Need help on GU-50 plasma flame circuit.

[@T2D.SteveArcs]

Interesting theres no bypass cap close to the oscillator, these things are so sensitive, just changing the length of the breakout point can reduce the output drastically.. are you getting red plating? My thoughts are Its either out of tune, not enough grid drive/correct bias or maybe a bad tube, probably exactly what your already thinking 😀

[@CoolDudeClem]

I never run it long for the tube to start red plating as I don't want to destroy what little I have, so it's just a brief on for a few seconds, then off. So should the two coils need to be in tune with each other?

[@Timothycan]

Maybe check valve biasing against the valve data, check frequency, reposition coils or something. I don't know, as I've never tried to build one of these.

[@CoolDudeClem]

The annoying thing is though, in the past I built one of these that used a different circuit, it still used a GU 50 tube, but that circuit worked, I did get a plasma flame, only a small one, so maybe I'll re-visit that.

[@ritalinproductions]

@@CoolDudeClem wasn’t that outside in a garage?! I tried same circuit years ago and got no results. I did. blame it on the tube cause it had a miniature crack in it. Please keep us updated and good to see you building coils and stuff again!

[@Earthcomputer-d9t]

Brother I know how to make vacuum tube can you tell me brother what are the dimensions of gu50 vacuum tube internal parts size please brother

[@janno288]

Puh, my first guess would be wire connections, what does your scope say to it, hows the waveform? Also those small resistors can have quite the inductance. you can use a single resistor, it doesnt need to he a high wattage one 100k shouldnt get too warm. the capacior for feedback looks okay to me. Leon from Teslaundmehr made a video using the same schematic, you should take a look at that maybe. the shield put that to mains earth rather than ground, see what happens

[@CoolDudeClem]

I'm not sure if I could measure the waveform as it's such a high frequency, and the power is supplied on a half wave rectifier, there would be a lot of mains frequency noise in there too, it's gonna be really difficult go get a good look. I have seen Teslaundmehr's video on it and I'm quite envious, although he destroyed the tube by over-driving it at the end which I didn't think was necessary.

[@janno288]

@@CoolDudeClem True. tho a half wave rectified wouldnt cause anything funny, or did you mean halt wave doubled? half wave doubled could be a problem for these sorts of oscillators

[@janno288]

@@CoolDudeClem Any Update?

[@CoolDudeClem]

@@janno288 I'm currently comming up with a few ideas.

[@LarsToft-e4h]

Should both coils be "magnetically" connected, meaning wound on the same pipe? Greetings from Denmark.

[@unrealcrafter2671]

Not for this circuit. They can be separate

[@CoolDudeClem]

I don't think they need to be, I'm pretty sure they need to be seperate, that why I oriented one of them sideways and the other one vertical.

[@chilledoutpaul]

Hi Clem Its Paul here again 🙄. Anyway I'm not sure why they used a pentode, normally I would have used a triode unless they are using a grid as an accelerator, I dont see how the valve is being driven in to oscillation. I would try a low competence, high voltage capacitor where the 10k 50w resistor is, Also leave pin 5 no connection, as you have it to negative it would close the the valve down and could damage the valve. (First try removing pin 5's link to negative and leave it floating)

[@CoolDudeClem]

Pin 5 is plate 3, I thought that must always be connected to the cathode, even some tubes are internally wired that way.

[@chilledoutpaul]

@@CoolDudeClem Hi Clem, Yeah I have also seen that too, but knowing how valves work I imagined that the grid going to negative would cut the electrons getting through to the cathode, Anyway it was just an idea or a thought. When I was your age our lives are very similar, I use to spend all my life in my bedroom designing and fiddling with electronics, I first started electronics when I was about 10, no girlfriends, even when i left school I would work as an electrician, then get home to electronics lol.

[@Elnufo]

Did you check the frequency? Maybe you are way off?

[@CoolDudeClem]

I don't think I have any thing that could measure such high frequencies. my scope could be fast enough to measure that but the circuit as powered a half wave rectifier with no smoothing I'm not all that sure if I could get a reliable reading from it.

[@Clancydaenlightened]

Build a class E In mhz range inductance and wiring can make a difference Want to have short connects as possible And a pcb

[@Clancydaenlightened]

Reduce as much stray capacitance and inductance wherever you can

[@Clancydaenlightened]

Unfortunately I cannot post urls But uzzors2k and Steve ward class E on Google should get schematics

[@Clancydaenlightened]

Wonder how drsstc driver topology would work in such modes

[@Clancydaenlightened]

Can use a zvs driver too

[@Clancydaenlightened]

Build a zvs using tubes Or zero current switch since tubes are more current controlled than a MOSFET They switch either when apparent voltage across it is zero (anode-cathode, or drain-source) Or switch when the apparent current through the primary is zero (current sensing) Allows greater efficiency, faster switching at higher power with minimal waste dissipation

[@janno288]

After finally making my on GU-50 Plasma Flame generator (video on it on my channel) I think I know what the reason is, your feedback is way too tiny to make the tube oscillate real hard. Scrap the capacitor plate idea, have a loop of wire around your resonator coil end (a bit bellow where the flame is supposed to be) and that should make the tube oscillate well. Also have a mA meter hooked up, because it really shouldnt be drawing over 100mA current at like 200V b+ that would mean that you are barely oscillating and the tube isnt properally shutting off.