ÊŽ/2 DRSSTC
omegalabs, Mon Dec 15 2014, 03:30PMHello!
I’m currently working on a big DRSSTC, it’s for a customer and not for myself. They want a dual coil, because they want the sparks to reach each other while they are standing on the top. They will use it mostly indoors, so I thought I should build a coil without RF ground, needing only the mains connection „plug and playâ€. There was 2 option, a bipolar coil or a ÊŽ/2 magnifier (or a 3rd, a dual coil with separated primaries and secondaries grounded to each other, but I’m not sure if that would work without RF ground).
Sparklengh needed is a minimum of 3 meters between coils. With the IGBTs I already have, it’s not a problem. I don’t want to push the coil to the maximum efficiency, instead for maximum reliability. So every parameter should below the maximum ratings of the individual parts.
A few parameters and parts:
Power: 8kW 3 Phase
Primary topology: H bridge
IGBT: Mitsubishi CM600-HA-24H
Primary capacitor: 625nF 16kVDC (80x Aerovox 2uF 1kVDC snubber caps)
Primary impedance: 6,7 ohms
Controller: Flexibrute 1.0 Universal DRSSTC Controller from Eastern Voltage Research
Modulator: Advanced modulator with fiber optic cable
Energy storage capacitors: 2x 3300uF 350VDC Powerlytics 36DX + 48uF PP film capacitor
Bus voltage: 550VDC
Primary frequency: 38kHz
Extra resonator coils: 250mm PVC tube with 0,5mm copper wire 2540 turns.
I built many spark gap coils, a few SSTCs, and no DRSSTCs. So if you have any suggestions please don’t hesitate to share it with me
.Pictures and additional informations will be added soon.
Re: ÊŽ/2 DRSSTC
Sulaiman, Mon Dec 15 2014, 04:43PM
out of my league but
if you have a 1/2 wavelength system with no centre earth,
if one topload arcs to anywhere other than the other topload (e.g. 'earth')
then the centre will be at eht.
I'd expect flashovers to the primary etc.
Sulaiman, Mon Dec 15 2014, 04:43PM
out of my league but
if you have a 1/2 wavelength system with no centre earth,
if one topload arcs to anywhere other than the other topload (e.g. 'earth')
then the centre will be at eht.
I'd expect flashovers to the primary etc.
Re: ÊŽ/2 DRSSTC
teravolt, Mon Dec 15 2014, 05:15PM
others are more expierinced than I but can it be treated like a half wave cener tap so there is one drver unit and two teslas with the same resonant frequency that are grounded
teravolt, Mon Dec 15 2014, 05:15PM
others are more expierinced than I but can it be treated like a half wave cener tap so there is one drver unit and two teslas with the same resonant frequency that are grounded
Re: ÊŽ/2 DRSSTC
Dr. Dark Current, Mon Dec 15 2014, 06:53PM
To me it looks like you don't have enough capacitance on the DC bus, you'll surely be exceeding the max rms currents of the lytics. If you're going for reliability, use at least 4 of the caps.
And Sulaiman is correct about the flashovers.
Dr. Dark Current, Mon Dec 15 2014, 06:53PM
To me it looks like you don't have enough capacitance on the DC bus, you'll surely be exceeding the max rms currents of the lytics. If you're going for reliability, use at least 4 of the caps.
And Sulaiman is correct about the flashovers.
Re: ÊŽ/2 DRSSTC
omegalabs, Mon Dec 15 2014, 10:30PM
I'm aware of the flashovers. I'm going to use a lower coupling than other magnifier systems, + a few PE or PP sheets between coils, and a safety spark gap between the feedlines. I worry about the stray capacitances too.
Or it may be enough,if the center is grounded to a metal sheet on the floor. I made a small spark gap model of this topology, I can test a few things on it this week I hope.
omegalabs, Mon Dec 15 2014, 10:30PM
I'm aware of the flashovers. I'm going to use a lower coupling than other magnifier systems, + a few PE or PP sheets between coils, and a safety spark gap between the feedlines. I worry about the stray capacitances too.
Or it may be enough,if the center is grounded to a metal sheet on the floor. I made a small spark gap model of this topology, I can test a few things on it this week I hope.
If you're going for reliability, use at least 4 of the caps.Then I'll order 4 of them. These are the only parts missing right now.
Re: ÊŽ/2 DRSSTC
Goodchild, Wed Dec 17 2014, 04:35PM
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
Goodchild, Wed Dec 17 2014, 04:35PM
omegalabs wrote ...
I'm aware of the flashovers. I'm going to use a lower coupling than other magnifier systems, + a few PE or PP sheets between coils, and a safety spark gap between the feedlines. I worry about the stray capacitances too.
Or it may be enough,if the center is grounded to a metal sheet on the floor. I made a small spark gap model of this topology, I can test a few things on it this week I hope.
I'm aware of the flashovers. I'm going to use a lower coupling than other magnifier systems, + a few PE or PP sheets between coils, and a safety spark gap between the feedlines. I worry about the stray capacitances too.
Or it may be enough,if the center is grounded to a metal sheet on the floor. I made a small spark gap model of this topology, I can test a few things on it this week I hope.
If you're going for reliability, use at least 4 of the caps.Then I'll order 4 of them. These are the only parts missing right now.
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
Re: ÊŽ/2 DRSSTC
omegalabs, Thu Dec 18 2014, 05:13PM
Their datasheet doesn't give any more information than "high ripple current". So that means it's way less than 20A I probably need?
I bought 4 used ones for 100$, so I'll buy new caps for the final version anyway, but I hope these will hold for at least some low power testings.
Can you give me an exact type I should use?
Right now I can't afford expensive capacitors.
omegalabs, Thu Dec 18 2014, 05:13PM
Goodchild wrote ...
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
Their datasheet doesn't give any more information than "high ripple current". So that means it's way less than 20A I probably need?
I bought 4 used ones for 100$, so I'll buy new caps for the final version anyway, but I hope these will hold for at least some low power testings.
Can you give me an exact type I should use?
Right now I can't afford expensive capacitors.
Re: ÊŽ/2 DRSSTC
loneoceans, Thu Dec 18 2014, 06:35PM
I agree with Goodchild, especially if you're building a coil for a customer. The caps you are using will probably work fine for a while, but I wouldn't place my bets on it for any sort of long-term reliability or even any sort of high duty cycle runs.
A quick search on ebay turned up a few good capacitors for reasonable prices. I know many folks have had great success with Rifa capacitors which are affordable if you are on a budget:
These are 3400uF 420VDC rated. Look at the size of their terminals compared to the ones you are currently using. These should work great as your bus capacitors. Two caps seem a bit on the low side though for a 8kW coil, I'd use no less than 4 of them. Suppose your coil is drawing 8kW at about 300bps, that's 27J per bang. I usually size my bus caps no smaller than 20x bang energy. At 550V on the bus, that's at least 3500uF required.
Finally for your MMC, I'm not sure how much primary current you are planning to run the coil at, but I wouldn't go more than about 1.2kA (given your primary impedance of 6.8R) to keep within the AC ratings of the capacitor. Again for long term reliability, I'd go for even more capacitors for a higher voltage rating. But otherwise it should work fine.
loneoceans, Thu Dec 18 2014, 06:35PM
omegalabs wrote ...
Their datasheet doesn't give any more information than "high ripple current". So that means it's way less than 20A I probably need?
I bought 4 used ones for 100$, so I'll buy new caps for the final version anyway, but I hope these will hold for at least some low power testings.
Can you give me an exact type I should use?
Right now I can't afford expensive capacitors.
Goodchild wrote ...
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
I wouldn't use those caps for bus capacitors at all. Any datasheet that doesn't spec their rated RMS current or ESR means that they are not intended to be used in a high current application. Those caps are most likely marketed at things like audio amplifiers and other application that have low RMS bus current. A another great indicator is the terminal size on the cap. Those capacitors will likely cook when presented with the kinds of currents you want to run.
I would recommend buying capacitor that are intended for inverter applications. Be prepared to pay at least double unless you buy from the surplus market.
Their datasheet doesn't give any more information than "high ripple current". So that means it's way less than 20A I probably need?
I bought 4 used ones for 100$, so I'll buy new caps for the final version anyway, but I hope these will hold for at least some low power testings.
Can you give me an exact type I should use?
Right now I can't afford expensive capacitors.
I agree with Goodchild, especially if you're building a coil for a customer. The caps you are using will probably work fine for a while, but I wouldn't place my bets on it for any sort of long-term reliability or even any sort of high duty cycle runs.
A quick search on ebay turned up a few good capacitors for reasonable prices. I know many folks have had great success with Rifa capacitors which are affordable if you are on a budget:
These are 3400uF 420VDC rated. Look at the size of their terminals compared to the ones you are currently using. These should work great as your bus capacitors. Two caps seem a bit on the low side though for a 8kW coil, I'd use no less than 4 of them. Suppose your coil is drawing 8kW at about 300bps, that's 27J per bang. I usually size my bus caps no smaller than 20x bang energy. At 550V on the bus, that's at least 3500uF required.
Finally for your MMC, I'm not sure how much primary current you are planning to run the coil at, but I wouldn't go more than about 1.2kA (given your primary impedance of 6.8R) to keep within the AC ratings of the capacitor. Again for long term reliability, I'd go for even more capacitors for a higher voltage rating. But otherwise it should work fine.
Re: ÊŽ/2 DRSSTC
omegalabs, Fri Dec 19 2014, 02:27PM
I searched a few containers in the university and found 2 RIFA PEH169 3300uF 350V capacitors which I can use, so I feel lucky today.
Yes the limit will be 1,2kA.
I have a total of 100 snubbers, so a I can go probably a little higher in voltage. Like 17 caps in series for and 6 strings for a total of 705nF @ 17kV. But these are used caps withouth guarantee, so I assume a few will fail on the test I'm going to make with them before use.
Today I tested the dual conception with a small spark gap driver around 350W (NST 7kV) and it worked. Unfortunately I made no video, but I'll make one next month, as the university will be closed from next week.
I made the driver transformer with helical primary and used a PA sheet between the coils. The primary wasn't tuned perfectly as I was needed to test the coil fast, but it made 35cm sparks without a problem. Small surface sparks appeared near the primary coil, but it was expected as the whole surface of the PA sheet was very smooth. The coupling was very high, probably near 0,5.
omegalabs, Fri Dec 19 2014, 02:27PM
loneoceans wrote ...
I agree with Goodchild, especially if you're building a coil for a customer. The caps you are using will probably work fine for a while, but I wouldn't place my bets on it for any sort of long-term reliability or even any sort of high duty cycle runs.
A quick search on ebay turned up a few good capacitors for reasonable prices. I know many folks have had great success with Rifa capacitors which are affordable if you are on a budget:
These are 3400uF 420VDC rated. Look at the size of their terminals compared to the ones you are currently using. These should work great as your bus capacitors. Two caps seem a bit on the low side though for a 8kW coil, I'd use no less than 4 of them. Suppose your coil is drawing 8kW at about 300bps, that's 27J per bang. I usually size my bus caps no smaller than 20x bang energy. At 550V on the bus, that's at least 3500uF required.
Finally for your MMC, I'm not sure how much primary current you are planning to run the coil at, but I wouldn't go more than about 1.2kA (given your primary impedance of 6.8R) to keep within the AC ratings of the capacitor. Again for long term reliability, I'd go for even more capacitors for a higher voltage rating. But otherwise it should work fine.
I agree with Goodchild, especially if you're building a coil for a customer. The caps you are using will probably work fine for a while, but I wouldn't place my bets on it for any sort of long-term reliability or even any sort of high duty cycle runs.
A quick search on ebay turned up a few good capacitors for reasonable prices. I know many folks have had great success with Rifa capacitors which are affordable if you are on a budget:
These are 3400uF 420VDC rated. Look at the size of their terminals compared to the ones you are currently using. These should work great as your bus capacitors. Two caps seem a bit on the low side though for a 8kW coil, I'd use no less than 4 of them. Suppose your coil is drawing 8kW at about 300bps, that's 27J per bang. I usually size my bus caps no smaller than 20x bang energy. At 550V on the bus, that's at least 3500uF required.
Finally for your MMC, I'm not sure how much primary current you are planning to run the coil at, but I wouldn't go more than about 1.2kA (given your primary impedance of 6.8R) to keep within the AC ratings of the capacitor. Again for long term reliability, I'd go for even more capacitors for a higher voltage rating. But otherwise it should work fine.
I searched a few containers in the university and found 2 RIFA PEH169 3300uF 350V capacitors which I can use, so I feel lucky today.
Yes the limit will be 1,2kA.
I have a total of 100 snubbers, so a I can go probably a little higher in voltage. Like 17 caps in series for and 6 strings for a total of 705nF @ 17kV. But these are used caps withouth guarantee, so I assume a few will fail on the test I'm going to make with them before use.
Today I tested the dual conception with a small spark gap driver around 350W (NST 7kV) and it worked. Unfortunately I made no video, but I'll make one next month, as the university will be closed from next week.
I made the driver transformer with helical primary and used a PA sheet between the coils. The primary wasn't tuned perfectly as I was needed to test the coil fast, but it made 35cm sparks without a problem. Small surface sparks appeared near the primary coil, but it was expected as the whole surface of the PA sheet was very smooth. The coupling was very high, probably near 0,5.
Re: ÊŽ/2 DRSSTC
omegalabs, Fri Dec 19 2014, 05:36PM
A few pictures. You can see some snubber caps are broken. Not sure if the the customs did that or it was like that before the shipping. But the seller was nice he sent a total of 110 caps instead of 100.
omegalabs, Fri Dec 19 2014, 05:36PM
A few pictures. You can see some snubber caps are broken. Not sure if the the customs did that or it was like that before the shipping. But the seller was nice he sent a total of 110 caps instead of 100.
Re: ÊŽ/2 DRSSTC
loneoceans, Fri Dec 19 2014, 05:57PM
Looks like you got everything up and running well. I've used those Aerovox capacitors before and they seem to do just fine in DRSSTC use. Most snubber capacitors are usually good choices as well due to their high dv/dt ratings and low inductance construction. Good find on the bus caps as well but you'd still want more for a big coil.
I don't know what EVR's driver is like but make sure it can handle the drive capability for the CM600 bridge. I suppose it's essentially the same as the UD2x driver?
loneoceans, Fri Dec 19 2014, 05:57PM
Looks like you got everything up and running well. I've used those Aerovox capacitors before and they seem to do just fine in DRSSTC use. Most snubber capacitors are usually good choices as well due to their high dv/dt ratings and low inductance construction. Good find on the bus caps as well but you'd still want more for a big coil.
I don't know what EVR's driver is like but make sure it can handle the drive capability for the CM600 bridge. I suppose it's essentially the same as the UD2x driver?
Re: ÊŽ/2 DRSSTC
omegalabs, Fri Dec 19 2014, 06:09PM
Wolfram told me about these capacitors and these were the cheapest and best choice right now. I saw someone used large GTO snubbers, those looks even better (gto snubber), but the new ones are really expensive.
The EVR driver has a maximum of 20A gate drive current capability. It has a dual UCC27423 + FDD8424H p- and n-channel half bridge as a GDT output.
omegalabs, Fri Dec 19 2014, 06:09PM
Wolfram told me about these capacitors and these were the cheapest and best choice right now. I saw someone used large GTO snubbers, those looks even better (gto snubber), but the new ones are really expensive.
The EVR driver has a maximum of 20A gate drive current capability. It has a dual UCC27423 + FDD8424H p- and n-channel half bridge as a GDT output.
Re: ÊŽ/2 DRSSTC
Kizmo, Sat Dec 20 2014, 08:55AM
UCC27423 + FDD8424H in UD2 can drive 8x cm600 equivalents at 75kHz just fine ;)
Kizmo, Sat Dec 20 2014, 08:55AM
UCC27423 + FDD8424H in UD2 can drive 8x cm600 equivalents at 75kHz just fine ;)
Re: ÊŽ/2 DRSSTC
omegalabs, Wed Dec 24 2014, 04:41PM
Today I tested every snubber capacitor at 1kVDC, all passed the test. Plus got a lot of copper yesterday. 0,5mm diameter a total of 4km in lenght.
omegalabs, Wed Dec 24 2014, 04:41PM
Today I tested every snubber capacitor at 1kVDC, all passed the test. Plus got a lot of copper yesterday. 0,5mm diameter a total of 4km in lenght.
Re: ÊŽ/2 DRSSTC
Hydron, Fri Dec 26 2014, 10:48AM
I used the Aerovox caps also with great success. Ran 2 parallel strings of 16-24 caps (depending on how i had the tank impedance setup). At 1-2kW I had them running for >10 min straight without any detectable heating (was testing the bridge by induction cooking eggs in a frypan).
If you got them from ctr surplus they also chucked a few extras in for me as well. Mine were shipped internationally (via a freight forwarder) but ctr knew I was doing this and packed them very well - cost a bit extra in shipping but they were all perfect.
Good luck with the coil!
Hydron, Fri Dec 26 2014, 10:48AM
I used the Aerovox caps also with great success. Ran 2 parallel strings of 16-24 caps (depending on how i had the tank impedance setup). At 1-2kW I had them running for >10 min straight without any detectable heating (was testing the bridge by induction cooking eggs in a frypan).
If you got them from ctr surplus they also chucked a few extras in for me as well. Mine were shipped internationally (via a freight forwarder) but ctr knew I was doing this and packed them very well - cost a bit extra in shipping but they were all perfect.
Good luck with the coil!
Re: ÊŽ/2 DRSSTC
omegalabs, Sat Dec 27 2014, 02:46PM
The funny thing, even the broken ones are working. I think I'll check the maximum voltage which these caps can whitstand.
omegalabs, Sat Dec 27 2014, 02:46PM
The funny thing, even the broken ones are working. I think I'll check the maximum voltage which these caps can whitstand.
Re: ÊŽ/2 DRSSTC
Hydron, Sun Dec 28 2014, 11:34PM
Yeah it just looks like the epoxy potting is chipped rather than the actual capacitor. Some of mine had bent tabs but nothing like your bad ones.
Hydron, Sun Dec 28 2014, 11:34PM
Yeah it just looks like the epoxy potting is chipped rather than the actual capacitor. Some of mine had bent tabs but nothing like your bad ones.
Re: ÊŽ/2 DRSSTC
omegalabs, Mon Dec 29 2014, 07:47PM
I coiled the first resonator coil. I coated it with PU as it dries very fast unlike the epoxy resin I have.
I made a little coiling machine from my lathe. It's useful for big coils like this.
Video
omegalabs, Mon Dec 29 2014, 07:47PM
I coiled the first resonator coil. I coated it with PU as it dries very fast unlike the epoxy resin I have.
I made a little coiling machine from my lathe. It's useful for big coils like this.
Video
Re: ÊŽ/2 DRSSTC
omegalabs, Wed Dec 31 2014, 03:42PM
They're twins!
omegalabs, Wed Dec 31 2014, 03:42PM
They're twins!
Re: ÊŽ/2 DRSSTC
omegalabs, Mon Jan 19 2015, 12:40PM
A question: where should I set the pulse width lockout in the driver? As I understand, PW isn't a major problem, unless there is no load to the primary circuit (no spark at the output), but if this occurs the OCD will act and skip a few periods of the ringing.
Or the question more likely, how important is the setting of the PW lockout?
Other stuff: I realised that the modulator doesn't have any fiber optic transmitter so I bought one from ebay and it works now perfectly with the controller.
And a little criticism about the controller: the inner stabilizers (7824, 7809, 7805) are surrounded by the only heat sensitive elements, all the electrolytic capacitors without any space. More problem, that the 7805 is connected directly to the unstabilized 24V, so it's so hot, it burned my finger literally and I'm not sure how the 7824 will handle the current to the gate driver. I fixed a heatsink to the 7805 and 7824, but I think I should remove them from the PCB and attach all these to a large heatsink and wire them back to the PCB.
omegalabs, Mon Jan 19 2015, 12:40PM
A question: where should I set the pulse width lockout in the driver? As I understand, PW isn't a major problem, unless there is no load to the primary circuit (no spark at the output), but if this occurs the OCD will act and skip a few periods of the ringing.
Or the question more likely, how important is the setting of the PW lockout?
Other stuff: I realised that the modulator doesn't have any fiber optic transmitter so I bought one from ebay and it works now perfectly with the controller.
And a little criticism about the controller: the inner stabilizers (7824, 7809, 7805) are surrounded by the only heat sensitive elements, all the electrolytic capacitors without any space. More problem, that the 7805 is connected directly to the unstabilized 24V, so it's so hot, it burned my finger literally and I'm not sure how the 7824 will handle the current to the gate driver. I fixed a heatsink to the 7805 and 7824, but I think I should remove them from the PCB and attach all these to a large heatsink and wire them back to the PCB.
Re: ÊŽ/2 DRSSTC
loneoceans, Mon Jan 19 2015, 07:43PM
The original UD2 did suffer from the regulators getting quite hot. If I recall correctly, the 7809 was one of the hottest, dropping from 24V to 9V, and you are right that they are placed in-between the capacitors. As long as you're not maxing out to 40V input, the 24V regulator should be ok, However, I'd heat sink them to be safe. After all, this one one of the main reasons why I switched to a switching regulator design for my UD2.7 revision, where the 7824 and 7805 uses the main PCB as a heat sink and the 7809 is a switching regulator: . I've had no problems leaving the board on for a whole day in an enclosed box, so hopefully this has solved the regulator heating problem.
loneoceans, Mon Jan 19 2015, 07:43PM
omegalabs wrote ...
Other stuff: I realised that the modulator doesn't have any fiber optic transmitter so I bought one from ebay and it works now perfectly with the controller.
And a little criticism about the controller: the inner stabilizers (7824, 7809, 7805) are surrounded by the only heat sensitive elements, all the electrolytic capacitors without any space. More problem, that the 7805 is connected directly to the unstabilized 24V, so it's so hot, it burned my finger literally and I'm not sure how the 7824 will handle the current to the gate driver. I fixed a heatsink to the 7805 and 7824, but I think I should remove them from the PCB and attach all these to a large heatsink and wire them back to the PCB.
Other stuff: I realised that the modulator doesn't have any fiber optic transmitter so I bought one from ebay and it works now perfectly with the controller.
And a little criticism about the controller: the inner stabilizers (7824, 7809, 7805) are surrounded by the only heat sensitive elements, all the electrolytic capacitors without any space. More problem, that the 7805 is connected directly to the unstabilized 24V, so it's so hot, it burned my finger literally and I'm not sure how the 7824 will handle the current to the gate driver. I fixed a heatsink to the 7805 and 7824, but I think I should remove them from the PCB and attach all these to a large heatsink and wire them back to the PCB.
The original UD2 did suffer from the regulators getting quite hot. If I recall correctly, the 7809 was one of the hottest, dropping from 24V to 9V, and you are right that they are placed in-between the capacitors. As long as you're not maxing out to 40V input, the 24V regulator should be ok, However, I'd heat sink them to be safe. After all, this one one of the main reasons why I switched to a switching regulator design for my UD2.7 revision, where the 7824 and 7805 uses the main PCB as a heat sink and the 7809 is a switching regulator:
. I've had no problems leaving the board on for a whole day in an enclosed box, so hopefully this has solved the regulator heating problem.Re: ÊŽ/2 DRSSTC
omegalabs, Mon Jan 19 2015, 11:11PM
I see, thanks for the reply.
I welded a frame for the box, which will contain almost everything of the system, the driver, primary circuit and the driving coil for the extra resonator (this is why it's so big). So basically the whole driver will be in a closed box and only 2 wires, or copper tubes needed to be connected to the resonators, plus the modulator and the mains cable of course.
The sides and upper plate of the box must be made with some insulator material, as the driver coil will create a few hundred kV's between its ends and it also prevents capacitive currents flowing back to ground.
omegalabs, Mon Jan 19 2015, 11:11PM
I see, thanks for the reply.
I welded a frame for the box, which will contain almost everything of the system, the driver, primary circuit and the driving coil for the extra resonator (this is why it's so big). So basically the whole driver will be in a closed box and only 2 wires, or copper tubes needed to be connected to the resonators, plus the modulator and the mains cable of course.
The sides and upper plate of the box must be made with some insulator material, as the driver coil will create a few hundred kV's between its ends and it also prevents capacitive currents flowing back to ground.
Re: ÊŽ/2 DRSSTC
omegalabs, Sat Mar 14 2015, 07:31PM
A little update. I brought back my small NST powered model from the university, so I made a few pictures and a video.
Video
There are problems with the air core transformer as there is no insulation on the surface, but I'm going to add something to prevent these nasty sparks.
350W NST and the max spark lenght is 50cm. But more importantly, it shows that the whole system is usable without a secondary ground.
Interestingly, the sparklenght depends way more on the spark gap, than the actual primary capacitance. I changed the capacitors two ways, +-30%, but there was no difference on the output (extremely low primary quality factor?), but when I changed the spark gap, spark grew way bigger immediately. The whole system is very sensitive to what happens in the spark gap, mostly the ON period, while the gap is in the conducting period.
omegalabs, Sat Mar 14 2015, 07:31PM
A little update. I brought back my small NST powered model from the university, so I made a few pictures and a video.
Video
There are problems with the air core transformer as there is no insulation on the surface, but I'm going to add something to prevent these nasty sparks.
350W NST and the max spark lenght is 50cm. But more importantly, it shows that the whole system is usable without a secondary ground.
Interestingly, the sparklenght depends way more on the spark gap, than the actual primary capacitance. I changed the capacitors two ways, +-30%, but there was no difference on the output (extremely low primary quality factor?), but when I changed the spark gap, spark grew way bigger immediately. The whole system is very sensitive to what happens in the spark gap, mostly the ON period, while the gap is in the conducting period.
Print this page