GU81M VTTC help needed please: resonance anomaly

astralhighway, Thu Sept 28 2017, 04:06PM

Hi again,

This is my third build but is bugging me more than the others ever did. Help and insight on my primary and feedback/ tickler coils please.

Here are the parameters: Single GU81M, doubled MOT power supply with chain of 3A HV diodes.

Secondary: 8in long, 3.5in diameter.
Around 150 turns, 1.5mm wire. 4.25in copper corona guard, 5in dia stainless steel toroid. 3in breakout poin. DC resistance, 0.7R

self-resonance of secondary measured with scope probe in the air, including an 8 in copper wire 'streamer' attached to breakout point to simulate a plasma arc; 1.96 MHz.

Original primary: 16turns 1mm wire on 6 in dia former This is now discarded, but I've included it to show how I got the Fres of the secondary
Primary capacitor: 150pF vacuum type
Feedback/ tickler coil: 4 turns, 1 inch below primary

Grid leak 2.5K 10nF

I measured resonance by sweeping the tank circuit (primary plus 150pF parallel cap) with my signal generator - as I say, scope probe in the air, so unloaded.

Curiously, the tank measured just around 1Mhz when swept with the signal generator physically coupled to the tank via a 100K resistor. The secondary also measured only around 1MHz when tested in this way, a whole 900Khz lower F res than when measured coupled to the primary with the scope probe in the air.

I had an immediate flashover between the bottom of the primary and the top of the tickler coil. The scope picked up a huge pulse but I couldn't get a reading before the tickler flamed. ill

[color]I am now redesigning the primary coil and tickler.

My calculations give the following for the new primary: 13turns x 2mm wire for inductance of 43uH Please comment, thank you.
This is close to the original parameters but with three less turns and with thicker wire.

Please can you comment on the ideal position of the primary? I've seen some that seem to be almost mid-way up the secondary, especially in a case like mine where the secondary is rather short.

I also calculate only 4nF rather than my original 10nF for the grid leak capacitor, which gives around 5.25 time constants with 2K5 grid leak resistor. Is this about right?

Thank you folks