No oscillation with dummy load connected with loudspeaker cable.
OK, connect then cap as I described in my previous post.
The gm of the valve does not depend on which circuit arrangement it is in, but merely what the bias is: anode voltage (roughly half supply voltage), cathode voltage (or cathode current). I am not asking for gm of the circuit but gm of the valve. If you are unable to work it out just use the typical value given in the datasheet.
Alternatively, measure the impedance seen at the feedback point.
Alternatively, measure the impedance seen at the feedback point.
In my Williamson amp, I found by using a CCS in the common cathode of the 12BY7A config as triodes, this results in a gm roughly around 8; compared to your 3.5 of the
12BH7; would imply the 12BH7's should be pretty stable.
The 12BY7's are flighty tubes and using a SS current source, the stage oscillated on a X10 scope probe connected to an anode; the oscillation stopped when I chose a CCS pass transistor with a lower ft in the CCS.
This tells me something; the tube with a CCS in the cathode makes it behave like a grounded grid amplifier as the input impedance becomes lower than the AC CCS impedance of the cathode circuit. The AF balance becomes excellent, but remember the ccs AC imped magnification approaches near infinity and I found some damping is neccessary.
I'm only mentioning what I found with my circuit; the other problem is when the o/p stage hits the true AB quadrant with peak currents; Miller effect is flying back into the driver stage anodes whether one likes it or can desettle the capacitance of the stage if the phase margin is already slim and create oscillation on signal peaks....complicated and nasty debugging can be.
Too much CCS with too high ft, can be a bad thing, and with high b/w video tubes anything is possible.
My instinct is to temporarily substitute a resistor instead of the CCS and observe the reaction.
Morgan Jones Valveamps 3rd ed page 140.
richy
12BH7; would imply the 12BH7's should be pretty stable.
The 12BY7's are flighty tubes and using a SS current source, the stage oscillated on a X10 scope probe connected to an anode; the oscillation stopped when I chose a CCS pass transistor with a lower ft in the CCS.
This tells me something; the tube with a CCS in the cathode makes it behave like a grounded grid amplifier as the input impedance becomes lower than the AC CCS impedance of the cathode circuit. The AF balance becomes excellent, but remember the ccs AC imped magnification approaches near infinity and I found some damping is neccessary.
I'm only mentioning what I found with my circuit; the other problem is when the o/p stage hits the true AB quadrant with peak currents; Miller effect is flying back into the driver stage anodes whether one likes it or can desettle the capacitance of the stage if the phase margin is already slim and create oscillation on signal peaks....complicated and nasty debugging can be.
Too much CCS with too high ft, can be a bad thing, and with high b/w video tubes anything is possible.
My instinct is to temporarily substitute a resistor instead of the CCS and observe the reaction.
Morgan Jones Valveamps 3rd ed page 140.
richy
Try this:
Add resistors. If oscillations do not stop decrease R12 and increase C31 until it stops.
Play with R12 and C31.
One more option is to add a loop across output tubes, like 240K 2W resistors from anodes of output tubes to anodes of driver tubes.
Add resistors. If oscillations do not stop decrease R12 and increase C31 until it stops.
Yes, Effectively from anodes to grids, local feedback loop across output tubes, directly without secondary of the output transformer.You mean a resistor from each KT88 plate to the 12BH7 plates?
- I know am missing something here. Wouldn't the feedback network see R10 in series with R5?
- As far as measuring the phase goes, my scope has a math function so can you use that to measure phase? I am asking this cause someone showed me the math function to perform something, but I cannot remember if it was related to measuring phase difference or not.
- Why are you mentioning square waves? Do you measure phase difference using square waves or sine waves? I would think you would use a sine wave at various frequencies.
I think DF96 has answered the feedback network temp loading questions better than I can. If you can get within a few dB you're plenty close enough to find your way, and if the circuit's so twitchy that tiny differences matter it'll never stay still anyway and needs further work.
The square wave thing is only for old sound cards as sources and real emergencies - please ignore (and I'll try to pretend I never wrote it!).
All good fortune,
Chris
Try this:
Add resistors. If oscillations do not stop decrease R12 and increase C31 until it stops.
OK, a 3.3k resistor and 1000pF cap stopped the oscillation. 240k resistors were installed from anode to anode.
One problem. The amount of power the amp puts out is significantly reduced. 130 watts into the dummy load without resistors, and now less than 100 before it clips.
Should I change their value to restore original power output?
One problem. The amount of power the amp puts out is significantly reduced. 130 watts into the dummy load without resistors, and now less than 100 before it clips.
Should I change their value to restore original power output?
Still power drop problem after moving resistors to proper place. Are those resistors stealing power?
Try 330K, resistors, if still driver has no guts, increase up to 470K. Should be fine.
I have no 470k 2 watters on hand. I put two 330k's in series. My spectrum analyzer registers more distortion at 130 watts, and it requires another .25 volts input than without the resistors.
Would going up to 1MEG be OK? I would try it if I had some on hand. Just thinking about what I need to order.
Would going up to 1MEG be OK? I would try it if I had some on hand. Just thinking about what I need to order.
I have no 470k 2 watters on hand. I put two 330k's in series. My spectrum analyzer registers more distortion at 130 watts, and it requires another .25 volts input than without the resistors.
Would going up to 1MEG be OK? I would try it if I had some on hand. Just thinking about what I need to order.
It is up to you, whether you prefer more of global feedback through transformer, or of local feedback around output toobs. When you increase portion of local feedback gain through global one is lower, that helps to improve stability without sacrificing higher frequency end.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- Stubborn Oscillation