Author Topic: DRSSTC TESLA COIL problem and MONEY reward for working solution  (Read 2600 times)

0 Members and 1 Guest are viewing this topic.

Offline justin66Topic starter

  • Regular Contributor
  • *
  • Posts: 115
hello guys. fist i want to say  i will pay 20 euros for a person who can help me sort this problem out i can give more if you offer extended help

short about this project it is my graduation project that i cant get to work. I have to finish it as quick as possible that is why im offering money. I hope its enough because i have spent too much on this coil already. its almost 350euros for all the components that keep blowing or not working. im tired of wasting money on it and i need smarter people advise.

short about me and my equipment i have multimeter, clamp meter for measuring current ac/dc, osciloscope dso138 witch is good for 100khz at most. i can access a good one in my college.
i have decided to make drsstc  TC for my finals that i kind of regret it, because it didnt turn out the way i wanted. im stuck with it and i need to finish it ASAP. i have electronics as a hobby and my speciality is electrican

short about my TC i decided to use already polished designs that are available on the internet. but it still doesnt work for some odd reason. i use DRSSTC design witch contain Universal driver UD2.7c full Hbridge bridge and an interupter. it seems easy at first, just order some pcbs, put it together and wuala. but i got stuck now. i think i have assembled my devices correctly. i got it working several times, but the result is not what i hoped.
UD2.7 http://www.loneoceans.com/labs/ud27/
H bridge design https://github.com/profdc9/DRSSTC-PCB-Pack/tree/master/full-bridge-transistor
interupter https://easyeda.com/stuarth/steve-ward-s-drsstc-interrupter-with-burst-mode

my trial and error srory i goofed a lot of trial and error tries to get it working. i blew my transistors at least 4 times now.
on first try i can verify, my boards should be working to some extent interupter is definetly working. UD2.7 should be working because my gate drive transformer makes noise according to interupter pulses.
my h bridge should be working, verified with osciloscope, although the drive is not looking the best, at least on the first power up when i had a good osciloscope to probe around. on feedback i ran signal generator to check if it is working.
first i blew a single audyn brand polypropilene capacitor rated 550VAC 0.1uf i saw sparks of 1mm  when metal object is near  secondary.
then by grounding my primary coil i blew IGBT transistors
then i bought new audyn polypropilene caps, this time 4 in series making it 2kv ac capacitor close to 0.1uf i saw 1mm of sparks as well
then i blew transistors again, probably because my Gate drive transformer was 1:2 ratio making it 48v on IGBT gates
then i bought expensive snubber capacitors, they are called CDE capacitors 0.068uf, steve ward used them. it is 12 pcs bank, 3 series, 4 paralel, making it 0.09uf bank of caps
i changed up my primary design a bit, because it seemed that the most voltage is at the middle of the secondary. so i changed cylinder shape primary to pankake. my coil worked like a charm, except it was making sparks 3mm to air and maybe 3cm to grounded object. not the best result. the worst part i had to turn up my interupter and my variac. my variac  was at maximum 250vac and my interupter was turned to pretty high pulse frequency and probably maximum pulse with. suddenly my apartments lights went black. i blew IGBTs again.  |O
Now i found two older pair of transistors that seems to be ok, at least with multimeter diode test mode, so i was lucky and im left with two blown and two kind of ok transistors, so i slapped old pair and new pair of transistors, seems ok everything is working again, this time my GDT seems to change pitch quite differently. i kind of got it working with variac at 180vac, before last blow up by the GDT sound the circuit was geting into a working state as low as 90vac on variac, now it need at least 160vac to go to working state. this state is when the gdt is silent it means circuit is working. and then gdt is noisy it means the circuit is not working correctly.
so this circuit seems to go to working state. the GDT is silent, so my current transformers should be seeing current and giving it to the feedback. i can hear EMF disruptions through my PC speakers. but the funny thing is, no sparks what so ever. i tried different MMC(capacitror bank) and the old one with no sparks this time

additional information my GDT and current transformers are ferrite toroids from car audio  amplifier boost converter circuits. 500:1
the MMC capacitors is CDE cap bank with 6kvdc rating, i also have ceramic type mmc with 5kvdc rating, the little equipment i have says that they are good. transistor tester mk-168 and multimeter shows correct capacitance.
my tesla coil primary is about 3 turns, the secondary is about 1050 turns with former with unknown plastic. it has 0.4mm copper wire and former radius is 4.3cm and diameter is 8.6cm. winding height is about 43.5cm
i also have top load toroid that doesnt seem to affect things
there is optional resistor values on the Hbridge circuit that im not sure witch one to use and im using IGBTs that are suggested in base of materials list for the Hbridge circuit

couple of pictures of my project:

https://ibb.co/album/c7yhfa
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #1 on: May 09, 2019, 08:26:35 pm »
I am seeing a hell of a lot of lead inductance there and copious potential for noise pickup in important things like transistor drive connections, also some very small leads in what I would expect to be high current places.

Can we have a circuit diagram of exactly how you have that lot hooked up (including such things as exactly what is connected to what ground wise).

Have you set the phase lead and over current thresholds as described in the UD2.7 documentation to suit your primary and cap bank (These will need readjusting any time you change either)? You should not be blowing IGBTs.

Does the secondary resonate at the same frequency as the primary? Not going to get much spark action if not, but the electronics setup can be done without the secondary even installed (Tesla coils are LOOSELY coupled systems)!

 
The following users thanked this post: justin66

Offline justin66Topic starter

  • Regular Contributor
  • *
  • Posts: 115
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #2 on: May 09, 2019, 09:29:39 pm »
hi, thanks for the reply,
i used internet cat5 cable for this task as a steve ward did for the GDT. this is just a temporary setup to get it working with at least 5cm sparks its meant to be easy to change things up. and after some reasonable results assemble a proper stand for the TC and put all LV electronics inside aluminum and sheet boxes for higher power use. at least this was the plan, correct me if im wrong. im planing on building wood stand today.

should i remove all the connectors and solder in the wires directly? im not sure if it will help that, but i will do this if you say so.

the leads are 1.5mm2 just a temporary/testing solution as well. what thickness wires i should use?

connection diagram is this https://ibb.co/zhnWbXQ BTW ground is my apartment central heating radiators that is for sure ground. we dont have ground coming out of outlets.

i used to use phase inductor as fixed value 10micro henries. recently i recieved DIY adjustable inductor and i made it roughly 10microhenries. it did change the sound of the GDT when adjusting. so i set it to the most audible sound setting. i didnt have a chance to use osciloscope yet, i can probably arrange it tomorrow.

as of adjusting the OCD i didnt think it is necessary as im not going to run it to high sparks, i also keep track of the variac current and it doesnt rise what so ever using my digital clamp-multimeter

i didint think measuring resonances mattered much because i calculated this in java tc. im going to do this probably tomorrow as i get a chance to use the expensive equipment.

the strange part is before the last blow it kind of worked with the sparks. now im not sure if CDE caps are blown they measure mostly full capacitance with in my eqipment mesuring tolerances. some read a little bit less. when i change phase jumpers the TC refuse to work by the sound of GDT.

i tried using steel pancake for primary testing, the results are the same, no current, my pc speakers react to the operation by making interupter sounds assuming that the UD adjusts the drive frequency  to the primary it is very strange that nothing happens, other people indicated that the pan should heat up. not in my case. and im sure it is compatible with induction heaters.

i know im doing something wrong but i cant pin point what it is. maybe some ideas?
 

Offline daedalux

  • Contributor
  • Posts: 34
  • Country: es
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #3 on: May 09, 2019, 09:45:28 pm »
Have you tried to drive an artificial load made out of resistors with simulated feedback? If that works okay shield as well as you can. You can serve yourself with aluminium tape and perhaps coaxial cable in some places. Any IGBT or MOSFET is very likely to get killed with high voltage and frequency just around.
 
The following users thanked this post: justin66

Offline justin66Topic starter

  • Regular Contributor
  • *
  • Posts: 115
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #4 on: May 09, 2019, 09:49:16 pm »
thanks for the reply, i dint tried this, how do i make this? do i just connect some resistance across Hbridge output? what resistor value i should use for this application? and what variac voltage?
BTW im using 4pcs of FGY75N60 IGBT
the Hbridge driver has some TVS diodes, i thought they are supposed to take care of the overvoltage spikes? the schematic is in the first post at the Hbridge link
« Last Edit: May 09, 2019, 09:52:20 pm by justin66 »
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #5 on: May 09, 2019, 09:58:01 pm »
What are you doing wrong? Just about everything. Starting with choosing the most difficult kind of Tesla Coil to get working properly. You have a bit of a learning curve ahead of you. The first thing to be aware of is that when you _do_ succeed you will be generating extremely high voltages which laugh at insulation. The only thing that will prevent damage to equipment and components is proper layout and _spacing_. It isn't just a matter of stray inductances although that is important also.
My advice is to build your primary and secondary in such a manner as to respect the high voltages in the secondary and the high currents available in the primary. Clean up the layout and construction, then work on refining the electronics, first and foremost by getting rid of the stray inductances. You are going to be generating a big oscillating EM field and all those wires will be picking it up and causing crazy behaviour of your circuit.
Here's a coil that was designed using the same Javascript designer, a hybrid SS-SG TC running on 24 VDC:




The easiest person to fool is yourself. -- Richard Feynman
 
The following users thanked this post: justin66

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #6 on: May 09, 2019, 10:10:12 pm »
There are HUGE currents flowing in the primary circuit loop (primary coil, cap bank, connections to the IGBTs, and these connections need to be REALLY robust (It is these currents that the over current sense is designed to protect things from, and there is a reason the documentation talks about hundreds of amps).

The calculator will get you in the ballpark, but you HAVE to measure, Q factors are (should be) quite high, and no calculator application will deal with the specifics of your particular coil.

I am highly suspicious of a gate driver that is making significant noise, they really should not be running anything like that much flux.

Please tell me there is an isolation transformer in play after that variac? Otherwise you have exposed mains (Actually worse, rectified mains!) all over the shop, and that is way more dangerous then the secondary high voltage.

As to a resistive test load, I would suggest maybe an old school 100W filament bulb with the Variac set to 70V or so, bulb in place of the primary and cap bank, and a signal generator at whatever you calculated resonance to be hooked up to the appropriate spot on the driver board.

Be careful scoping this thing, you really want a differential probe as as a lot of this is high voltage and fairly high energy, lots of easy ways to die.

I notice the lack of a grounded (at one end) anti arc ring above the primary, this is helpful as it both makes a spark going back to the secondary and damaging it less likely and prevents an arc between the HV end of the secondary and the primary from causing dangerous amounts of current to flow.

Regards, Dan.
 

Offline justin66Topic starter

  • Regular Contributor
  • *
  • Posts: 115
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #7 on: May 09, 2019, 10:52:58 pm »
alsetalokin4017 thanks for the repply i will clean up my circuit ASAP and place everything in faradays cage underneeth
 

Offline justin66Topic starter

  • Regular Contributor
  • *
  • Posts: 115
Re: DRSSTC TESLA COIL problem and MONEY reward for working solution
« Reply #8 on: May 09, 2019, 11:12:37 pm »
dmills, thanks for the reply. as of right now it is not isolated, it is just for testing to get it working. final design will have isolation transformer. i still yet to determine the optimal power and i dont want to waste money on wrong parts. im not anywhere near primaries when operating. im aware of dangers associated with operating and handling such voltages

final design will have strike rail as well, i dont need it yet, no sparks achieved  |O
the GDT makes noise only when the TC is not operating correctly, as soon as i get it working the GDT goes silent and my speakers start making interupter poping sounds, assuming it is working.
i rethought my design again. i think i killed that expensive cde cap bank :(  i plan to clean it up, increase primary windings to at least 9 turns or more and decrease tank capacitor
acording to java tc calculations it should look like this. what do you think about that?
https://ibb.co/GV6Wgrn

OFC i will run light bulb test with frequency generator to check for potential flaws.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf