Hi Steve. You’re the man who started it all for me with the excellent schematics you’ve designed and shared. I think most Teslacoilers would greatly appreciate a TH-cam series of your classic teslacoils like your DRSSTC.5 and 1etc.
The man, the myth, the legend! So grateful for the information you've given to us all. Awesome to see a new video and it's even an HFSSTC. This output and efficiency is mindblowing, I thought you were using a TUBE at first!! Just when I thought I was understanding these things...
Man, you are a genius! Do you know that? I am always stunned by your builds and they have been quite an inspiration for me. It's good to see another video from you. This flame looks awesome, and the efficiency is outstanding. This really shows that with proper tuning and knowledge, one can build systems unlike any before. Anyway, could you please explain, what are the advantages of SiC fets compared to regular MOSFETs? Thanks and keep up the great work!👍
As an alternative to traditional silicon MOSFETs, silicon carbide MOSFETs offer the advantages of higher blocking voltage, lower on-state resistance, and higher thermal conductivity. The devices can replace silicon MOSFETs and IGBTs in many applications.
Since SiC is more efficient (lower resistance material) the chips are much smaller for a given current capacity, so the internal parasitic capacitance between Gate, drain and source are also much smaller, which is helpful compared to Si devices. The JFET Cascode parts im using from United SiC offer an additional advantage for this circuit, which does not use a gate driver to clean up the gate voltage. The cascode uses a low-voltage Si MOSFET to control the JFET, this internal Si fet gives the device a higher gain (more drain amps per gate volt) which is beneficial when the gate signal is a sine wave! I have not tried using *actual* SiC MOSFETs in this duty, there are some concerns over needing high gate voltage for low drain resistance, but being also limited to -5 or -10V on the gate because of breakdown or reverse conduction degradation issues. In this case, using a gate driver to buffer and control the drive voltage would be a good call.
Dang that's cool! Is it possible to audio modulate it? If you made the secondary of litz wire, could you decrease the power loss in it? I can imagine it being more practical to use copper pipe for the primary and dealing with the power loss by running coolant through it if nessesary. Probably mineral oil or other nonpolar inert liquid.
I originally used litz wire, but it could not stand the heat, and i believe the insulation between strands melted as the wire became stiff and crunchy. The "200*C" magnet wire im using seems to hold up perfectly fine. Primary doesnt need any extra cooling, but i might build a more powerful one that does. Might be neat to re-use a CPU cooler with liquid coolant loop. Audio modulation can be achieved by injecting audio signal (about 1Vpk works well) into the gate bias voltage via 10u coupling cap. It's not as good as "bus" or supply modulation, but it was pretty decent for almost zero effort!
What would the output be like if the secondary was actually tuned to the primary as closely as possibe, or would that lessen output while overloading the fet ?
Great project ! Can you change the output frequancy of circuit to about 100MHz and convert it to a radio frequancy transmmiter and connect an amplifier output to the circuit ?
As always, awesome build. Did you know with double resonant (primary and secondary), you can tune into 2 resonant frequencies? the first (transverse resonance) is in phase between primary and secondary voltages, The second (longitudinal resonance) which is tuned higher, has 180 degrees out of phase resonance, with much higher voltage on the secondary.
Hey Steve, Your coil is great! Ive built one of these myself but yours performs at least 100% better at the same voltages. I was using cheap 460's but i bought some 1200v SiC MOSFETs from aliexpress that are legitimate and wow, what a difference it made! What is the purpose of the coil between the voltage input and the MOSFET gate, and how does it affect performance?
Hmm, coil on MOSFET gate? I didnt think i posted anything about that. I have been using resistance/inductance in the "feedback path" to the gate to properly tune the phasing between drain (driving) voltage and primary (load) current. It seems everyone else missed this detail in their HFSSTC, and its crucial to getting significant power output for a given drain voltage. Its like running a bridge driven tesla coil with bogus phase in the feedback loop, it CAN oscillate, but the power factor can be garbage, so its mostly just reactive power and not useful for pumping the flame. The drain voltage pulse should start at the negative voltage peak on the tank cap (coincides with tank current zero crossing). Most circuits dont have the right delay for the gate drive, and so the drain voltage pulse happens early before the primary current zero crossing, so all that driving voltage that's too early, cancels out some of the useful part. Study: power factor, dot-product... if this comment doesn't make sense :P.
What is the temperature of such a plasma flame? Could it be used, for example, instead of a gas flame in a gas stove to heat a house?What is the temperature of such a plasma flame? Could it be used, for example, instead of a gas flame in a gas stove to heat a house?
It's going to be the same spectral emission as most other air-plasmas, albeit much hotter than typical Tesla coil sparks, so it may be a broader spectrum than normal streamers. Of course, that spectrum will also contain emission lines from the metal ions, vaporizing off of the metal's surface. I was able to find an open-access article, where they measured the emission of an air plasma, generated with what looks to be a Jacob's-Ladder: www.researchgate.net/figure/Emission-spectrum-of-a-gliding-arc-discharge-in-atmospheric-air-at-a-wavelength-range-of_fig11_312492221 That spectrum should be pretty close to the plasma spectrum from Steve's coil. Note that it looks to extend well-past the visible band, in both directions. So you are correct; there is plenty of UV and IR. Much more than visible light!
Could the output power be controlled using a delta sigma pulse skipping approach ? I was thinking in something like that to control the output power of an 80m band class E transceiver
Possibly, but that would require enough Q-factor to keep the drain voltage oscillating during the skipped pulse, so that there is still ZVS turn-on opportunity when the pulsing resumes. Practically speaking, i have my doubts, since maintaining enough Q factor is sort of tricky already, but if you arent making plasma flame, it might just work fine.
@@SteveWard It did not work! The Q was not enough to sustain oscilation at the output center frequency. Skiping half the cycles was already bad, as the output frequency dropped to the new half frequency.. really interesting.. the Q would need to be at least 10x bigger, or more.
This schematic is not on your web site , would aprecciated if you can show it , anyways it seems pretty tricky to get it working but I would still like to learn more about it
Haha, thanks! I do a lot of power electronic design/testing for my "real" work... this $7 fet doesnt even scratch the surface of what I deal with on a regular basis, but at the same time it "stings" when i break anything. Helps to have 20 years experience with these things... surely blew up way more mosfets in those first few years than i have in the last 10. I generally work very hard to find root-cause for any failure, it can be expensive, but capturing waveforms as the transistor lets out the smoke can often reveal the issue, and experience can provide the solution. I see you working on these machines without a scope... that's a difficult way to learn about what works and what doesnt! If you can put some transistor budget aside for a cheap used scope, it should pay for itself eventually :-).
I don't think it would need to be water-cooled; a similar sized piece of plain-old tungsten would already be a massive improvement over the stainless steel, considering the fact that tungsten's melting point is over twice that of stainless steel! Plus, I think a water cooling setup might become a little complicated, since it needs to run with deionized water only (otherwise, most water is conductive), which, at this high of a frequency, may as well conduct a little bit of current, robbing your breakout of some voltage and diminishing potential flame-size.
@@nabzim I agree. Simply switching to a similar sized peice of tungsten would most likely be sufficient. Water cooling would add complexity but could be done with the proper setup. I mean, there were water cooled vacuum tubes that operated at 800 MHz.
A brief experiment with a "voltage doubler" (level shifter) circuit, just like microwave oven or VTTC uses, showed that these sparks need a very long time to heat up and grow. The "ramped" power input resulted in a flame of maybe 2/3 the length of CW mode, and the peak input power doubled. Will be making a modulator eventually, as ~4mS ramp was not satisfactory.
If a piece of tungsten is hard to obtain at the moment, then maybe consider a piece of graphite; at least it will never melt! (graphite doesn't have a liquid phase at atmospheric pressure, thus, no melting point.) Carbon sublimes close to 4000K (which is higher than tungsten's melting point) so the graphite would act as a consumable electrode that needs replacing every-so-often. Also, just curious: Is it annoying that most Tesla-coilers practically worship you? (I also slightly do... I apologize). I can understand that fame is sometimes unwanted and a burden to bear. I hope you find my suggestion helpful, and, maybe it's a nice break from the typical comments of praise (and requests) you normally get.
I think you're right, graphite is probably the best answer. Pencil lead held up better than i expected - ive got some gouging rods on order. Any metals ive tried (including tungsten, brass, stainless steel, tool steel, aluminum, copper) eventually reach a surface temperature that causes some metal vaporization, and that effect is too bright to look at, hurts the eyes. I don't mind some attention for doing cool stuff, hopefully my influence on anyone that is a fan of mine is a good one. I hope that if nothing else, I can inspire others to experiment and learn, and that can carry over to other parts of their lives that might actually be more useful than just making pretty sparks :-).
Just tried 1/4" tungsten rod. Even the square-cut end turns as bright as a lightbulb filament, painful to look at, no bueno. I'm waiting for some carbon gouging rods, I think any metal will eventually vaporize in a brilliant bright light.
@@SteveWard yeah sounds like you need some way to dissipate that heat from the electrode.. maybe a solid copper electrode with heatsink fins and then a forced air blower with the air coming up from the bottom though the primary and secondary to give it enough cooling so it never reaches glowing temps. This should also help cool your secondary if you want longer run times.
Hi Steve. You’re the man who started it all for me with the excellent schematics you’ve designed and shared. I think most Teslacoilers would greatly appreciate a TH-cam series of your classic teslacoils like your DRSSTC.5 and 1etc.
Thanks! Sorry, none of those old machines even exist anymore, but I do want to start publishing projects from the past 10 years or so.
Have you published any of these projects yet?@@SteveWard
Awesome as usual!
The man, the myth, the legend! So grateful for the information you've given to us all. Awesome to see a new video and it's even an HFSSTC. This output and efficiency is mindblowing, I thought you were using a TUBE at first!! Just when I thought I was understanding these things...
Holy shit bro, you never cease to amaze me! Damn..
Man, you are a genius! Do you know that? I am always stunned by your builds and they have been quite an inspiration for me. It's good to see another video from you. This flame looks awesome, and the efficiency is outstanding. This really shows that with proper tuning and knowledge, one can build systems unlike any before. Anyway, could you please explain, what are the advantages of SiC fets compared to regular MOSFETs? Thanks and keep up the great work!👍
As an alternative to traditional silicon MOSFETs, silicon carbide MOSFETs offer the advantages of higher blocking voltage, lower on-state resistance, and higher thermal conductivity. The devices can replace silicon MOSFETs and IGBTs in many applications.
Since SiC is more efficient (lower resistance material) the chips are much smaller for a given current capacity, so the internal parasitic capacitance between Gate, drain and source are also much smaller, which is helpful compared to Si devices. The JFET Cascode parts im using from United SiC offer an additional advantage for this circuit, which does not use a gate driver to clean up the gate voltage. The cascode uses a low-voltage Si MOSFET to control the JFET, this internal Si fet gives the device a higher gain (more drain amps per gate volt) which is beneficial when the gate signal is a sine wave! I have not tried using *actual* SiC MOSFETs in this duty, there are some concerns over needing high gate voltage for low drain resistance, but being also limited to -5 or -10V on the gate because of breakdown or reverse conduction degradation issues. In this case, using a gate driver to buffer and control the drive voltage would be a good call.
VERY IMPRESSIVE.
I could watch it all day. Nice work
Hey Steve! Thanks again for helping me out with my senior project all those many years ago!
Steve - Father of Solid State Tesla Coil! Thanks!
Exactly🔥💯💯💯
Nearly beyond belief, Steve! Excellent job & thanks 🙏 for the interesting overview!
That is super cool! :) Good job sir!
Just as all your other vids are, this one does not fail to impress!
Hi Steve. This is an incredible discharge! You are very cool!
Awesome output as per your usual efficiency !
hi, I am wondering why the secondary oscillation frequency is two times that of the primary one?
Awesome 👍
Very interesting!
Dang that's cool! Is it possible to audio modulate it? If you made the secondary of litz wire, could you decrease the power loss in it? I can imagine it being more practical to use copper pipe for the primary and dealing with the power loss by running coolant through it if nessesary. Probably mineral oil or other nonpolar inert liquid.
I originally used litz wire, but it could not stand the heat, and i believe the insulation between strands melted as the wire became stiff and crunchy. The "200*C" magnet wire im using seems to hold up perfectly fine. Primary doesnt need any extra cooling, but i might build a more powerful one that does. Might be neat to re-use a CPU cooler with liquid coolant loop.
Audio modulation can be achieved by injecting audio signal (about 1Vpk works well) into the gate bias voltage via 10u coupling cap. It's not as good as "bus" or supply modulation, but it was pretty decent for almost zero effort!
It's not JUST a plasma flame, it's a SUPER PLASMA FLAME!!
What would the output be like if the secondary was actually tuned to the primary as closely as possibe, or would that lessen output while overloading the fet ?
Great project !
Can you change the output frequancy of circuit to about 100MHz and convert it to a radio frequancy transmmiter and connect an amplifier output to the circuit ?
As always, awesome build.
Did you know with double resonant (primary and secondary), you can tune into 2 resonant frequencies?
the first (transverse resonance) is in phase between primary and secondary voltages,
The second (longitudinal resonance) which is tuned higher, has 180 degrees out of phase resonance, with much higher voltage on the secondary.
It is show (TEM and LMD resonance) in this video:
th-cam.com/video/c2_rKc7sHGk/w-d-xo.html
nice to see a new one
Cool coil Steve
DAMN Nice! Greetings from Russia)
Hey Steve, Your coil is great! Ive built one of these myself but yours performs at least 100% better at the same voltages. I was using cheap 460's but i bought some 1200v SiC MOSFETs from aliexpress that are legitimate and wow, what a difference it made! What is the purpose of the coil between the voltage input and the MOSFET gate, and how does it affect performance?
Got a link to those fets?
search C2M0080120D on aliexpress
@@iLikeDots1 cool thanks
Hmm, coil on MOSFET gate? I didnt think i posted anything about that. I have been using resistance/inductance in the "feedback path" to the gate to properly tune the phasing between drain (driving) voltage and primary (load) current. It seems everyone else missed this detail in their HFSSTC, and its crucial to getting significant power output for a given drain voltage. Its like running a bridge driven tesla coil with bogus phase in the feedback loop, it CAN oscillate, but the power factor can be garbage, so its mostly just reactive power and not useful for pumping the flame. The drain voltage pulse should start at the negative voltage peak on the tank cap (coincides with tank current zero crossing). Most circuits dont have the right delay for the gate drive, and so the drain voltage pulse happens early before the primary current zero crossing, so all that driving voltage that's too early, cancels out some of the useful part. Study: power factor, dot-product... if this comment doesn't make sense :P.
Impressive!
Circuit diagram
So where do we order our solid state not-a-flamethrowers?
That's awesome like it :)
Circuit diagram and some information about coils? Plz? 😄
What is the temperature of such a plasma flame? Could it be used, for example, instead of a gas flame in a gas stove to heat a house?What is the temperature of such a plasma flame? Could it be used, for example, instead of a gas flame in a gas stove to heat a house?
quite ballzy to run that thing that close to your expensove looking (rigol 7000 series?) scope :P
Hello Steve Ward~I can't find your article about HFSSTC introduction~Can you give me an address?
Any schematics for this? Looks sick man awesome video
You kent privyu schematic detalizid?!))
Schematic please
Wow very impressive! Has anyone done spectroscopy to see what wavelengths are being emitted from this type of discharge? Are we getting some UV, IR?
It's going to be the same spectral emission as most other air-plasmas, albeit much hotter than typical Tesla coil sparks, so it may be a broader spectrum than normal streamers. Of course, that spectrum will also contain emission lines from the metal ions, vaporizing off of the metal's surface.
I was able to find an open-access article, where they measured the emission of an air plasma, generated with what looks to be a Jacob's-Ladder: www.researchgate.net/figure/Emission-spectrum-of-a-gliding-arc-discharge-in-atmospheric-air-at-a-wavelength-range-of_fig11_312492221
That spectrum should be pretty close to the plasma spectrum from Steve's coil. Note that it looks to extend well-past the visible band, in both directions. So you are correct; there is plenty of UV and IR. Much more than visible light!
Сool
Could the output power be controlled using a delta sigma pulse skipping approach ?
I was thinking in something like that to control the output power of an 80m band class E transceiver
Possibly, but that would require enough Q-factor to keep the drain voltage oscillating during the skipped pulse, so that there is still ZVS turn-on opportunity when the pulsing resumes. Practically speaking, i have my doubts, since maintaining enough Q factor is sort of tricky already, but if you arent making plasma flame, it might just work fine.
@@SteveWard It did not work! The Q was not enough to sustain oscilation at the output center frequency.
Skiping half the cycles was already bad, as the output frequency dropped to the new half frequency.. really interesting.. the Q would need to be at least 10x bigger, or more.
Sheesh it is so little but it got so much power
:D
This schematic is not on your web site , would aprecciated if you can show it , anyways it seems pretty tricky to get it working but I would still like to learn more about it
SiC like a boss
Other youtubers - "man these fets keep blowing unexpectedly"
Steve Ward - "Hold on let me crank this way more expensive fet to 1000 watts"
Haha, thanks! I do a lot of power electronic design/testing for my "real" work... this $7 fet doesnt even scratch the surface of what I deal with on a regular basis, but at the same time it "stings" when i break anything. Helps to have 20 years experience with these things... surely blew up way more mosfets in those first few years than i have in the last 10. I generally work very hard to find root-cause for any failure, it can be expensive, but capturing waveforms as the transistor lets out the smoke can often reveal the issue, and experience can provide the solution. I see you working on these machines without a scope... that's a difficult way to learn about what works and what doesnt! If you can put some transistor budget aside for a cheap used scope, it should pay for itself eventually :-).
A water cooled tungsten electrode from an old xenon arc lamp could possibly be useful here.
I don't think it would need to be water-cooled; a similar sized piece of plain-old tungsten would already be a massive improvement over the stainless steel, considering the fact that tungsten's melting point is over twice that of stainless steel! Plus, I think a water cooling setup might become a little complicated, since it needs to run with deionized water only (otherwise, most water is conductive), which, at this high of a frequency, may as well conduct a little bit of current, robbing your breakout of some voltage and diminishing potential flame-size.
@@nabzim I agree. Simply switching to a similar sized peice of tungsten would most likely be sufficient. Water cooling would add complexity but could be done with the proper setup. I mean, there were water cooled vacuum tubes that operated at 800 MHz.
You'd think so, but nope, the attachment point of the flame still vaporizes metal and makes a blinding bright light after some 20 seconds.
What do you think would happen if you modulated the supply voltage as is done with QCW coils? Could you make this produce longer straight arcs?
A brief experiment with a "voltage doubler" (level shifter) circuit, just like microwave oven or VTTC uses, showed that these sparks need a very long time to heat up and grow. The "ramped" power input resulted in a flame of maybe 2/3 the length of CW mode, and the peak input power doubled. Will be making a modulator eventually, as ~4mS ramp was not satisfactory.
Way to Smoke the competition
What's tank cap model number ?
Very cool
Circuit diagram ..?
🔥
If a piece of tungsten is hard to obtain at the moment, then maybe consider a piece of graphite; at least it will never melt! (graphite doesn't have a liquid phase at atmospheric pressure, thus, no melting point.) Carbon sublimes close to 4000K (which is higher than tungsten's melting point) so the graphite would act as a consumable electrode that needs replacing every-so-often.
Also, just curious:
Is it annoying that most Tesla-coilers practically worship you? (I also slightly do... I apologize). I can understand that fame is sometimes unwanted and a burden to bear. I hope you find my suggestion helpful, and, maybe it's a nice break from the typical comments of praise (and requests) you normally get.
Hi could you spare a moment to talk about our Lord and Savior Steve Ward?
I think you're right, graphite is probably the best answer. Pencil lead held up better than i expected - ive got some gouging rods on order. Any metals ive tried (including tungsten, brass, stainless steel, tool steel, aluminum, copper) eventually reach a surface temperature that causes some metal vaporization, and that effect is too bright to look at, hurts the eyes.
I don't mind some attention for doing cool stuff, hopefully my influence on anyone that is a fan of mine is a good one. I hope that if nothing else, I can inspire others to experiment and learn, and that can carry over to other parts of their lives that might actually be more useful than just making pretty sparks :-).
hey!!! they say "candle like" not "wildfire like" :D
Looks like something from Teslaundmehr xD
Steve Ward is the OG
You need a tungsten breakout point for that thing ;)
Just tried 1/4" tungsten rod. Even the square-cut end turns as bright as a lightbulb filament, painful to look at, no bueno. I'm waiting for some carbon gouging rods, I think any metal will eventually vaporize in a brilliant bright light.
@@SteveWard yeah sounds like you need some way to dissipate that heat from the electrode.. maybe a solid copper electrode with heatsink fins and then a forced air blower with the air coming up from the bottom though the primary and secondary to give it enough cooling so it never reaches glowing temps. This should also help cool your secondary if you want longer run times.
@@SteveWard the gouging rod won't melt but it will still emit bright blinding light I'd guess
nice