SSTC V4.0 - Large Solid State Tesla Coil @ ~1kW of Power!
HTML-код
- Опубликовано: 6 сен 2024
- Hey everyone, this is an old video of an old coil I had running. I've since made better and more powerful or more advanced SSTCs, so be sure to check out the other videos on my channel. I've made all the SSTC data available for free to download from my blog here: sigurthrenterpr...
Please feel free to check it out and make use of the data where needed! Just be sure to give credible mention if you post part or whole of the data elsewhere, thanks!
------------------- Original Description --------------------
I finally fixed and rebuilt the driver for my large Tesla Coil. This uses the resonator from my old large SGTC. Here's the results!
FDL100N50F Half Bridge powered by rectified mains wall power.
UCC37321 UCC37322 Gate Drive Chips Pair
74HC14 Hex Schmitt Trigger Inverter antenna feedback system.
Pri: 10T of 10AWG hi-temp 600VAC rated insulated wire directly on secondary coilform with 1mm polyethylene insulator.
Sec: 18.5" of 30AWG on 4.5" OD PVC coilform.
f-res: 198.5KHz
Input Power: 1110VA (120Vrms @ 9.25A)
Output Power: ~965W of RF.
1 foot long very hot CW streamers.
In the configuration shown in the video it is Steve Ward's SSTC 5 (just google that). I rater revised it by going to my own take on the famous "DWSSTC" which is 4046 PLL driven. Up to 3kW power now.
Hello again,
Yeah, just noticed the 74HC14 H S T I antenna feedback system.
(DUHHH!!! I cant believe I missed it)
I'll have no problem finding the part from somewhere. (Digi-Key wasn't carrying)
Keep up the high quality, very good work.
All the best!
Ben
have you made this yet?
Tesla Explorer No, I bought the transistors and blew them up with other projects... They are expensive transistors... Also, I have been finding it hard to achieve time for anything due to working 50+ hours a week at a shit-hole factory and my living conditions are constantly changing... Its a long story.. Anyways, this project is still on the to do list... I have an extensive to do list :)
Great video! I'm going to follow your design for my 1.0v attempt at a SSTC, thanks again!
I have been using the fairchild (now part of ON semi) FDP33N25's and they are a joy to use too! They have very simalar ratings to IRFP250N's, but are MUCH cheaper, (about $150 a piece,) in a more manageable TO-220 package, have lower gate charge, etc! They can take a bit of abuse too! (although I do still kill them occasionally.)
I keep getting a lot of comments and questions on this old video. All of the schematics, Bill of Materials (parts list), and build specifications are located here: sigurthrenterprises.blogspot.com/2015/12/sigurthrs-unified-sstc-data-file-all.html
Wow nice job :) cant wait for this "supprise"
I just checked and even at full power both of my UCC chips stay below 85F. My runtimes are never over 10min at 500W or 30sec at 1500W though. That being said I chose the mosfets I use for their low RdsON and low gate capacitance and I run my UCC chips on 12V (gate sees +-12v for 24Vp-p) instead of the popular 20V.
hey what are your thoughts on this driver into a full bridge of 260ns, do the ucc combos have the "driveablitiy"? great coil again!
hey, do you suppose you can give me the link to this circuit or post it if its your design? it would be very helpful!
Why not add a pwm/freq interrupter so that you can modulate arcs while achieving greater length at much lower currents?
At the time the coil shown only used antenna feedback and relied upon the gate chips to ping the circuit to initiate resonant feedback. The problem with not running CW (using an interrupter) is that what was only a transient response (running with no resonant feedback) becomes a regular occurrence. These chips ping at around 7.8MHz, which is WAY too fast for normal MOSFETs and IGBTs to handle, so subjecting them to repeated pings while there is current flowing will slowly destroy them.
I later switched to an oscillator driven PLL based design which allows for modulation and interruption without any drawbacks.
That being said, it isn't always about getting larger arcs. Having a huge spectrally pure E-field to play around with is a lot of fun too. Besides, modulation is loud as hell above 1kW.
Matt Giordano That is an interesting bit of information. Probably has caused me to alter the design of my next SSTC from antenna feedback to PLL, too. Nice job Matt, when you cranked it up and got the yellow flame, you know you are cooking.
You bet, stacatto mode in a kW class sstc will wake up the neighbors for sure.
Matt Giordano so theres no way to prevent the often hf buildup from damaging the 260n's within time while modulating? ive reviewed your site its very nice. it seems you have moved away from the ucc chips
Tesla Explorer Sure there is, you run a startup oscillator that is switched out (or often overpowered) when feedback is received. The PLL chip drivers use this technique, but it requires the driver be tuned to the individual resonator coil - which is almost impossible without a scope.
5:00: I don't know much about tesla coils circuits, but I have played with enough electricity to know by sight that something is wrong here. The green-white color of the arc is caused by persistent ionization of the air, either because the current is anomalously high, or because the frequency is anomalously low. Most coils of that sort would probably just give you a really unpleasant burn if they accidentally arced to you, this one might just kill you outright.
***** You said it yourself; you're not familiar with tesla coils. The reason the plasma isn't the typical color or appearance is because it is running CW or continuous wave, not interrupted/pulsed. RMS current at the output is only about 30ma, and because the output is RF and beyond the frequency response of the human nervous system it is totally painless and cannot cause cardiac arrest or arrhythmia. It can of course cause serious, but non-lethal RF burns.
An aside, this video is painful for me to watch now, haha. I've made so much better systems since. It's tempting to take it down but it still get regular attention.
Estimates vary but the low end of the threshold considered to be lethal is 60mA. Ingoring the immediate consequences of direct exposure to current, that thing would give you a hell of a burn if you touched it for more than a tiny fraction of a second. What frequency is it actually running at? I can't exactly tell by looking but it doesn't have the continuous blue-white characteristic appearance of a well tuned coil, which implies it's probably lower than would be desired. If it's less than 1khz, it would probably kill you immediately if you touched it.
***** I don't remember the f0 exactly but the secondary resonates around 300KHz with that tiny topload. It resonates at around 170KHz with the new topload in newer videos. Arc color isn't a good indicator because not only current is a factor but also pulse width and rate if pulsed and RF envelope shaping if CW/QCW. Yeah it would toast your finger like a hot dog though!
With SSTCs (not DRSSTC, OLTC, VTTC, or SGTC) the secondary is always driven at resonance. The bandwidth of the coil is so low that driving away from resonance, as in a poorly tuned spark gap or DR type, results in no visible output what-so-ever. On my smaller coil which runs at 337KHz, the total (not classical FWHM) bandwidth is plus/minus 10KHz (327-347KHz). The FWHM bandwidth is only 7KHz (-+3.5KHz). So there is no way to have these run out of tune.
Matt Giordano Nice circuit. No it's not lethal. I can tell you that my VTTC designs have very similar outputs when set at some levels and so do many on line. Some cameras also make corona's look very different from others. This is not really the point here, but a thing to keep in mind for other videos. The most dangerous voltage in here is the line AC voltage mains. Period. This coil is not as dangerous as many think also since this it's RF here. Radio frequency at 300khz is actually pretty high. Too high for muscles to react the same as say 60, 1000hz. A few spikes are being produced by the AC ripple current in the DC (larger caps will bring this down but also make your sprites smaller, you may want to use less) Some may argue about how I describe this, but still we can agree you should never touch a coil like this. To say that it would be lethal due to persistence of ionization here however, is not true. While coils such as arc-gap and base-fed VTTC types should never be touched since they often have a secondary connected to the high current/voltage DC circuit itself or produce large low frequency currents. Many coils that run pretty well on RF are actually very safe to touch on the outside but ONLY to a certain point! (So don't try--data obtained from tests) After getting too close (trust me) burns happen very fast. An RF burn, since these things are nothing more then RF transmitters as mentioned above with voltages stepped up. My VTTC1200 (see on Mostlymacros channel) run's enough output to break a jar in about 3 seconds and thus I assume it would give a really nasty RF burn, however--my tests on fruit and current indicate that momentary exposure to the secondary of that VTTC would not require a hospital trip, just first aid. As far as RF AC, the way it travels through the body, it is far less likely to be lethal because it's at such a high frequency. The old timer TV repair guys were all too aware of with this as flybacks switch to DC output only! The body reacts differently to low frequency and DC. Low frequency is most dangerous. 60hz (120hz in the case of this circuit is seen as ripple current created by the bridge) and such is so dangerous because it moves muscles and interrupts nervous system. RF into 300kc (just above nav. beacons and below AM BC band) would give you a burn, and this kind of flame will cook a hotdog, but it's not quickly lethal. You got to get to far larger coils or currents for that in this type of situation. I've seen a guy get 20kv from an arc as thick as your leg go straight through him and he survived at 60hz no less! Don't get me wrong, 25years of experience tells me electricity is DANGEROUS and must be respected--but the only really lethal voltage here is the mains which you have done a good job of isolating and it looks like filtering. Coil not bad at all, well laid out. Normally, only 30ma at 30volts is considered "potentially dangerous to humans", that actually depends on a lot of factors. But it is INCREDIBLE what the human body can withstand. Tesla himself used to light up his ENTIRE BODY with flames like this and that's WITH no Faraday suit on!
Matt Giordano Hey, I'm kinda new to RUclips got a channel up called "Mostlymacros". If you want have a look at my VTTC and other stuff! More to come when my new tripod and 4KV cap gets here. I noticed you got -strike through- writing too. Don't use the "-" when you write. I had the same problem until I figured it out. Once before and once after will auto-format. I thought this was supposed to be ASCII! I can't stand that junk. Anyway, nice circuit.
Where did you get from the variac?
if ya didnt use halfbridge but fullbridge ya could get same result at half the voltage
excelente me gusta porque es sencillo
Please , can you help me? I made it , driver works , on the main transistors are being hot , but I don't get light
+Egister & Vladgister TV There are many factors involved, so I'll need as many details as you can provide. Assuming you used my data file I can provide possible causes of issue.
That being said usually if the FETs are getting hot it means one or more of a few things:
1) poor or insufficient heatsinking
2) insufficient gate drive (could be too low of voltage to the driver, if the power supply is sagging under the load)
3) poor quality gate drive (leads too long, wires too thin, wires not twisted in pairs, poorly made GDT, wrong GDT core, too much gate resistance used)
4) too high of resonant frequency chosen (higher the frequency, the more gate current needed, and the more heat dissipated in the fets)
5) no feedback signal present (this causes parasitic oscillation of the gate drive chips at very high frequencies, use an oscillation starter to manually "ping" the resonator, move antenna closer, etc)
Likely issues for low or no visible output are:
1) coupling is too low (raise primary, increase turns, use smaller diameter primary coilform)
2) grounding is too high impedance for RF (use thicker wire, move closer to ground, use a real RF ground)
3) too many turns on primary (low current draw)
4) too few turns on primary (poor coupling, high peak currents)
5) enable lines on gate drive chips are getting interference, causing random interruption of oscillation
6) low or no feedback signal present (move antenna, ping the resonator, etc).
Hope this helps!
+Matt “Sigurthr” Giordano GDT is perfect; there are 8 turns on primary; one question: what size will be perfect from secondary to antenna ?
+Matt “Sigurthr” Giordano and how many turns on GDT ? I made four , is it good ?
Egister & Vladgister TV No! No wonder you are having issues, that's no where enough drive the FETs. 12 turns minimum, I typically use 15 or 16. Make sure you wind it trifiliar/pentafiliarly and not as separate windings. Use this as a guide: thedatastream.4hv.org/gdt_index.htm
+Matt “Sigurthr” Giordano thank a
Lot , for your help
Nice :) Thumbs up :D
Alex
Where is interrupter ? I think , if you connect it , it will be better
schematic for the things?
+Keith Nikolaevich This is a very outdated video, however full schematics and data files are available for download on my blog site: www.SigurthrEnterprises.blogspot.com just look for the tesla coil related posts (there's not a lot of posts)
Matt Giordano thanks, i have built a standard tesla coil, and am interested in making one play music, but dont really know how, thanks
Keith Nikolaevich I didn't post the add-on board schematic for audio modulation, but it's simply a VCO (like a LTC6990 for example) running at the coil's resonant frequency connected to the USSTCC board input, and you feed headphone level audio into the VCO's input. Simple to add on once you get the basic system built and running.
How many terns I must make on antenna ? Please help!
+Egister & Vladgister TV It doesn't matter if the antenna is coiled or not, you only coil it to save space so it is farther away from the high voltage areas. Never let high voltage near the antenna.
+Matt “Sigurthr” Giordano I know about it , but thanks a lot
+Katarina Radivojevic Hello! You can find full specifications in my data file at this link. It isn't for this exact coil but it contains all the information to make a working coil very close to this one (better, even!) tinyurl.com/nmkm2vd
+Katarina Radivojevic Sorry, I'd been busy so I hadn't checked google notifications. Fully completed from all new parts a coil of my design costs about $100 USD for all the electronics components, heatsinks, wire, and electrical related hardware (prototype circuit board, solder, screws, low voltage power transformer, etc). The cost of the physical hardware varies by location; for me it is quite expensive - over $70 USD, but my customers have said they can get the coliforms, coatings, and base materials for under $40 USD total. Unfortunately I am not able to build Tesla Coils for other people. I live deep in the rural north, so it is very difficult to acquire materials here, especially in the winter like it is now. Also, shipping of a completed high power tesla coil is very expensive from here. I recently sold a few spare parts and an empty printed circuit board to a customer and it cost over $30 shipping, and he was only half a country away.
If you are unable to build a high power tesla coil yourself you can buy a low power one already made. A Russian acquaintance of mine sells them inexpensively on eBay and privately, his coils may be suitable for you. Like my coils they are best appreciated by those interested in illuminating gas tubes or experimenting with wireless power transfer.
Here is his coil on eBay: tinyurl.com/gms622p
Here is his private site: tinyurl.com/hqlk9bt
Katarina Radivojevic Yes, coils like what Nikola himself made, we call SGTC - spark gap tesla coil, are the simplest and cheapest to make. Good luck!
Are the irfp260's and 250n's just destine to blow? Why did I even buy these lol! So far 10 of them have bit the dust... Silently too... I'm going for 450n's in my near future, the datasheets show they double the 250's/260s....
I later rebuilt all my coils to use a different driver and bridge configuration. A major factor in FET lifetime is the use (or lack there) of protection, isolation, and flyback diodes. See my site for specific details and free schematics. www.SigurthrEnterprises.com . I haven't blown a single fet since.
Man thank you so much. I'm on it!
it's,very good
What transistors do you use ?
+Egister & Vladgister TV Fairchild's FDL100N50F
+Matt “Sigurthr” Giordano thanks a lot
+Matt “Sigurthr” Giordano thanks a lot
+Matt “Sigurthr” Giordano can you said me characteristics of this transistor
Egister & Vladgister TV Here is the datasheet www.fairchildsemi.com/datasheets/FD/FDL100N50F.pdf
Can it be bigger? like a lightsaber? :D
It can XD
But for tesla coils with a light saber length, you will need more width. So that would be a quite big lightsaber and the heatsink of transistors will make it heavy XD