High Voltage Forum
Tesla coils => Solid State Tesla Coils (SSTC) => Topic started by: bogdan on February 03, 2019, 10:06:21 AM
-
Hello, i am new to this forum and i am searching for some answers, i build a secondary for my tc, It is build from 80cm of pvc pipe with 110mm diameter, magnet wire is 0,118mm...a lot of windings, top load 110mm aluminium pipe, primary 10 turns of 2.5mm copper wire, i made a slayer exciter circuit and found out it resonates at ~141kHz, now i am stuck at the question of how to drive it, any answer is welcome
-
Hi bogdan and welcome to HVF
You have made a very high ratio secondary coil, that can hinder performance a bit, but do not worry too much about that now, you are on a learning path :)
Could you please post some pictures of your coil?
I would say you should start with a half-bridge steve ward mini sstc / sstc5: http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-i/ and when you have that working you can upgrade to a full-bridge version: http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-ii/
The SSTC is fairly simple in terms of driving topology and feedback, contains very few parts and will give you a fair spark length.
A DRSSTC is more advanced and here there is many more fallpits if you are not familiar with Tesla coils in general, so a SSTC is a good place to start.
-
here are some photos, i checked the schematic, and i ca't find the UCC37322 and UCC37321, can you please provide some similar ic that are easier to find?
[ Invalid Attachment ] [ Invalid Attachment ]
-
Search for TC/MCP/MIC preffix parts like the TC4429,just make sure you have one inverted and one non-inverted version,with high output current capability and enable pin if you want to easily use an interrupter.
The second option is to use an NPN/PNP output stage,like this:
(https://www.mikrocontroller.net/attachment/284427/U3RBsZAI.png)
The secondary coil ratio is really high,as Mads said,with just under a 1:4 ratio secondary coil,the streamers were limited in my case,so had to wind a smaller secondary.
-
The first schematic looks a lot more easier, i have tc4427, but it doesn't have the enable pin for the interrupter, so i still need some replacement options.
-
So i woke up this morning, i am free from work, and what can i do better that online searching, among replacement for the ic i am missing. I found out that i can do better with my slayer exciter circuit, with a mosfet and a mosfet driver. I had lying around an irfp150n and a tc4427, said and done, grabbed my breadboard, and started doing stuff. I decided that i will power the ic from a 5v battery power pack, so i did, the + from power supply, and from the power bank i connected together. At around 8 volts it draws about 1A, and at 12V about 3A, happy about it, grabbed a screwdriver and made about 1 cm of arcs, turned on my oscilloscope, and i realized 2 things happened
1-the frequency dropped
2-i forgot to turn on the battery pack
i turned on my battery pack, the power draw went down, and the sparks were gone
wo power bank (http://i66.tinypic.com/kb93t5.jpg)
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2Fi66.tinypic.com%2Fkb93t5.jpg&hash=281e5183466fc161356903f0826df444b4c61081)
with power bank (http://i63.tinypic.com/2i7nxoo.jpg)
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2Fi63.tinypic.com%2F2i7nxoo.jpg&hash=75bc035111c7fd82485a1146a779c027a852692f)
now, i don't know what happened, but i would like for someone to explain.
edit: aparently, idk how to insert links here...
-
4046 drivers can be dificult to use and you should have a solid 12 and 5v reglated supply to start this so you can rule it out.
a simlpe 7812,7805 could be built. even some ebay power supply or switcher will do. do you have a picture of your circuit
-
yes, this is the circuit
-
So i decided to go with a half-bridge steve ward mini sstc, after a lot of reading and reading, i still have a few questions
-is it ok to use polyester caps for the 100nF including the one right before GDT?
-is there some way to test the permeability of the GDT (i saw a lot of problems regarding this one)
The rest of my components are on their way, so only waiting now.
-
Both polypropylene and polyester have about the same properties when it comes to be used as coupling/DC blocking capacitor. So yes you have use a polyester capacitor, only disadvantages
Testing a GDT could be done from these two resources, albeit it might be easier to just order a suitable core. http://thedatastream.4hv.org/gdt_index.htm and http://www.richieburnett.co.uk/temp/gdt/gdt2.html
-
ok, thank you very much, i will wait for my components, and keep you updated :)
-
Hi bogdan, I'm new to this forum as well. I started building solid state circuits for Tesla coils about a year ago, but for GDT's, I have had good experiences with EPCOS toroids made from type n30 material and cat 5 wire, as well as commercially available pre-wound transformers like 673-PH9400233ANLT from Mouser.
If you do end up winding a smaller secondary, the slayer resonator circuit you were using is where I started as well. The power supply I used was just a mid-sized 60hz iron core transformer, bridge rectifier and electrolytic as a bare bones unregulated supply.
-
So, after a long long wait, i finally got all of my components, and here is the pcb, not my best work, but i promise if this works, i will make a perfect one for the full bridge version. Now what is the best way to test this ?
-
I'am not an tesla coil expert, but you could try to use very low supply voltage like 12V and verify the signal and the resonance.
-
So, after a long long wait, i finally got all of my components, and here is the pcb, not my best work, but i promise if this works, i will make a perfect one for the full bridge version. Now what is the best way to test this ?
Driver:
Use a multimeter to verify that GND and 12V isn't shorted (large cap might look like a short), or use a short-proof powersupply (like a computer psu), or . Disconnect the GDT and hook the 12V up. Then use a multimeter to test the voltages (5V, 12V). Then apply a squarewave (at your wanted frequency) to the feedback and use an oscilloscope to see if the signal gets processed correctly (hex inverter, ucc input and then the ucc output). Then attach the GDT and make sure that the mosfets/igbts get the drive signal.
Further testing:
Make sure that the mosfets don't turn on at the same time/phasing. Test if you have the mosfets connected the right way. Then apply a low voltage (12V - 30V) to the bridge for a short time and check if the mosfets are warm. If not everything works fine. Then you can hook the bridge to the primary and make sure that it doesn't draw too much current. Then add the secondary and it should give you the first arks. (At this point you probably can increase the voltage, but always be careful and check if anything gets warm (heatsink, caps, chips,...) and never let it run too long without checking if anything gets hot)
Feedback:
Then you can add your feedback (antenna/secondary base transformer) and try it out. You might need to change the phasing if it doesn't work the first time.
Honestly I think your board look great, but you will most likely need a bigger heatsink. Those Mosfets will heat up quite a bit with this small heatsink. Just adding a fan will probably be not enough.
Some questions:
- Did you test the frequency of your secondary coil with a function generator?
- Did you calculate the Impedance of your Primary coil?
- Do you have a variac?
-
Make sure to never run the UCCs without load.If you only want to check the logic circuit,better pull them out from the sockets or leave the GDT in place.The input signals to the UCCs should be the same as one of them inverts it so they switch alternately.Do you have gate resistors,reverse diodes and zener/TVS diodes for the MOSFETs?Is your GDT suitable for this application?
-
Ok, a lot of testing for this weekend, my resonance should be at @141kHz(measured with slayer exciter circuit), i don't know my impedance of my primary, i don't have a variac, but i have several transformers with various dc output, up tp 60v should be fine for test purposes, gonna be harder to change the voltage but, oh well...
I have no short on the pcb, i powered it up with the ic out of the socket, everything is ok
"Make sure that the mosfets don't turn on at the same time/phasing. " i don't own a 2ch oscilloscope, can i verify in other way ? There were no zeners in the schematic, and the gdt should do fine, it's from a buck converter which supposedly worked at 300kHz
-
"Ok, a lot of testing for this weekend, my resonance should be at @141kHz(measured with slayer exciter circuit), "
That should be quite accurate, but look at the next point
"I don't know my impedance of my primary"
You should check your coil with JavaTC, so you have an idea about how much current can flow, and many more useful informations. Better safe than sorry.
"I don't have a variac, but i have several transformers with various dc output, up tp 60v should be fine for test purposes, gonna be harder to change the voltage but, oh well..."
That should work fine
"I have no short on the pcb, i powered it up with the ic out of the socket, everything is ok"
Great
""Make sure that the mosfets don't turn on at the same time/phasing. " i don't own a 2ch oscilloscope, can i verify in other way? "
Medium effort:
Disconnect the Primary. Put some kind of load between your positive input and the positive bus of your board. If the mosfets are in phase you will get current flowing. If the is close to no current everything is fine.
Low effort:
Disconnect the Primary. Connect the positive voltage to your positve bus for a short time. If it sparks, thats bad. If not, then not.
"There were no zeners in the schematic, and the gdt should do fine, it's from a buck converter which supposedly worked at 300kHz"
You can always add electrical safety component to your board, even if the schematic doesn't show any. You can look at other schematics and compare some components.
About safety components in general: Most of the time you don't need them, but sometimes they safe your work.
"Make sure to never run the UCCs without load."
Actually you can run them without any load, just don't run them for 5 minutes straight and check if they get warm. That would be a great time for a break.
EDIT: I am actually wrong here
"If you only want to check the logic circuit,better pull them out from the sockets or leave the GDT in place."
The first part if true. But if you have no input signal and have the GDT connected the UCC's don't switch. That basically shorts (through the UCC's andGDT) 12V and GND together. Thats how I lost 9$ once.
-
"Make sure to never run the UCCs without load."
Actually you can run them without any load, just don't run them for 5 minutes straight and check if they get warm. That would be a great time for a break.
I'd never run them without load.Many failed for me right after powering them up without load. ;)
"If you only want to check the logic circuit,better pull them out from the sockets or leave the GDT in place."
The first part if true. But if you have no input signal and have the GDT connected the UCC's don't switch. That basically shorts (through the UCC's andGDT) 12V and GND together. Thats how I lost 9$ once.
There's a DC blocking cap in series to the GDT.Not sure if its inrush current is enough to destroy the UCCs. :)
-
so, i ran some tests, everything was fine, i connected the primary and tested it at 12v, but i added a 0.47 resistor in series so i don't burn things, was this a mistake ? and then ran it for a few minutes, it draws about 100mA@12v with the resistor, everything is cool, except the ucc, they got pretty warm, what do i do in this case? The whole experiment was done at 100kHz from a crappy signal generator.
-
anyone any idea ?
-
Please upload some gate waveforms to see how the UCCs perform.