scoped the primary, i cant tune it any lower than 56 KHz so there is a 3KHz difference between the primary and secondary.
I could easily squash the ducting to make a smaller toroid to raise the frequency.
Im a bit lost here, as my initial calculations with javaTC were almost 10KHz higher.
Any ideas?
Will i have to add a larger primary at some point?
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Sun Jun 05 2016, 06:00AM
If you have a primary coil already made, adding some more capacitance to the MMC is the "easiest" solution, maybe not the cheapest :)
I plotted your numbers in JavaTC and that is quite different from your measurements, something is not right here in the numbers you gave. Did you mistake some of the inputs and gave diameter instead of radius?
Mads Barnkob, Sun Jun 05 2016, 06:00AM
If you have a primary coil already made, adding some more capacitance to the MMC is the "easiest" solution, maybe not the cheapest :)
I plotted your numbers in JavaTC and that is quite different from your measurements, something is not right here in the numbers you gave. Did you mistake some of the inputs and gave diameter instead of radius?
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 05 2016, 06:58AM
I take it you selected auto tune and its set the radius of 16.4 for the primary and around 5 turns?
I last used JavaTC about 12 months ago and it was telling me about 64 KHz, which was what i was going by.
I still have all my notes and re-entered them, so dont know what went wrong, it baffled me, only difference was that my toroid is about 70mm larger in diameter and my MMC is slightly larger at 0.24uF, but i dont think thats enough to put things out of whack.
Here is my last screenshot from JavaTC at that time.
nzoomed, Sun Jun 05 2016, 06:58AM
Mads Barnkob wrote ...
If you have a primary coil already made, adding some more capacitance to the MMC is the "easiest" solution, maybe not the cheapest :)
I plotted your numbers in JavaTC and that is quite different from your measurements, something is not right here in the numbers you gave. Did you mistake some of the inputs and gave diameter instead of radius?
No, everything looks right that you have entered, except you have got way less turns on there than what i worked out. I worked out about 2621 turns, another calculator said around 2400 turns for 30AWG.If you have a primary coil already made, adding some more capacitance to the MMC is the "easiest" solution, maybe not the cheapest :)
I plotted your numbers in JavaTC and that is quite different from your measurements, something is not right here in the numbers you gave. Did you mistake some of the inputs and gave diameter instead of radius?
I take it you selected auto tune and its set the radius of 16.4 for the primary and around 5 turns?
I last used JavaTC about 12 months ago and it was telling me about 64 KHz, which was what i was going by.
I still have all my notes and re-entered them, so dont know what went wrong, it baffled me, only difference was that my toroid is about 70mm larger in diameter and my MMC is slightly larger at 0.24uF, but i dont think thats enough to put things out of whack.
Here is my last screenshot from JavaTC at that time.
Re: Building my first DRSSTC - Updated thread...
Mads Barnkob, Sun Jun 05 2016, 08:47AM
Oops, I looked at wire length and not winding length from my calculator :)
You are right, your numbers are correct.
If I were you, I would double the MMC capacitance to 0.48 uF and that would give you a tuning point around 7 turns.
Mads Barnkob, Sun Jun 05 2016, 08:47AM
Oops, I looked at wire length and not winding length from my calculator :)
You are right, your numbers are correct.
If I were you, I would double the MMC capacitance to 0.48 uF and that would give you a tuning point around 7 turns.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 05 2016, 09:03AM
I also dont know where to put them, would be cheaper and easier for me to make the toroid smaller i guess?
Anyway, I did not measure the primary with the secondary in place, so i expect this will raise the frequency somewhat?
nzoomed, Sun Jun 05 2016, 09:03AM
Mads Barnkob wrote ...
Oops, I looked at wire length and not winding length from my calculator :)
You are right, your numbers are correct.
If I were you, I would double the MMC capacitance to 0.48 uF and that would give you a tuning point around 7 turns.
Thats alot more $$$ ive got to spend on some CDE942 caps! :POops, I looked at wire length and not winding length from my calculator :)
You are right, your numbers are correct.
If I were you, I would double the MMC capacitance to 0.48 uF and that would give you a tuning point around 7 turns.
I also dont know where to put them, would be cheaper and easier for me to make the toroid smaller i guess?
Anyway, I did not measure the primary with the secondary in place, so i expect this will raise the frequency somewhat?
Re: Building my first DRSSTC - Updated thread...
Blackcurrant, Sun Jun 05 2016, 10:46PM
Try putting the coil in the middle of the room and up a bit ie. on a small table see if the sec coil res goes up a bit.
Blackcurrant, Sun Jun 05 2016, 10:46PM
Try putting the coil in the middle of the room and up a bit ie. on a small table see if the sec coil res goes up a bit.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 05 2016, 11:45PM
Ill try that, might have to do it outside, the toroid will hit the ceiling in my workshop.
I am going to shortly check the resonant frequency of the primary with the secondary and toroid in place, this should hopefully raise it a bit.
In the meantime i have been scoping the bus with 15VAC fed into the voltage doubler.
Its too low voltage on the primary to give me a signal from the CT so am unable to tune the phase lead until i put around 100v there, but here is a scope shot of the bus(yellow) and gates(blue)
below
nzoomed, Sun Jun 05 2016, 11:45PM
Blackcurrant wrote ...
Try putting the coil in the middle of the room and up a bit ie. on a small table see if the sec coil res goes up a bit.
Try putting the coil in the middle of the room and up a bit ie. on a small table see if the sec coil res goes up a bit.
Ill try that, might have to do it outside, the toroid will hit the ceiling in my workshop.
I am going to shortly check the resonant frequency of the primary with the secondary and toroid in place, this should hopefully raise it a bit.
In the meantime i have been scoping the bus with 15VAC fed into the voltage doubler.
Its too low voltage on the primary to give me a signal from the CT so am unable to tune the phase lead until i put around 100v there, but here is a scope shot of the bus(yellow) and gates(blue)
below
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed Jun 08 2016, 03:25AM
Got my first sparks after doing low voltage tests on the bus (15VAC doubled and rectified to 40V)
What is the correct way of connecting the scope to the bus and gates?
I had initially connected the probe to one of the gates and the ground clip to the corresponding emitter and i put both ends of the other lead across the bridge.
I was told that both ground clips should be connected to the negative rail on the bus.
This fixed the buzzing sound the scope was making, but im still having trouble getting a image to lock in to the scope, everything sweeps across the screen.
Is there anything i need to do with these digital scopes to lock in the waveform?
Im also sure with the phasing, if it matters what gate i connect the scope probe up to.
I still cant drive the coil without feeding the signal generator into the driver board, i expect this is because im running at such a low bus voltage. I was getting a slight tingle on my hand if i touched the earth terminal of the scope, is this a concern? I dont know why this is possible with a 40V bus voltage.
I want to adjust the phase lead now, but will need to feed in 100V or more i expect.
My main concern is if this will damage the IGBT's if the phase lead is not set correctly, but i need to power up the coil to set the phase lead, so i expect this is not too much concern.
Anyway, below is an image of the coil operating currently.
Thanks for everyone's help :)
nzoomed, Wed Jun 08 2016, 03:25AM
Got my first sparks after doing low voltage tests on the bus (15VAC doubled and rectified to 40V)
What is the correct way of connecting the scope to the bus and gates?
I had initially connected the probe to one of the gates and the ground clip to the corresponding emitter and i put both ends of the other lead across the bridge.
I was told that both ground clips should be connected to the negative rail on the bus.
This fixed the buzzing sound the scope was making, but im still having trouble getting a image to lock in to the scope, everything sweeps across the screen.
Is there anything i need to do with these digital scopes to lock in the waveform?
Im also sure with the phasing, if it matters what gate i connect the scope probe up to.
I still cant drive the coil without feeding the signal generator into the driver board, i expect this is because im running at such a low bus voltage. I was getting a slight tingle on my hand if i touched the earth terminal of the scope, is this a concern? I dont know why this is possible with a 40V bus voltage.
I want to adjust the phase lead now, but will need to feed in 100V or more i expect.
My main concern is if this will damage the IGBT's if the phase lead is not set correctly, but i need to power up the coil to set the phase lead, so i expect this is not too much concern.
Anyway, below is an image of the coil operating currently.
Thanks for everyone's help :)
Re: Building my first DRSSTC - Updated thread...
orac12, Wed Jun 08 2016, 04:24AM
Adjust your trigger level on the scope to try and get a better lock. Just scope one thing at a time, less chance of messing things up that way.
Remember the phasing of your CT has to be correct for it to run, you either change the polarity or the jumpers on your driver board. I think I had to run around 70 volts on the bus to get enough before it would start oscillating it self. Just run short on times (keep current low) and enough voltage to get it oscillating to check phase lead, even if its way off, at low voltage and low on times you shouldn't damage anything.
orac12, Wed Jun 08 2016, 04:24AM
Adjust your trigger level on the scope to try and get a better lock. Just scope one thing at a time, less chance of messing things up that way.
Remember the phasing of your CT has to be correct for it to run, you either change the polarity or the jumpers on your driver board. I think I had to run around 70 volts on the bus to get enough before it would start oscillating it self. Just run short on times (keep current low) and enough voltage to get it oscillating to check phase lead, even if its way off, at low voltage and low on times you shouldn't damage anything.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed Jun 08 2016, 06:49AM
Yes, ive tried that with not much success, i can seem to lock the gates OK, its the bus that i cant get a lock onto.
Im pretty sure ive got the phase jumper set right, but i can configure the power transformer to about 110VAC.
Will try that and see how it goes.
I remember ive got a small variac that i can use to adjust the voltage more also.
Will keep you updated, if i dont get anywhere i may need a better scope.
nzoomed, Wed Jun 08 2016, 06:49AM
orac12 wrote ...
Adjust your trigger level on the scope to try and get a better lock. Just scope one thing at a time, less chance of messing things up that way.
Remember the phasing of your CT has to be correct for it to run, you either change the polarity or the jumpers on your driver board. I think I had to run around 70 volts on the bus to get enough before it would start oscillating it self. Just run short on times (keep current low) and enough voltage to get it oscillating to check phase lead, even if its way off, at low voltage and low on times you shouldn't damage anything.
Adjust your trigger level on the scope to try and get a better lock. Just scope one thing at a time, less chance of messing things up that way.
Remember the phasing of your CT has to be correct for it to run, you either change the polarity or the jumpers on your driver board. I think I had to run around 70 volts on the bus to get enough before it would start oscillating it self. Just run short on times (keep current low) and enough voltage to get it oscillating to check phase lead, even if its way off, at low voltage and low on times you shouldn't damage anything.
Yes, ive tried that with not much success, i can seem to lock the gates OK, its the bus that i cant get a lock onto.
Im pretty sure ive got the phase jumper set right, but i can configure the power transformer to about 110VAC.
Will try that and see how it goes.
I remember ive got a small variac that i can use to adjust the voltage more also.
Will keep you updated, if i dont get anywhere i may need a better scope.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 09 2016, 04:51AM
OK, i tested the coil with an isolation transformer and a small variac.
I turned the bus voltage up to around 100VDC and was getting huge sparks with a loud crack about 60mm long or more.
The coil is able to drive itself on its own, but im unsure if my phase lead is set right, is it possible i will need a different tunable inductor?
Its hard to explain, but ive added some shots from the scope. These tests were done with a heavy hammer in the centre.
The yellow trace is the bus voltage and the blue is the gate voltage.
This is the first Shot, as you can see there is significant ringing, you will see this drop with each cycle from left to right. It always seems to be the first pulse that gives the ringing and turning in the inductor does not do anything noticeable on the first cycle, but i can remove all the ringing to the right of this image as you will see on the next picture.
You can see here with the ringing removed, the ringing after the second cycle was virtually invisible before, and now appears removed completley.
here is a shot with the whole waveform showing with the timebase division shortened
and one with it expanded out
Are these results normal to expect?
Amy other ideas to improve this?
The ringing still present is my main concern
nzoomed, Thu Jun 09 2016, 04:51AM
OK, i tested the coil with an isolation transformer and a small variac.
I turned the bus voltage up to around 100VDC and was getting huge sparks with a loud crack about 60mm long or more.
The coil is able to drive itself on its own, but im unsure if my phase lead is set right, is it possible i will need a different tunable inductor?
Its hard to explain, but ive added some shots from the scope. These tests were done with a heavy hammer in the centre.
The yellow trace is the bus voltage and the blue is the gate voltage.
This is the first Shot, as you can see there is significant ringing, you will see this drop with each cycle from left to right. It always seems to be the first pulse that gives the ringing and turning in the inductor does not do anything noticeable on the first cycle, but i can remove all the ringing to the right of this image as you will see on the next picture.
You can see here with the ringing removed, the ringing after the second cycle was virtually invisible before, and now appears removed completley.
here is a shot with the whole waveform showing with the timebase division shortened
and one with it expanded out
Are these results normal to expect?
Amy other ideas to improve this?
The ringing still present is my main concern
Re: Building my first DRSSTC - Updated thread...
loneoceans, Thu Jun 09 2016, 05:08AM
Looks just fine. My UD page explains why. Plug in your secondary already and you're good to go!
loneoceans, Thu Jun 09 2016, 05:08AM
Looks just fine. My UD page explains why. Plug in your secondary already and you're good to go!
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 09 2016, 05:40AM
OK, thanks, i see what you are meaning by you cant remove all switching spikes.
I will firstly set the OCD to 800Apk and fire it up, i doubt I will be getting anywhere near that at this point anyway.
My only concern is the coil tuning, im hoping that it will be tuned near enough to give me sparks 2M or more.
Ill give you the dimensions later to add it to your DRSSTC's around the world if you like :)
nzoomed, Thu Jun 09 2016, 05:40AM
loneoceans wrote ...
Looks just fine. My UD page explains why. Plug in your secondary already and you're good to go!
Looks just fine. My UD page explains why. Plug in your secondary already and you're good to go!
OK, thanks, i see what you are meaning by you cant remove all switching spikes.
I will firstly set the OCD to 800Apk and fire it up, i doubt I will be getting anywhere near that at this point anyway.
My only concern is the coil tuning, im hoping that it will be tuned near enough to give me sparks 2M or more.
Ill give you the dimensions later to add it to your DRSSTC's around the world if you like :)
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 09 2016, 12:06PM
I got it running tonight with a higher bus voltage around 150-200VDC and its putting out sparks a good foot or more.
The higher notes seem to be less powerful and not want to produce a decent arc, but i think that this is probably because im running through a low current power supply and the pulse width of the lower notes allows for more duration between each cycle to charge the bus caps perhaps?
Some MIDI files are also better than others it seems.
nzoomed, Thu Jun 09 2016, 12:06PM
I got it running tonight with a higher bus voltage around 150-200VDC and its putting out sparks a good foot or more.
The higher notes seem to be less powerful and not want to produce a decent arc, but i think that this is probably because im running through a low current power supply and the pulse width of the lower notes allows for more duration between each cycle to charge the bus caps perhaps?
Some MIDI files are also better than others it seems.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Thu Jun 09 2016, 08:22PM
Congrats on getting it running. Can you put another current transformer on the primary circuit with burden resistor and scope the output? That will show you what the current is doing during each pulse. Helpful in tuning too. Can easily see double envelopes, peak currents, OC protection kick-in. I just like watching what's happening to the primary circuit while it's running at full power.
What MIDI controller are you using? Some MIDI files I have to adjust note length because they are so short. Still sounds good enough. Also drop the whole song by an octave or two to avoid those really high notes. I also try avoid complex chords too. It's good that high notes are producing short sparks - the controller is protecting your coil for you :)
Graham Armitage, Thu Jun 09 2016, 08:22PM
Congrats on getting it running. Can you put another current transformer on the primary circuit with burden resistor and scope the output? That will show you what the current is doing during each pulse. Helpful in tuning too. Can easily see double envelopes, peak currents, OC protection kick-in. I just like watching what's happening to the primary circuit while it's running at full power.
What MIDI controller are you using? Some MIDI files I have to adjust note length because they are so short. Still sounds good enough. Also drop the whole song by an octave or two to avoid those really high notes. I also try avoid complex chords too. It's good that high notes are producing short sparks - the controller is protecting your coil for you :)
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 09 2016, 10:49PM
The MIDI files are ones i downloaded from the onetesla and EVR forums, supposedly they were designed for tesla coil use.
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
nzoomed, Thu Jun 09 2016, 10:49PM
Graham Armitage wrote ...
Congrats on getting it running. Can you put another current transformer on the primary circuit with burden resistor and scope the output? That will show you what the current is doing during each pulse. Helpful in tuning too. Can easily see double envelopes, peak currents, OC protection kick-in. I just like watching what's happening to the primary circuit while it's running at full power.
What MIDI controller are you using? Some MIDI files I have to adjust note length because they are so short. Still sounds good enough. Also drop the whole song by an octave or two to avoid those really high notes. I also try avoid complex chords too. It's good that high notes are producing short sparks - the controller is protecting your coil for you :)
OK, ill experiment with another CT attached.Congrats on getting it running. Can you put another current transformer on the primary circuit with burden resistor and scope the output? That will show you what the current is doing during each pulse. Helpful in tuning too. Can easily see double envelopes, peak currents, OC protection kick-in. I just like watching what's happening to the primary circuit while it's running at full power.
What MIDI controller are you using? Some MIDI files I have to adjust note length because they are so short. Still sounds good enough. Also drop the whole song by an octave or two to avoid those really high notes. I also try avoid complex chords too. It's good that high notes are producing short sparks - the controller is protecting your coil for you :)
The MIDI files are ones i downloaded from the onetesla and EVR forums, supposedly they were designed for tesla coil use.
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Fri Jun 10 2016, 03:35PM
Ok, I have the same controller - works well. For just playing basic MIDI and being able to quickly and easily modify the notes I use SpeedyMidi. To sync with the other tracks through a synthesizer I have to switch to AnvilStudio. Even still I purchased an add-on to allow me to output individual MIDI tracks to different ports. It also has the capability to advance or retard the tracks to keep them in synch as they are often a little off between tesla and synth ports. You need a good amp to match the volume of the spark noise which can't be adjusted. It does sound cool though.
Graham Armitage, Fri Jun 10 2016, 03:35PM
nzoomed wrote ...
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
Ok, I have the same controller - works well. For just playing basic MIDI and being able to quickly and easily modify the notes I use SpeedyMidi. To sync with the other tracks through a synthesizer I have to switch to AnvilStudio. Even still I purchased an add-on to allow me to output individual MIDI tracks to different ports. It also has the capability to advance or retard the tracks to keep them in synch as they are often a little off between tesla and synth ports. You need a good amp to match the volume of the spark noise which can't be adjusted. It does sound cool though.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Fri Jun 10 2016, 09:46PM
OK, i will check those programs out.
I know what you mean about advancing or retarding.
I tried this in FLstudio with not alot of success, i found that the notes on the MIDI track were slightly longer and the more you got into the song, the worse the lag was.
I was able to speed up the midi tracks and make the playback the same length, but it still sounded wrong.
Hopefully anvilstudio can do it better :)
nzoomed, Fri Jun 10 2016, 09:46PM
Graham Armitage wrote ...
Ok, I have the same controller - works well. For just playing basic MIDI and being able to quickly and easily modify the notes I use SpeedyMidi. To sync with the other tracks through a synthesizer I have to switch to AnvilStudio. Even still I purchased an add-on to allow me to output individual MIDI tracks to different ports. It also has the capability to advance or retard the tracks to keep them in synch as they are often a little off between tesla and synth ports. You need a good amp to match the volume of the spark noise which can't be adjusted. It does sound cool though.
nzoomed wrote ...
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
Im still trying to learn the ropes in FL studio, there is so much to know.
I also want to find a way to sync midi files to play along to the real soundtrack like is seen on a few youtube videos.
The MIDI controller is the basic Mini MIDI controller by Phillip Slawinski
Ok, I have the same controller - works well. For just playing basic MIDI and being able to quickly and easily modify the notes I use SpeedyMidi. To sync with the other tracks through a synthesizer I have to switch to AnvilStudio. Even still I purchased an add-on to allow me to output individual MIDI tracks to different ports. It also has the capability to advance or retard the tracks to keep them in synch as they are often a little off between tesla and synth ports. You need a good amp to match the volume of the spark noise which can't be adjusted. It does sound cool though.
OK, i will check those programs out.
I know what you mean about advancing or retarding.
I tried this in FLstudio with not alot of success, i found that the notes on the MIDI track were slightly longer and the more you got into the song, the worse the lag was.
I was able to speed up the midi tracks and make the playback the same length, but it still sounded wrong.
Hopefully anvilstudio can do it better :)
Re: Building my first DRSSTC - Updated thread...
Benjamin, Sat Jun 11 2016, 05:08PM
I have good results with anvil studio, but I haven't figured out how to sync audio tracks to midi files yet.
Benjamin, Sat Jun 11 2016, 05:08PM
I have good results with anvil studio, but I haven't figured out how to sync audio tracks to midi files yet.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sat Jun 11 2016, 10:21PM
OK, ill see if i can figure it out.
It may be a good idea that we add a section here to share MIDI files and synced tracks so we can build up a good database to share with other builders.
I will see how i get on.
The place i want to take my coil has a sound stage, so i might be able to plug in to their large PA system!
nzoomed, Sat Jun 11 2016, 10:21PM
Benjamin wrote ...
I have good results with anvil studio, but I haven't figured out how to sync audio tracks to midi files yet.
I have good results with anvil studio, but I haven't figured out how to sync audio tracks to midi files yet.
OK, ill see if i can figure it out.
It may be a good idea that we add a section here to share MIDI files and synced tracks so we can build up a good database to share with other builders.
I will see how i get on.
The place i want to take my coil has a sound stage, so i might be able to plug in to their large PA system!
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Sat Jun 11 2016, 10:55PM
You need the Pro-Mix Accessory Pack for Anvil Studio for this to work. The free version does not allow it. Once you have it installed, you can select each track in the midi file and which port it should output on. So if the guitar track is what you want the Tesla to play, select that track for Tesla Output. Then send the other racks to the synthesizer out port which you can then connect to the receiver or amp. Then play with the timing on the Tesla track and it should stay in sync. It's pretty slick.
I built a MIDI over ethernet module as I was getting too much interference with the USB-MIDI cable and notes kept hanging. Now I can operate the coil wirelessly from the laptop and theoretically control it from anywhere in the world over the internet - can you say GEEKY ?? With this config I actually have no lag at all between the different output ports. Have fun.
Graham Armitage, Sat Jun 11 2016, 10:55PM
You need the Pro-Mix Accessory Pack for Anvil Studio for this to work. The free version does not allow it. Once you have it installed, you can select each track in the midi file and which port it should output on. So if the guitar track is what you want the Tesla to play, select that track for Tesla Output. Then send the other racks to the synthesizer out port which you can then connect to the receiver or amp. Then play with the timing on the Tesla track and it should stay in sync. It's pretty slick.
I built a MIDI over ethernet module as I was getting too much interference with the USB-MIDI cable and notes kept hanging. Now I can operate the coil wirelessly from the laptop and theoretically control it from anywhere in the world over the internet - can you say GEEKY ?? With this config I actually have no lag at all between the different output ports. Have fun.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 12 2016, 04:37AM
OK, i will check it out. I must say im having RF issues with my laptop from the tesla coil, how many metres should the laptop be away from the coil?
I thought 5 metres would be sweet, even at low power, the laptop will lock up when playing a MIDI file.
nzoomed, Sun Jun 12 2016, 04:37AM
Graham Armitage wrote ...
You need the Pro-Mix Accessory Pack for Anvil Studio for this to work. The free version does not allow it. Once you have it installed, you can select each track in the midi file and which port it should output on. So if the guitar track is what you want the Tesla to play, select that track for Tesla Output. Then send the other racks to the synthesizer out port which you can then connect to the receiver or amp. Then play with the timing on the Tesla track and it should stay in sync. It's pretty slick.
I built a MIDI over ethernet module as I was getting too much interference with the USB-MIDI cable and notes kept hanging. Now I can operate the coil wirelessly from the laptop and theoretically control it from anywhere in the world over the internet - can you say GEEKY ?? With this config I actually have no lag at all between the different output ports. Have fun.
You need the Pro-Mix Accessory Pack for Anvil Studio for this to work. The free version does not allow it. Once you have it installed, you can select each track in the midi file and which port it should output on. So if the guitar track is what you want the Tesla to play, select that track for Tesla Output. Then send the other racks to the synthesizer out port which you can then connect to the receiver or amp. Then play with the timing on the Tesla track and it should stay in sync. It's pretty slick.
I built a MIDI over ethernet module as I was getting too much interference with the USB-MIDI cable and notes kept hanging. Now I can operate the coil wirelessly from the laptop and theoretically control it from anywhere in the world over the internet - can you say GEEKY ?? With this config I actually have no lag at all between the different output ports. Have fun.
OK, i will check it out. I must say im having RF issues with my laptop from the tesla coil, how many metres should the laptop be away from the coil?
I thought 5 metres would be sweet, even at low power, the laptop will lock up when playing a MIDI file.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 12 2016, 11:09AM
OK, so i fried my secondary tonight!
Not all bad as i hope i should get by with removing the last 40mm or so off the bottom and re connect it at the base.
I heard a loud pop when i hit a high note and didnt think much of it until it did it two or three more times.
I then saw a small spark at the base and realised what had happened.
Its really weird, its blown off and shattered the lacquer in large flakes at the base and higher up above the burnt patch its expanded and buckled but with no sign of arcing.
I did notice some dark spots all over the secondary that were not there after i wound it a year ago.
I dont know if this is some sort of breakdown on the varnish insulation and/or oxidization on the copper, but its well sealed with several thick coats of floor polyurethane.
Is bad earthing to blame for this?
I earthed the secondary to my heatsink which is directly grounded to my main earth point that is connected to a copper water pipe in my basement, i thought this should be more than adequate, since it goes into the ground.
Any other suggestions?
Should i use a higher quality wire next time (i was led to believe it was good quality)?
Should i use epoxy resin instead of polyurethane?
Thanks for everyone's help, i hope i can still get it down to a mid winter arts festival next friday.
I was blown away with the power the coil is putting out despite it running at roughly half its bus voltage.
I stood some florescent tubes all around the coil and they were sparking from the base to the concrete, and thats without the streamers even hitting it!
I really dont know how i can continue to operate this in a residential area though, seriously its loud and i could hear dogs barking miles away, i will have alot of angry neighbours if i have this running every week lol!
nzoomed, Sun Jun 12 2016, 11:09AM
OK, so i fried my secondary tonight!
Not all bad as i hope i should get by with removing the last 40mm or so off the bottom and re connect it at the base.
I heard a loud pop when i hit a high note and didnt think much of it until it did it two or three more times.
I then saw a small spark at the base and realised what had happened.
Its really weird, its blown off and shattered the lacquer in large flakes at the base and higher up above the burnt patch its expanded and buckled but with no sign of arcing.
I did notice some dark spots all over the secondary that were not there after i wound it a year ago.
I dont know if this is some sort of breakdown on the varnish insulation and/or oxidization on the copper, but its well sealed with several thick coats of floor polyurethane.
Is bad earthing to blame for this?
I earthed the secondary to my heatsink which is directly grounded to my main earth point that is connected to a copper water pipe in my basement, i thought this should be more than adequate, since it goes into the ground.
Any other suggestions?
Should i use a higher quality wire next time (i was led to believe it was good quality)?
Should i use epoxy resin instead of polyurethane?
Thanks for everyone's help, i hope i can still get it down to a mid winter arts festival next friday.
I was blown away with the power the coil is putting out despite it running at roughly half its bus voltage.
I stood some florescent tubes all around the coil and they were sparking from the base to the concrete, and thats without the streamers even hitting it!
I really dont know how i can continue to operate this in a residential area though, seriously its loud and i could hear dogs barking miles away, i will have alot of angry neighbours if i have this running every week lol!
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 12 2016, 11:14AM
.
nzoomed, Sun Jun 12 2016, 11:14AM
.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Sun Jun 12 2016, 11:57AM
Horrible when that happens - but consider it a learning opportunity. I prefer to ground my coil to the mains earth wire. I have a gounding bus on the base and everything connects to that. You can't be sure the copper pipes in your place are grounded properly. I have never had an issue connecting to the earth wire - that way I can run my coil at most venues without an issue.
Always used Poly, as do others, when sealing the coil. Don't think that's the problem. If it's arcing at the base, the bottom of your coil is probably not at ground potential. Lucky that's all you fried. Remember that taking off 40mm of windings is going increase your Fres significantly. Good luck.
Graham Armitage, Sun Jun 12 2016, 11:57AM
Horrible when that happens - but consider it a learning opportunity. I prefer to ground my coil to the mains earth wire. I have a gounding bus on the base and everything connects to that. You can't be sure the copper pipes in your place are grounded properly. I have never had an issue connecting to the earth wire - that way I can run my coil at most venues without an issue.
Always used Poly, as do others, when sealing the coil. Don't think that's the problem. If it's arcing at the base, the bottom of your coil is probably not at ground potential. Lucky that's all you fried. Remember that taking off 40mm of windings is going increase your Fres significantly. Good luck.
Re: Building my first DRSSTC - Updated thread...
orac12, Sun Jun 12 2016, 09:46PM
I had same problems, make sure mmc is connected to outer end of your coil and inverter connected to inside end. I also took one turn off the inner of my primary to lower the coupling. I then had to also raise the secondary another inch.
orac12, Sun Jun 12 2016, 09:46PM
I had same problems, make sure mmc is connected to outer end of your coil and inverter connected to inside end. I also took one turn off the inner of my primary to lower the coupling. I then had to also raise the secondary another inch.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 12 2016, 10:12PM
OK, well i can see the water pipe going into the ground, so i thought that would be sweet.
I was worried about using mains earth as others here have advised me its a bad idea and increases the risk of it damaging electrical appliances in the house. I am already still worried that its potentially striking mains cabling that is running in the basement already as its a short stud height.
Do you think it potentially could have damaged my IGBT's if it was not at ground potential?
Im not worried about raising the frequency, i could afford to raise it if anything, the secondary is currently running about .6KHz lower than the primary can maxed out at 10 turns.
nzoomed, Sun Jun 12 2016, 10:12PM
Graham Armitage wrote ...
Horrible when that happens - but consider it a learning opportunity. I prefer to ground my coil to the mains earth wire. I have a gounding bus on the base and everything connects to that. You can't be sure the copper pipes in your place are grounded properly. I have never had an issue connecting to the earth wire - that way I can run my coil at most venues without an issue.
Always used Poly, as do others, when sealing the coil. Don't think that's the problem. If it's arcing at the base, the bottom of your coil is probably not at ground potential. Lucky that's all you fried. Remember that taking off 40mm of windings is going increase your Fres significantly. Good luck.
Horrible when that happens - but consider it a learning opportunity. I prefer to ground my coil to the mains earth wire. I have a gounding bus on the base and everything connects to that. You can't be sure the copper pipes in your place are grounded properly. I have never had an issue connecting to the earth wire - that way I can run my coil at most venues without an issue.
Always used Poly, as do others, when sealing the coil. Don't think that's the problem. If it's arcing at the base, the bottom of your coil is probably not at ground potential. Lucky that's all you fried. Remember that taking off 40mm of windings is going increase your Fres significantly. Good luck.
OK, well i can see the water pipe going into the ground, so i thought that would be sweet.
I was worried about using mains earth as others here have advised me its a bad idea and increases the risk of it damaging electrical appliances in the house. I am already still worried that its potentially striking mains cabling that is running in the basement already as its a short stud height.
Do you think it potentially could have damaged my IGBT's if it was not at ground potential?
Im not worried about raising the frequency, i could afford to raise it if anything, the secondary is currently running about .6KHz lower than the primary can maxed out at 10 turns.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 12 2016, 10:16PM
OK, thats interesting, ive got my MMC going to the centre and the other IGBT is going to the outer winding.
I couldnt find any info on this and its hard to decipher what way from the schematics ive seen.
nzoomed, Sun Jun 12 2016, 10:16PM
orac12 wrote ...
I had same problems, make sure mmc is connected to outer end of your coil and inverter connected to inside end. I also took one turn off the inner of my primary to lower the coupling. I then had to also raise the secondary another inch.
I had same problems, make sure mmc is connected to outer end of your coil and inverter connected to inside end. I also took one turn off the inner of my primary to lower the coupling. I then had to also raise the secondary another inch.
OK, thats interesting, ive got my MMC going to the centre and the other IGBT is going to the outer winding.
I couldnt find any info on this and its hard to decipher what way from the schematics ive seen.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 16 2016, 09:16AM
Here is a video of my first high powered test.
I think the streamers are currently capable of going much longer, its just that they want to keep hitting the steel lamp shade from the fluorescent lights in the basement, its not even grounded, yet the streamers are hitting that and not the fluorescent tube that is earthed.
Bus voltage is around 240V and the OCD will kick in a fair bit depending on the tune played. Ive only set it to around 600A, but i may take it up to 750A soon and see how that plays out.
At the end of the video you can hear it locking up on the same note, this is caused by RF from the coil i think, its messing up with the USB MIDI adaptor.
I wrapped tin foil round it and put in an aluminium box and that helps a little, but as you can see in the video, its still happening, any ideas?
Perhaps i will have to make a Faraday cage for the laptop.
How far do you typically have to keep your computer away?
I thought a 5M optical cable would have been plenty.
nzoomed, Thu Jun 16 2016, 09:16AM
Here is a video of my first high powered test.
I think the streamers are currently capable of going much longer, its just that they want to keep hitting the steel lamp shade from the fluorescent lights in the basement, its not even grounded, yet the streamers are hitting that and not the fluorescent tube that is earthed.
Bus voltage is around 240V and the OCD will kick in a fair bit depending on the tune played. Ive only set it to around 600A, but i may take it up to 750A soon and see how that plays out.
At the end of the video you can hear it locking up on the same note, this is caused by RF from the coil i think, its messing up with the USB MIDI adaptor.
I wrapped tin foil round it and put in an aluminium box and that helps a little, but as you can see in the video, its still happening, any ideas?
Perhaps i will have to make a Faraday cage for the laptop.
How far do you typically have to keep your computer away?
I thought a 5M optical cable would have been plenty.
Re: Building my first DRSSTC - Updated thread...
Hydron, Thu Jun 16 2016, 08:31PM
hahah yeah the MIDI problem isn't a surprise - I had exactly the same issue. For starters USB _HATES_ being around tesla coils (rather susceptible to common mode interference - it has a bus reset signal that isn't differential), and the cheapo converters are poorly designed to make it worse. I found the best solution short of buying or making a better MIDI source was to move it further away (I have a 10m optical cable).
Other than that, looks great, congrats! You should be able to get at least 2m streamers/arcs without much fuss, probably up to ~2.5 if you have the space and tune it right. (my coil has extremely similar dimensions and does a bit over 2m with no problems with OCD set to ~750-1000A)
Hydron, Thu Jun 16 2016, 08:31PM
hahah yeah the MIDI problem isn't a surprise - I had exactly the same issue. For starters USB _HATES_ being around tesla coils (rather susceptible to common mode interference - it has a bus reset signal that isn't differential), and the cheapo converters are poorly designed to make it worse. I found the best solution short of buying or making a better MIDI source was to move it further away (I have a 10m optical cable).
Other than that, looks great, congrats! You should be able to get at least 2m streamers/arcs without much fuss, probably up to ~2.5 if you have the space and tune it right. (my coil has extremely similar dimensions and does a bit over 2m with no problems with OCD set to ~750-1000A)
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jun 16 2016, 09:43PM
OK, thanks, i was suspecting its because its a cheap midi adapter, but i dont know what if any brands are better, there is not many midi adapters available in New Zealand.
Ill see if anything is better on ebay, i will try a tin foil box today and see if that improves everything.
Im keeping my fingers crossed for 2m+ streamers if i set the OCD up to 800A, i may look at upgrading the MMC to 1000A or higher rating, but thats probably overboard for a coil this size.
nzoomed, Thu Jun 16 2016, 09:43PM
OK, thanks, i was suspecting its because its a cheap midi adapter, but i dont know what if any brands are better, there is not many midi adapters available in New Zealand.
Ill see if anything is better on ebay, i will try a tin foil box today and see if that improves everything.
Im keeping my fingers crossed for 2m+ streamers if i set the OCD up to 800A, i may look at upgrading the MMC to 1000A or higher rating, but thats probably overboard for a coil this size.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Sat Jun 18 2016, 02:58PM
I found by wrapping the USB-MIDI cable around a large ferrite toroid helped significantly. Did not eliminate it completely and I had to move further away. It's the USB on the laptop that is the problem. That is why I eventually built the arduino based MIDI over Ethernet so I could have the laptop a LONG way from the coil. I occasionally still get a hanging note, but nowhere near as bad as the USB cable.
Graham Armitage, Sat Jun 18 2016, 02:58PM
I found by wrapping the USB-MIDI cable around a large ferrite toroid helped significantly. Did not eliminate it completely and I had to move further away. It's the USB on the laptop that is the problem. That is why I eventually built the arduino based MIDI over Ethernet so I could have the laptop a LONG way from the coil. I occasionally still get a hanging note, but nowhere near as bad as the USB cable.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sat Jun 18 2016, 10:12PM
Ive put a spare toroid that i was going to originally use for my GDT, but it was too small, so ive wrapped it on the USB cable instead, and its worked a beaut!
I had it hang on one note, but i was a little close to the coil, 5m away seems fine so far without any issues!
Oh, and with Anvil studio, is it possible to sync the actual MP3 recording of the song to the MIDI?
It seems people are doing this somehow.
nzoomed, Sat Jun 18 2016, 10:12PM
Ive put a spare toroid that i was going to originally use for my GDT, but it was too small, so ive wrapped it on the USB cable instead, and its worked a beaut!
I had it hang on one note, but i was a little close to the coil, 5m away seems fine so far without any issues!
Oh, and with Anvil studio, is it possible to sync the actual MP3 recording of the song to the MIDI?
It seems people are doing this somehow.
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Sun Jun 19 2016, 03:32AM
Cool - glad that worked for you. Cheap and easy is always best
I have not tried synching the mp3 and MIDI together. Coming from two different sources I can imagine that's tricky to sync.
Graham Armitage, Sun Jun 19 2016, 03:32AM
Cool - glad that worked for you. Cheap and easy is always best
I have not tried synching the mp3 and MIDI together. Coming from two different sources I can imagine that's tricky to sync.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jun 19 2016, 05:27AM
I think people are doing it in FL studio, but i think they are creating the whole MIDI from scratch which would be a lot of work, and unfortunately im no professional musician even though i play guitar
nzoomed, Sun Jun 19 2016, 05:27AM
Graham Armitage wrote ...
Cool - glad that worked for you. Cheap and easy is always best
I have not tried synching the mp3 and MIDI together. Coming from two different sources I can imagine that's tricky to sync.
Cool - glad that worked for you. Cheap and easy is always best
I have not tried synching the mp3 and MIDI together. Coming from two different sources I can imagine that's tricky to sync.
I think people are doing it in FL studio, but i think they are creating the whole MIDI from scratch which would be a lot of work, and unfortunately im no professional musician even though i play guitar
Re: Building my first DRSSTC - Updated thread...
nzoomed, Sun Jul 17 2016, 09:36PM
So I used it at a public show last week, and it awed the crowd.
Then it started to lightly rain, not recommended to use these in the rain! lol
The coil was sweet, but the fine water droplets seem to absorb alot of energy from the coil's arcs, and they were not as long, the arcs made a different sound, more like a popping i suppose. It also made the USB lock up issue alot worse.
Anyway i quickly switched everything off and told the spectators these dont like the rain, but they were pretty happy with what they saw :)
I have to laugh how the media get so much stuff wrong though!
nzoomed, Sun Jul 17 2016, 09:36PM
So I used it at a public show last week, and it awed the crowd.
Then it started to lightly rain, not recommended to use these in the rain! lol
The coil was sweet, but the fine water droplets seem to absorb alot of energy from the coil's arcs, and they were not as long, the arcs made a different sound, more like a popping i suppose. It also made the USB lock up issue alot worse.
Anyway i quickly switched everything off and told the spectators these dont like the rain, but they were pretty happy with what they saw :)
I have to laugh how the media get so much stuff wrong though!
Re: Building my first DRSSTC - Updated thread...
Graham Armitage, Mon Jul 18 2016, 02:19PM
Congrats on the performance. Crowds love a Tesla Coil. Brave man running it in the rain :) Interesting to hear the weather made the MIDI lockup worse. I will have to try during the next downpour - NOT !!!
Graham Armitage, Mon Jul 18 2016, 02:19PM
Congrats on the performance. Crowds love a Tesla Coil. Brave man running it in the rain :) Interesting to hear the weather made the MIDI lockup worse. I will have to try during the next downpour - NOT !!!
Re: Building my first DRSSTC - Updated thread...
nzoomed, Wed Jul 20 2016, 12:36AM
I agree, water does not mix with electricity.
Anyway, i would love to learn more of the physics involved with the dampness and the MIDI issue.
Im only guessing this was the reason, it may not have been why, but I cant afford to have any failures with the MIDI on future performances.
Any ideas?
Ive got toroids on both ends of the midi cable.
nzoomed, Wed Jul 20 2016, 12:36AM
Graham Armitage wrote ...
Congrats on the performance. Crowds love a Tesla Coil. Brave man running it in the rain :) Interesting to hear the weather made the MIDI lockup worse. I will have to try during the next downpour - NOT !!!
Congrats on the performance. Crowds love a Tesla Coil. Brave man running it in the rain :) Interesting to hear the weather made the MIDI lockup worse. I will have to try during the next downpour - NOT !!!
I agree, water does not mix with electricity.
Anyway, i would love to learn more of the physics involved with the dampness and the MIDI issue.
Im only guessing this was the reason, it may not have been why, but I cant afford to have any failures with the MIDI on future performances.
Any ideas?
Ive got toroids on both ends of the midi cable.
Re: Building my first DRSSTC - Updated thread...
Justin, Wed Jul 20 2016, 12:40AM
If it has a plastic body like the inexpensive usb to midi converters you might try a shielding box or a similar arrangement.
Justin, Wed Jul 20 2016, 12:40AM
If it has a plastic body like the inexpensive usb to midi converters you might try a shielding box or a similar arrangement.
Re: Building my first DRSSTC - Updated thread...
nzoomed, Thu Jul 21 2016, 04:59AM
I have it wrapped in tinfoil, but its not making any connection to the shielding, so this may not help things.
nzoomed, Thu Jul 21 2016, 04:59AM
Justin wrote ...
If it has a plastic body like the inexpensive usb to midi converters you might try a shielding box or a similar arrangement.
If it has a plastic body like the inexpensive usb to midi converters you might try a shielding box or a similar arrangement.
I have it wrapped in tinfoil, but its not making any connection to the shielding, so this may not help things.
Print this page