SGTC underperforming, suspecting sparkgap
Pcnerd, Wed Apr 09 2014, 12:55PMHi,
Together with a friend we are currently finishing our SGTC with a 80mm secondary. Picture is here: http://berg-vanden.nl/dump/4hv/total.jpg. Power supply is 7.5kv NST.
We are not seeing the expected performance and were are suspecting our sparkgap, here is the story:
At first we suffered from issues which we determined to be in our filter. The issue was that the TC would not start quit good, or it would stop out of itself. A video of our best run is found here: https://www.youtube.com/watch?v=49POf1jr1yo
After this we created a new filter: http://berg-vanden.nl/dump/4hv/filter.jpg
And then we re-ran the tesla coil and stumbled into more problems, in particular with our sparkgap. A picture of our sparkgap consisting of two brass threaded rods with 2 acorn nuts fitted inside a PVC T-joint with a hairdryer blower in the leg of the T. The top is cut open to allow for the air to get out. Picture: http://berg-vanden.nl/dump/4hv/sparkgap.jpg.
At first we started the TC by first starting the blower and then the NST. Setting the width for the sparkgap here resulted in a very small width giving very poor performance with bursts of long streamers every now and then, similar to the streamers of the firstrun, together with a "sparky" sound from the sparkgap. See this video:http://youtu.be/yRgCDiD0Cvk.
Then we figured that it might be smarter to start the TC first and then turn on the blower, this resulted in better performance, but not yet near that of the first run with the old filter. Video here:https://www.youtube.com/watch?v=MnIbSPDN2bc
Since then we expected our sparkgap to be not well designed, and we would like your expertise on this subject. Is the sparkgap indeed at fault and what would be a good (though simple) design to outfit our TC with?
For references, here is our javatc: http://tomzooi.com/dump/4hv/javatc.txt
cheers
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Thu Apr 17 2014, 01:24PM
Found myself the materials for a synchronous rotary spark gap(see attachement, isn't it cute?), going to test it soon, will let you know soon.
In the meanwhile I'm wondering the following:
-Where would be the "tipping point" where one would use a RQ gap instead of a normal sparkgap,
-and where would it be to change from RQ sparkgap to rotary SG?
Pcnerd, Thu Apr 17 2014, 01:24PM
Found myself the materials for a synchronous rotary spark gap(see attachement, isn't it cute?), going to test it soon, will let you know soon.
In the meanwhile I'm wondering the following:
-Where would be the "tipping point" where one would use a RQ gap instead of a normal sparkgap,
-and where would it be to change from RQ sparkgap to rotary SG?
Re: SGTC underperforming, suspecting sparkgap
Sulaiman, Thu Apr 17 2014, 05:31PM
your javaTC output indicates a 0.35cm static spark gap,
it looks like more than 3.5 mm in the photo'
are you sure that the motor for your rsg is synchronous?
it looks to me like an induction (squirrel cage) motor,
so you would have to grind two flats in the rotor to make it synchronous
an arsg can put a lot of stress on an NST
... make sure that your safety spark gap is adjusted properly
the two critical things to adjust are the spark gap
and the primary resonant frequency / primary tap point.
The spark gap should make a very sharp snapping sound
and the arc should be thin and blue/white
if the spark is wide/fuzzy/non-white then more air blast is needed.
Sulaiman, Thu Apr 17 2014, 05:31PM
your javaTC output indicates a 0.35cm static spark gap,
it looks like more than 3.5 mm in the photo'
are you sure that the motor for your rsg is synchronous?
it looks to me like an induction (squirrel cage) motor,
so you would have to grind two flats in the rotor to make it synchronous
an arsg can put a lot of stress on an NST
... make sure that your safety spark gap is adjusted properly
the two critical things to adjust are the spark gap
and the primary resonant frequency / primary tap point.
The spark gap should make a very sharp snapping sound
and the arc should be thin and blue/white
if the spark is wide/fuzzy/non-white then more air blast is needed.
Re: SGTC underperforming, suspecting sparkgap
Hazmatt_(The Underdog), Fri Apr 18 2014, 12:56AM
1. Get rid of Aligator clips! If you haven't done this, do it now. Get some really beefy wire for your interconnects, like 10 AWG silver teflon, works great. Remember that every inch of wire connecting to the primary side IS A LOSS!!! Treat it as such. Anything inbetween the gap, cap, and primary IS.. I repeat IS the primary! So make those leads very short.
2. SINGLE BREAKOUT POINT. You want a single long streamer?... SINGLE BREAKOUT POINT. It's not a big secret, but it tends to make your coil seem like a really hot performer.
3. It is a lovely gap. I like the advance/retard feature, very cute.
Hazmatt_(The Underdog), Fri Apr 18 2014, 12:56AM
1. Get rid of Aligator clips! If you haven't done this, do it now. Get some really beefy wire for your interconnects, like 10 AWG silver teflon, works great. Remember that every inch of wire connecting to the primary side IS A LOSS!!! Treat it as such. Anything inbetween the gap, cap, and primary IS.. I repeat IS the primary! So make those leads very short.
2. SINGLE BREAKOUT POINT. You want a single long streamer?... SINGLE BREAKOUT POINT. It's not a big secret, but it tends to make your coil seem like a really hot performer.
3. It is a lovely gap. I like the advance/retard feature, very cute.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Mon Apr 28 2014, 08:51PM
haha, no worries guys, it's synchronous allright (2 sides milled off indeed). Filter is there with MOV's to protect even against a failing safety gap.
The alligators clip were for testing and are the 230v supply for the motor, the coil runs on 10 awg you can see in the earlier pictures of the filter etc.
We're working on making the leads as short as possible, but we'd like to find the proper tap point first.
Thanks for the tips at least :).
Pcnerd, Mon Apr 28 2014, 08:51PM
haha, no worries guys, it's synchronous allright (2 sides milled off indeed). Filter is there with MOV's to protect even against a failing safety gap.
The alligators clip were for testing and are the 230v supply for the motor, the coil runs on 10 awg you can see in the earlier pictures of the filter etc.
We're working on making the leads as short as possible, but we'd like to find the proper tap point first.
Thanks for the tips at least :).
Re: SGTC underperforming, suspecting sparkgap
redruM69, Tue Apr 29 2014, 07:50AM
Have you verified with a sync stobe that the RSG is in correct phase? Sounds like its close to 90 degrees out.
Is there a reason you need it synchronous? I could never get my sync coil even close to the performance I gained by switching to async. It went from being a mediocre coil to an absolute beast. I understand that theoretically a sync coil should be the most efficient, but it was difficult for me to get it dialed in.
redruM69, Tue Apr 29 2014, 07:50AM
Have you verified with a sync stobe that the RSG is in correct phase? Sounds like its close to 90 degrees out.
Is there a reason you need it synchronous? I could never get my sync coil even close to the performance I gained by switching to async. It went from being a mediocre coil to an absolute beast. I understand that theoretically a sync coil should be the most efficient, but it was difficult for me to get it dialed in.
Re: SGTC underperforming, suspecting sparkgap
Steve Conner, Tue Apr 29 2014, 09:09AM
A synchronous gap will only give a performance improvement if your tank capacitor is big enough to fully load the power supply at a low break rate, like 100 or 120bps.
If you have a smaller tank capacitor, then a higher break rate will give best performance, and beyond about 240bps, the difference between sync and async is hardly worth worrying about.
TLDR: It's actually the big tank capacitor that improves performance, the sync gap is just required to charge a big capacitor at a low break rate in a consistent way without massive voltage surges.
Steve Conner, Tue Apr 29 2014, 09:09AM
A synchronous gap will only give a performance improvement if your tank capacitor is big enough to fully load the power supply at a low break rate, like 100 or 120bps.
If you have a smaller tank capacitor, then a higher break rate will give best performance, and beyond about 240bps, the difference between sync and async is hardly worth worrying about.
TLDR: It's actually the big tank capacitor that improves performance, the sync gap is just required to charge a big capacitor at a low break rate in a consistent way without massive voltage surges.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Fri May 16 2014, 04:09PM
Today some testing, the tesla coil performed way better then in previous test with the normal spark gap (videos coming soon), however the motor appeared not to be very apt for the job. It's a 20-30watt microwave oven fan motor with two sides milled flat. Without any load I manages to get in sync and stay there, but run for some time it will get really hot and since it wouldn't get back ing sync very well. Next step is to try the vacuum motor synchronous trick and replace the old motor with that one (have to adapt the SRSG slightly for that). Thanks for the tips so far btw. We calculated the javaTC with a rotary spark gap and caps are loaded up to 93% which should give 70+cm spark in perfect circumstances, which isn't bad).
Pcnerd, Fri May 16 2014, 04:09PM
Today some testing, the tesla coil performed way better then in previous test with the normal spark gap (videos coming soon), however the motor appeared not to be very apt for the job. It's a 20-30watt microwave oven fan motor with two sides milled flat. Without any load I manages to get in sync and stay there, but run for some time it will get really hot and since it wouldn't get back ing sync very well. Next step is to try the vacuum motor synchronous trick and replace the old motor with that one (have to adapt the SRSG slightly for that). Thanks for the tips so far btw. We calculated the javaTC with a rotary spark gap and caps are loaded up to 93% which should give 70+cm spark in perfect circumstances, which isn't bad).
Re: SGTC underperforming, suspecting sparkgap
Graham Armitage, Fri May 16 2014, 05:47PM
Can I ask why you would not use a simple static spark gap, perhaps with some good air quenching? With a small NST like that, a rotary gap seems like overkill. I have also killed many NSTs with a rotary gap - never quite understood why, but just stopped doing it :( Nice looking rotary gap though - great workmanship!
Graham Armitage, Fri May 16 2014, 05:47PM
Can I ask why you would not use a simple static spark gap, perhaps with some good air quenching? With a small NST like that, a rotary gap seems like overkill. I have also killed many NSTs with a rotary gap - never quite understood why, but just stopped doing it :( Nice looking rotary gap though - great workmanship!
Re: SGTC underperforming, suspecting sparkgap
GrantX, Sat May 17 2014, 06:39AM
Just had a look at the photos of your setup and filter and in both photos it appears as if the strings of MOVs (blue disks, 6 in series) are only connected at one end, at the NST outputs. The other end appears to be floating, instead of connecting to ground. This won't be the cause of your performance woes, but currently the MOVs aren't going to do their job of suppressing voltage spikes that make it into the filter. Just noticed it and thought I should point it out in case you weren't already aware.
I don't have much experience with rotary gaps yet, but I use a static gap in a small SGTC with a 12.6nF tank cap, which is almost the same size as yours. I found that large electrode mass/surface area combined with a very high airflow helped immensely with quenching problems. I also noticed that increasing the coupling of the coils had a MUCH larger effect on performance than fiddling with the spark gap distances.
GrantX, Sat May 17 2014, 06:39AM
Just had a look at the photos of your setup and filter and in both photos it appears as if the strings of MOVs (blue disks, 6 in series) are only connected at one end, at the NST outputs. The other end appears to be floating, instead of connecting to ground. This won't be the cause of your performance woes, but currently the MOVs aren't going to do their job of suppressing voltage spikes that make it into the filter. Just noticed it and thought I should point it out in case you weren't already aware.
I don't have much experience with rotary gaps yet, but I use a static gap in a small SGTC with a 12.6nF tank cap, which is almost the same size as yours. I found that large electrode mass/surface area combined with a very high airflow helped immensely with quenching problems. I also noticed that increasing the coupling of the coils had a MUCH larger effect on performance than fiddling with the spark gap distances.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Sat May 17 2014, 01:44PM
Thanks for the filter heads up, meant to reconnect that (we undid it on purpose).
The primary acrilic design doesn't really allow for higher coupling, but maybe that'll change in the future.
Got myself a vacuum motor made synchronous (2 3A diodes) which runs nicely at 50volts, but the "outer" coils (cage?) do get quit warm, and I have no experience with these motor types to know what warm and what to warm is. The motor draws 4A at 50 volts and is a 1200watt vacuum cleaner motor. I've got a short movie of it running synchronously here:
Pcnerd, Sat May 17 2014, 01:44PM
Thanks for the filter heads up, meant to reconnect that (we undid it on purpose).
The primary acrilic design doesn't really allow for higher coupling, but maybe that'll change in the future.
Got myself a vacuum motor made synchronous (2 3A diodes) which runs nicely at 50volts, but the "outer" coils (cage?) do get quit warm, and I have no experience with these motor types to know what warm and what to warm is. The motor draws 4A at 50 volts and is a 1200watt vacuum cleaner motor. I've got a short movie of it running synchronously here:
Re: SGTC underperforming, suspecting sparkgap
Hazmatt_(The Underdog), Sat May 17 2014, 04:01PM
Vacuum motors are designed such that the air the motor engulfs, also acts to cool the motor! They are not designed to be run without forced air cooling.
Hazmatt_(The Underdog), Sat May 17 2014, 04:01PM
Vacuum motors are designed such that the air the motor engulfs, also acts to cool the motor! They are not designed to be run without forced air cooling.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Sat May 17 2014, 04:26PM
That explains, I'll see if I can make it as such it will cool itself, still interesting that it gets this hot even with only 200 of the 1200watts it is rated for going in.
I mounted a fan and read on that it might get uptil 80 degrees without issues (hope the epoxy around my diodes do not melt at that point), so I will be running a lengthy temperature test as soon as the motor is cooled down again to room temp.
Pcnerd, Sat May 17 2014, 04:26PM
That explains, I'll see if I can make it as such it will cool itself, still interesting that it gets this hot even with only 200 of the 1200watts it is rated for going in.
I mounted a fan and read on
that it might get uptil 80 degrees without issues (hope the epoxy around my diodes do not melt at that point), so I will be running a lengthy temperature test as soon as the motor is cooled down again to room temp.Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Tue May 20 2014, 06:15PM
So far so good, I've bolted on a small pc-fan which was able to keep the motor below 50 degrees for 10 minutes. Next step is to decently mount it, plus the electrode holder on the shaft and fitting it in the original design (Luckily I can keep the part that holds the electrodes. I'll keep you posted.
Pcnerd, Tue May 20 2014, 06:15PM
So far so good, I've bolted on a small pc-fan which was able to keep the motor below 50 degrees for 10 minutes. Next step is to decently mount it, plus the electrode holder on the shaft and fitting it in the original design (Luckily I can keep the part that holds the electrodes. I'll keep you posted.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Sat Sept 06 2014, 02:01PM
It's been a long time, but I got the new setup tested today and I'm satisfied with the result, the TC is probably not tuned very well but she runs stable and streamers so far are not bad. I made a video to show:
We're planning on some further tuning and creating the rest of the acryillic base to make it a transportable product that looks nice :).
Pcnerd, Sat Sept 06 2014, 02:01PM
It's been a long time, but I got the new setup tested today and I'm satisfied with the result, the TC is probably not tuned very well but she runs stable and streamers so far are not bad. I made a video to show:
We're planning on some further tuning and creating the rest of the acryillic base to make it a transportable product that looks nice :).
Re: SGTC underperforming, suspecting sparkgap
Hazmatt_(The Underdog), Sat Sept 06 2014, 05:02PM
Okay, you're still missing your single breakout point.
Your toroid might be a bit small, but try the single breakout point because it forces all the little streamers into a single large streamer.
That's how small systems can be made to seem big!
Hazmatt_(The Underdog), Sat Sept 06 2014, 05:02PM
Okay, you're still missing your single breakout point.
Your toroid might be a bit small, but try the single breakout point because it forces all the little streamers into a single large streamer.
That's how small systems can be made to seem big!
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Sat Sept 06 2014, 05:46PM
As soon as my NAS is finished repairing it's raid setup (apparently one failed during a TC run) and I'm able to shut it down I'm gonna try that. I'm wondering if that happened through the net and I could install some filter or just by the megnatic field (since the NAS is somewhat close) and I need to start running my TC elsewhere instead of in my room ;).
Pcnerd, Sat Sept 06 2014, 05:46PM
As soon as my NAS is finished repairing it's raid setup (apparently one failed during a TC run) and I'm able to shut it down I'm gonna try that. I'm wondering if that happened through the net and I could install some filter or just by the megnatic field (since the NAS is somewhat close) and I need to start running my TC elsewhere instead of in my room ;).
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Sun Sept 07 2014, 10:53AM
With a single breakout point I still get other breakouts, but the performance does not look that bad, purple streamers with decent length, a bit thin though:
next up: creating the rest of the acryllic base, wil keep you posted (here or in some showoff topic) when it is finished :).
Pcnerd, Sun Sept 07 2014, 10:53AM
With a single breakout point I still get other breakouts, but the performance does not look that bad, purple streamers with decent length, a bit thin though:
next up: creating the rest of the acryllic base, wil keep you posted (here or in some showoff topic) when it is finished :).
Re: SGTC underperforming, suspecting sparkgap
Hazmatt_(The Underdog), Sun Sept 07 2014, 03:44PM
Yea it's looking better though!
I think you might need a bigger toroid.
I put a pretty decent sized toroid on mine a long time ago and got decent results. It's the second half of the video.
Hazmatt_(The Underdog), Sun Sept 07 2014, 03:44PM
Yea it's looking better though!
I think you might need a bigger toroid.
I put a pretty decent sized toroid on mine a long time ago and got decent results. It's the second half of the video.
Re: SGTC underperforming, suspecting sparkgap
loneoceans, Mon Sept 08 2014, 07:12AM
Great job getting it working :) Everything seems to be working as expected. If you really want bigger sparks, more power is the way to go. NSTs parallel quite well, and if you can double your input power from the current ~375W that you have, you'll get much better output. :) For now, I think you could try to smooth out your toroid a bit more to prevent the side-breakouts.
loneoceans, Mon Sept 08 2014, 07:12AM
Great job getting it working :) Everything seems to be working as expected. If you really want bigger sparks, more power is the way to go. NSTs parallel quite well, and if you can double your input power from the current ~375W that you have, you'll get much better output. :) For now, I think you could try to smooth out your toroid a bit more to prevent the side-breakouts.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Wed Oct 15 2014, 02:20PM
I hope to work on the toroid ASAP, in the meanwhile I constructed a base met out of acryllic as well, and it looks the part (see image below)
does anyone have experience using something like a rescue blanket as a ground plane? I am getting tired of taping my room floor down with aluminum foil ;).
Full image:
(my god, including that with [img] tags messes up the screen...)
Pcnerd, Wed Oct 15 2014, 02:20PM
I hope to work on the toroid ASAP, in the meanwhile I constructed a base met out of acryllic as well, and it looks the part (see image below)
does anyone have experience using something like a rescue blanket as a ground plane? I am getting tired of taping my room floor down with aluminum foil ;).
Full image:
(my god, including that with [img] tags messes up the screen...)
Re: SGTC underperforming, suspecting sparkgap
loneoceans, Wed Oct 15 2014, 04:23PM
For a reasonably small size coil like yours, I think just a small rod hammered into the ground/grass would work well and you wouldn't need a grounding plane. I'm not sure if some rescue blankets are conductive or not, but if they are, no reason why you couldn't use it as a ground plane. One thing I would do is to definitely lower the toroid so that it's just above the top of the secondary winding. You're going to get severe corona at the top of the windings otherwise. Great looking build :)
loneoceans, Wed Oct 15 2014, 04:23PM
For a reasonably small size coil like yours, I think just a small rod hammered into the ground/grass would work well and you wouldn't need a grounding plane. I'm not sure if some rescue blankets are conductive or not, but if they are, no reason why you couldn't use it as a ground plane. One thing I would do is to definitely lower the toroid so that it's just above the top of the secondary winding. You're going to get severe corona at the top of the windings otherwise. Great looking build :)
Re: SGTC underperforming, suspecting sparkgap
redruM69, Fri Oct 17 2014, 07:20AM
OP, you never did answer my question. Are you set on a sync gap? I feel it would be much easier to get the performance you desire with an async.
redruM69, Fri Oct 17 2014, 07:20AM
OP, you never did answer my question. Are you set on a sync gap? I feel it would be much easier to get the performance you desire with an async.
Re: SGTC underperforming, suspecting sparkgap
Pcnerd, Fri Oct 17 2014, 03:39PM
For now I have made a sync gap indeed. As for the high toroid, the discs you see just below them can be taken away to lower the toroid, but since our TC is not that strong we noticed we could place the toroid rather high without getting corona.
We hope to do a simple test with a "bigger" toroid ind the current setup, but since I'm quit busy at the moment that will take up to a couple of weeks I am afraid. I am planning to also post all the design files of the acryllic online at some point.
Pcnerd, Fri Oct 17 2014, 03:39PM
For now I have made a sync gap indeed. As for the high toroid, the discs you see just below them can be taken away to lower the toroid, but since our TC is not that strong we noticed we could place the toroid rather high without getting corona.
We hope to do a simple test with a "bigger" toroid ind the current setup, but since I'm quit busy at the moment that will take up to a couple of weeks I am afraid. I am planning to also post all the design files of the acryllic online at some point.
Print this page