First sstc is not working

[Recep talip]

hello my name is Recep, first sorry for my english  ,https://www.vn-experimenty.eu/teslov-transformator/sstc/sstc-3.html I am trying to copy this coil
 I have been dealing with this project for a while, but it is not working. UCCs are always on fire. I bought them from china.
I ordered some parts from the mouser, but I haven't received it yet.
I couldn't test the GDT because I don't have an oscilloscope material N30 cat5wire
where could the error be?

[ZakW]

Hello Recep, welcome!

Your English is fine

Thank you for the included schematic and pictures. Troubleshooting without an oscilloscope can be difficult.

Some Ideas:

• First thing that comes to mind is counterfeit UCC gate drive chips. I know several other users have had issues with them. From what I have read the ENABLE pin does not work on them. This might lead to the driver running uninterrupted which could stress them a lot leading to excess heating.
• 15v gate drive might be a bit high for MOSFETs, you could overvolt/damage them
• It is hard to tell from the pictures, are the MOSFETs on separate pads to insulated them from the heatsink? If not, they can be shorted.
• How have you been testing the coil? Are you only powering up the driver and it gets hot?
• You said it is not working, have you seen any output from the coil at all?
• Do you hear any clicking or chirping from the GDT when the interrupter is running? That could be an indication the ENABLE pin is working.

Your GDT looks nicely done. From what I have seen the schematic looks pretty standard, I hope you have better results with your parts from Mouser.

[Recep talip]

First thing that comes to mind is counterfeit UCC gate drive chips. I know several other users have had issues with them. From what I have read the ENABLE pin does not work on them. This might lead to the driver running uninterrupted which could stress them a lot leading to excess heating.
15v gate drive might be a bit high for MOSFETs, you could overvolt/damage them
It is hard to tell from the pictures, are the MOSFETs on separate pads to insulated them from the heatsink? If not, they can be shorted.
How have you been testing the coil? Are you only powering up the driver and it gets hot?
You said it is not working, have you seen any output from the coil at all?
Do you hear any clicking or chirping from the GDT when the interrupter is running? That could be an indication the ENABLE pin is working.

Thank you for your answer.  bus voltage 24v
yes my UCC they get warm after a little bit of work
There is thermal paste between the heatsink and the mosfet. pad is safer, i will replace it
there is no output from the coil I am testing with a fluorescent lamb
There was a ringing at first but after the uccs burned out, I replaced them with new ones. It doesn't respond at all anymore.

[ZakW]

yes my UCC they get warm after a little bit of work

If they are not getting really hot in a short amount of time that could be a good sign. Still not ideal, but as long as they are not rapidly heating and smoking. If they are outputting constantly at a high frequency that can lead to them warming up. That is why I mentioned the enable pin might not be functioning properly causing them to output all the time.

Is the bridge drawing any power? Do your MOSFETs heat up?

There is thermal paste between the heatsink and the mosfet. pad is safer, i will replace it

I don't think thermal paste is sufficient insultation, at least not that I have heard of. This could be a separate issue. Best practice is to use pads or separate heatsinks. 

There was a ringing at first but after the uccs burned out, I replaced them with new ones. It doesn't respond at all anymore.

A ringing sound could be a sign of high frequency gate drive (causing heating), at lower frequencies (BPS) there should be a clicking sound from the GDT that matches the output of the interrupter.

there is no output from the coil I am testing with a fluorescent lamb

Throughout your testing, as long as nothing is burning up try flipping the primary output around to check if the phase is backwards.

Do you have an LED on the interrupter to see if it is outputting a pulse like it should be?

[NyaaX_X]

I'm not a expert. So please check and listen to others answers. Sorry for my english too.

For a antenna feedback halfbridge tesla coil. I don't know if 24V bus voltage on halfbridge power stage is high enough to create suitable field to transfer into the circuit.

I see there is a 74hc14 schmitt oscillator for starting up. But it's may output a signal way higher then your coil resonant frequency. If the antenna didn't catch enough power to feedback the oscillation the UCCs will work in too high frequency. And if the "enable" not working in the same time. The UCCs would burn hot. Check if the "enable" is working and tune the "10k" potentiometer let the frequency close to your coil (well you don't have a oscilloscope...). By the way according to others experience it's better to connect the not using input pin of 74HC14 to gnd or VCC.

And I see there're something like voltage regulators? I don't ever use them to power the UCCs cause the output current are usually 1A max. Although there're few 1000uF capacitors. So I don't know if thet're good enough in this using.

[ZakW]

And I see there're something like voltage regulators? I don't ever use them to power the UCCs cause the output current are usually 1A max. Although there're few 1000uF capacitors. So I don't know if thet're good enough in this using.

My understanding is that gate drive chips like the UCC's don't require large amounts of constant current. Instead they output short high current pulses that are sustained by adequate bypass capacitance. The UCC37XX chips can output 9A peak but don't draw 9 amps from the supply. My driver powering an IGBT half bridge (UCC27425 4A ouput) only draws about 30-40mA during operation.

Source: https://highvoltageforum.net/index.php?topic=1903.msg14206#msg14206

UCC output stage is a bipolar/MOS combination.  Output impedance is much higher close to supply rails than would be implied by 9A peak current.

Additional bypass capacitance at the UCC chips is always a good idea, especially with breadboard construction (no ground or power planes).  I'd suggest at least 0.47uF at each driver chip.  (1uF electrolytic capacitors likely have too high ESR, so likely need to be larger if electrolytic.)

I suspect the two 100uf caps in the schematic do little to supply current for the UCC chips.

[Recep talip]

Hello again
i tried with new ucc's There is a ringing sound from gdt but no arc, bus voltage is 100V ,I reversed the primary, nothing changed

[ZakW]

Sorry to hear that it is still not working. Remember when testing the coil, always turn the driver on first and then the bridge.

• Does the bridge draw any current at 100v?
• Can you test the MOSFETs with a multimeter to make sure they are not damaged?
• If the MOSFETs are okay, can you double check GDT phasing is correct?
• Have you tried adjusting how close the antenna is to the coil?

[Recep talip]

Hello again. new uccs arrived. I noticed a few errors in the bridge and fixed it now it works.
but for some reason, a fluorescent bulb barely burns at 100v dc. there are no arcs. How can I fix

[ZakW]

Hello, glad to hear you got it working!

Try flipping the primary phase again, you might have it backwards. Also, I counted 10 or so primary turns. You can try removing a turn or two to see if that helps once the phase is correct.

You might want to bend a small loop or curve the tip of your antenna to help reduce corona build up and limit the likelihood of arcs hitting it.

Nice work!

[Recep talip]

coil is not working CW mode. what could be the problem?

[ZakW]

Is the coil working with the interrupter at least? I am not sure if you got that portion working at this point or not.

When you switch off the interrupter and you have it in CW mode does the output stop? What is happening when "it is not working". Are the UCC's getting hot in CW mode?

Some additional information will hopefully help figure out what is going on.

[Recep talip]

coil is working with interrupter, At 100v dc, only fluorescent bulb lights up, no arcs
nothing happens when i switch to cw mode while working with interrupter
ucc's not warming up

[ZakW]

Two things:

1. Just to be clear, you are getting an output with the interrupter at 100v? Sparks are forming at the breakout point, right?


2. When you switch from the interrupter to CW mode that is when there is no output but only the fluorescent bulb lights up?

[Recep talip]

1: No arc, just fluorescent bulb lights up
2: nothing happens

[ZakW]

Got it, thanks for that. I misunderstood and thought you were getting sparks with the output.

I think your best bet is to get the coil working first with the interrupter. CW mode stresses a lot of components and can easily kill your coil.

Have you tried to disconnect the startup portion or your driver and just use the antenna? I wonder if your coil is not switching over to the antenna feedback once the coil starts running.

[NyaaX_X]

As always. We recommend to cover the driver part with grounded faraday cage (or aluminium foil) to avoid the EMF bothering them . And if you have a oscilloscope . Maybe you can check the driver waveform without powering up the power bridge stage first . And if running this circuit in CW mode under higher frequency . The gate driver IC and gate resistors may be heating up . Notice that .

[Recep talip]

Hello   again ı have a some result,but ı have a lot of problem again
anymore ı have an oscilloscope





+My first problem is that when I put one hand on the output of the interrupter and bring my other hand close to the coil, it works.normally there is just a sizzle

mosfet gate signal


driver output

[davekni]

Hello   again ı have a some result,but ı have a lot of problem again anymore ı have an oscilloscope

Great that you have a scope now!  That will be very helpful.  Likely worth reading more about scope use and probing techniques.
For almost all uses, avoid 1x probes.  Don't use 1x setting on switchable probes.  Bandwidth is low and signal loading high.
Adjust probe compensation using scope's square wave test signal output (provided for that purpose).  Read your scope and probe manuals to learn how.  Search internet if manuals aren't clear.
Connect scope probe ground clip close to signal being measured.  Of course, if probing half-bridge, use a bench supply or isolation transformer to power coil.  Never ground to non-isolated line voltage.
Set vertical channels for DC-coupled, not AC coupled.  (This is separate from trigger coupling mode.)  Looks like you are doing this already.

If there are still issues, pictures of scope probe connection will help.

[Recep talip]

I'm sorry for the late reply
The circuit I made is not working properly. I have been dealing with this for about 3 months.
It only works when I touch the interrupter with one hand and bring my other hand closer to the coil.i think i will copy another design

[davekni]

he circuit I made is not working properly. I have been dealing with this for about 3 months.
It only works when I touch the interrupter with one hand and bring my other hand closer to the coil.i think i will copy another design

Sounds disappointing!  Sometimes starting over is the easiest.  However, I don't think there is any inherent issue with the circuit you are using.  Now that you have a scope, debugging should be possible.  It will take some time to step through the circuit scoping signals at each point.  If you want to continue, I'd suggest testing entirely in interrupted mode with duty cycle set low.  That minimizes risk of part damage as average power remains low.

[Mads Barnkob]

Sounds like you need to find the fault in the feedback circuit. The gain is not high enough, there must be some faulty or wrong component. You are basically acting as a large antenna and amplifier, when you can make it work by touching it.

I have also started over completely on some projects, it happens to all of us, do not loose hope

The schematic seems to be a run of the mill version of Steve Wards Mini SSTC, which is really one of the simplest to get running, I would recommend to still go with that schematic. I built 3 of them myself: https://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-i/ and https://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-ii/ and https://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-iii/

[NyaaX_X]

My half-bridge SSTC is still been tested too. My experiences may not be helpful. 

The wiring might be as short as possible. But the primary coil leads be stranded and make the tesla coil tower far away from your driver at least about 15cm (for me) might be better. And away from interrupter like 30cm up and make the interrupter connect to power V- or ground. I saw many people put all the circuit under the tower but their circuit might be finished. And that position seems also have low EMI.

The unuse "input" pins of 74HC14 are better to be connected to ground or 5V.

And you can make a 200k ~ 500k Hz oscillator with ~50% dutycycle 5V squarewave output directly go into UCC "in" pin and catch the GDT waveforms with oscilloscope. I think here do not have to  power the half-bridge power stage while testing.  To test if your half-bridge driver working fine. (Also maybe disable the interrupter so driver run in CW mode. Watch out the UCC temperature)

[Recep talip]

Thank you all. I'll fix the feedback section soon and fix other issues. thank you all  for inspiring me
I will follow all your advice,

[Recep talip]

  hello again i'm back with more problems. I rebuilt the driver circuit, and I changed the UCCs and 74hc14 and to test gdt I  bought a signal genenator 
N30 material 10 turn CAT5 wire





driver output



interrupter output


my first problem is: a ringing comes from the bridge and comes a tittle from gdt. Trying to reduce the number of turns of the primer didn't help.I changed the direction of the primer but nothing changed.
all unused pins of 74hc14 connected to gnd. bus voltage is 80v dc   I still can't get this project working.

[davekni]

all unused pins of 74hc14 connected to gnd.

Only unused inputs should be connected to ground.  Leave unused outputs unconnected.  With inputs grounded, outputs will be high.  Grounding them will short outputs and make 74HC14 chip get hot.

interrupter output

I'd suggest testing at much lower duty cycle than shown 50%.  Reduces risk of something burning out.

Then scope GDT inputs (driver outputs) while trying to run in feedback mode.    Scope antenna too.  These scope traces should give some idea of what the issue may be.

[Recep talip]

Only unused inputs should be connected to ground.

yes that's how i did it

Ringing coming from primary and gdt in interrupter mode
No output in CW mode,touching the antenna did not work. This is the signal on the input pins of the uccs

[NyaaX_X]

If there are mistakes in my saying please check and correct them for safety, thank you very much

Here I have a few questions. Where do you connect your secondary coil ground to? A big piece metal or something else? And how much is the k-factor in your coil. The javaTC might help you to calculate the parameters of tesla coil. Sorry for I don't know how to use it.. Here is the url https://www.classictesla.com/java/javatc/javatc.html . And in your post the "driver output" 80V waveform seems to be measured from the half bridge? without secondary? If that's right I think the GDT is working. Maybe you can catch a VGS waveform and post it in the future. Here I don't recommend you probe your live wire connected circuit with a non-isolated oscilloscope. And the ringing may caused by the parasitic inductors and oscillation. You have to optimize your power stage wiring and also can add a capacitor between Vbus +- near your half-bridge transistors as near as possible to  absorb some switching ringing.

And you can also read about Steve ward's articles like https://www.stevehv.4hv.org/SSTC5.htm or the transformer feedback type from the left-up side of
https://www.stevehv.4hv.org/DRSSTC.5.htm, you can ignore the JK flip-flop part, focus on the 74HC14 and current transformer part. I'm shocked you buy a function generator.. But not bad if you like to play with electronics and you have enough resource 

Pay attention while playing with AC live wire...

[davekni]

This is the signal on the input pins of the uccs

That looks like a great 671kHz square wave.  As NyaaX_X said, JavaTC can tell you if that is expected frequency for your secondary and top load.  If so, then coil is running, just without enough power.  If JavaTC indicates a lower frequency, then the antenna is picking up the wrong signal and/or phase is inverted.

[Recep talip]

  Hello again, I'm back with some good results
I connect the secondary to the building ground line. there is a bus cap 1500uf 450v

when it started to work at first, the length of the arcs was short, I set the primary as 4 turns and it looks like this at 80v dc. but the coil sometimes doesn't work when playing with the interrupter.
the coil  not working in CW mode. I think what Davekni said is not true for my 671khz coil
 I will review and share when I have time
I am grateful for your help, thank you all

I don't know much about how to use javatc. need to do some research

[NyaaX_X]

So how do you get that 631kHz? Feedback form coil or function generator?
If you want to boost up the Vbus voltage then you might have to decrease the ringing on your power stage half bridge.