833c Vttc not working

[Thereuopoda]

Hello, I have been making several changes to my 833c vttc, but no luck. I have tried reversing the feedback and primary coils, as well as reducing the tank capacitance. Unfortunately, my secondary coil is not developing any voltage at all. I also have tried using a Grid leak circuit, consisting of a 250ohm resistor and 2 1200pf rf caps. Any ideas?

[MRMILSTAR]

I think that your primary circuit capacitance is too small. You won't have enough turns on the primary coil to tune it with your secondary circuit. For the number of turns on your primary coil, you probably want around 2 nF in order to tune it with your primary coil inductance and secondary coil size. For a sanity check, use JAVATC to confirm that you can achieve resonance on the primary and secondary circuits with your design parameters. Remember to add a few extra turns for tuning due to inaccuracies. Running out of tuning turns is very inconvenient. You can ignore the spark gap portions of JAVATC.

[Thereuopoda]

Thank you for your reply, I have been messing around with the capacitance, unfortunately when I added caps to get 2.8nf, nothing. I finally discovered that my tube appeared to have an internal short. I hooked up my oscilloscope to the grid and discovered a 60hz AC wave, as well as one on the anode. I believe this means that the 60hz 10v AC is somehow on the grid and anode. Furthermore, I measured 1.2kv DC on the cathode, this leads me to believe that my tube is bad. Hooking up a continuity tester told me that the Grid and Anode are both shorted to the Cathode, but It only happens once the filament is hot. This is the second tube I have ordered, the first one had lost its vacuum due to a glass fracture sometime before I got it probably during shipping. How can I get an 833 A or C tube that works and also not spend too much money? Brand-new in-the-box tubes are pretty expensive, unfortunately. Thank you for your reply!

[MRMILSTAR]

Bad tubes are always a possibility, especially if bought used. My first 833A was weak even though the filament glowed. I finally bought a NOS Ampex 833A and later a matched pair of NOS 833C tubes. Its always best to have two tubes on hand so that you can swap them out for testing. NOS 833A or 833C tubes aren't cheap though.

[janno288]

Can you measure the voltage on the grid when b+ is on and filament voltage has been applied?
Preferebly with a scope and please post the waveform.

4 Pictures, 2 gridleak, both polarities , 2 no girdleak both polarities.

Wait if your schematic is accurate, you dont have the filament (cathode) attached to your power supply ground, the tube obviously doesnt conduct if there is possibility for it to conduct

[Thereuopoda]

Thank you janno288 and MRMILSTAR,
Turns out the tube is fine. and after playing with the feedback we had some nice sparks. We also had to ground the filament, and also remove the grid leak resistor. We also ended up driving a ground pole into our garden. We still have one problem though, there is a nasty corona coming off the feedback coil, probably due to high coupling between the secondary and feedback. We will be away for a few days and won't be working on the coil, and we plan to make a spacer for a new feedback coil. For anybody wondering, the Feedback is at 25 turns of 24 AWG magnet wire, the primary is 27 turns of 8 AWG wire. We now have 6 470pf soviet RF caps for the tank circuit, and the secondary is ~1100 t of 27 AWG red magnet wire.
Here are the arcs it produces:

[MRMILSTAR]

Glad to hear that you got it running. I had to add an acrylic spark shield to the top of my grid feedback coil because I was getting arcing between it and the top of the secondary coil.

[Thereuopoda]

I have been working on tuning the coil, as well as making a new feedback coil that doesn't arc. I used 8 2'' long 3/4 PVC pipe sections to create space between the feedback and the secondary. I have gotten down from 25 to 12 turns on the feedback coil, this works very well. The system pulls around 25A and 120v at full power, as well as producing 16'' streamers without any staccato control. I have been pushing this 833c tube to the limit, with it red plating significantly at full power. The only problem is that the dual MOT setup gets hot after a while. My next idea is to make it able to play music, but I don't know where to start with that, any ideas?
Here is my feedback coil:

And Here are the streamers:

[Thereuopoda]

Hello,
I have been messing around with staccato control with the coil and when I use staccato in the system the microwave transformers violently arc over. Doesn't happen in uninterrupted mode so I am not sure why I have this problem, hope somebody might know why.

[davekni]

I have been messing around with staccato control with the coil and when I use staccato in the system the microwave transformers violently arc over. Doesn't happen in uninterrupted mode so I am not sure why I have this problem, hope somebody might know why.

What type of switch does your interrupter use?  Thyristor, IGBT, FET?
MOT cores run fairly hard into saturation in normal use.  When powered for a half-line-cycle, input volt-seconds is twice, causing extreme saturation and resulting very-high primary current.  Current is mostly limited by winding and line power resistance, typically 100A or more.  That occurs occasionally when a microwave oven switches on by chance near line zero volt point.  That's why there is sometimes a brief loud hum when switched on.
If your staccato controller switches off at end of half-cycle when current is extreme, resulting inductive voltage spike is high.  That would be a likely cause for arcs and/or damage to staccato controller.  However, if controller uses thyristors (TRIAC or other), current will continue through much of next line half-cycle, ramping current down in a more controlled way.  In that case, I can't think of any reason for arcing.  Might be some interaction of MOT's internal magnet shunt saturating earlier and/or coming out of saturation later (if you haven't removed MOT shunt).

[Thereuopoda]

It is thyristor based, I have not removed the shunts because I was told to leave them in there as a kind of ballast. If I were to remove them, are there any pointers to not damage the coils?
Thx for your reply
 

[davekni]

It is thyristor based, I have not removed the shunts because I was told to leave them in there as a kind of ballast. If I were to remove them, are there any pointers to not damage the coils?

Ballast is the shunt's purpose in microwave ovens.  Thinking in more detail about shunt saturation, I can't think of any reason that would be causing voltage spikes.  Only remaining possible cause that comes to mind is stored minority carrier charge in thyristor.  I'm not an expert on thyristors.  I believe stored charge causes some delay in turn-off, slightly past zero current point.  Perhaps that delay is followed by a relatively rapid drop in current, causing an inductive spike.  A scope capture of MOT primary voltage would capture useful information, especially when voltage spike occurs relative to line half-cycle.  If you don't have any probes capable of surviving a voltage spike, even a probe capacitively-coupled to MOT, primary or secondary, might be sufficient.  In other words, a probe placed in proximity to signal but not connected to it.

Whatever the exact issue is, I think the best solution is to use your interrupter instead.  I presume it is interrupting secondary side, either grid or cathode of your tube.  Set interrupter duration for half-line-cycle (8.33 or 10ms) and trigger on line zero-crossings.  Thyristor gate signal from your existing staccato controller might be suitable trigger, if it doesn't start way early on non-conducting half-cycle of thyristor.  Or load your staccato controller with an incandescent light bulb and place a small pulse transformer in series with bulb.  Presuming pulse transformer is small and saturates easily, secondary will have pulses at line zero crossings.