The protection circuit appears to be working as it should. (The schematic at post #19 shows high voltage from the bridge rectifier coupled to a potentiometer and voltage divider- can the protection voltage be adjusted?)
Assuming the protection circuit is working well, perhaps each triode of the 6336/6080 is drawing very different current? Manufacturers's data sheets recomend that, to help the each triode share current and to prevent one triode drawing excessive current, minimum values of cathode resistance should be used. In the MQ80, it appears that at DC the triodes are in parallel - only the small resistance of the cathode winding of the OPT is attached to each cathode.
Assuming the protection circuit is working well, perhaps each triode of the 6336/6080 is drawing very different current? Manufacturers's data sheets recomend that, to help the each triode share current and to prevent one triode drawing excessive current, minimum values of cathode resistance should be used. In the MQ80, it appears that at DC the triodes are in parallel - only the small resistance of the cathode winding of the OPT is attached to each cathode.
bondini ...high voltage from the bridge rectifier coupled to a potentiometer and voltage divider- can the protection voltage be adjusted?
in 60 s curent of each triode gos t 120 ma in 6080 !!! i use 800 ohm opt. chang socket....i dont I don't understand why the current is go high so fast?Burned fuse.........why? voltage is -120 in 4 and 1 pin . 😱 protect circuit conect to +b by capacitor ...
in 60 s curent of each triode gos t 120 ma in 6080 !!! i use 800 ohm opt. chang socket....i dont I don't understand why the current is go high so fast?Burned fuse.........why? voltage is -120 in 4 and 1 pin . 😱 protect circuit conect to +b by capacitor ...
OK, so the protection circuit works 👍
And the valves are OK outside the amp? Could you test each 6336A outside the amp to make sure that both triodes bias up the same? 😒
For the 6336A: Bias of -84V with Vak=250 gives 50mA for each triode, 100mA total. BUT if the bias drops to -75V then each triode will draw 100mA! The 6336A is a big powerful valve and anode current is VERY sensitive to cathode bias.
For the 6080: With Vak of 250V the bias would need to be -140V to hold anode current at 50mA.
If the valves are OK, then you could test the components in the bias circuit? Is it working well?
As I recall, Luxman were one of the few manufacturers to use the 6336A in audio and they put a lot of effort into getting it right. They had the only commercial designs I know of that used 6336A AND fixed bias. Kees Huevelman and Wim de Haan (Glass Audio, May 2005) found the 6336A is hard to use in fixed bias - I agree!!
And the valves are OK outside the amp? Could you test each 6336A outside the amp to make sure that both triodes bias up the same? 😒
For the 6336A: Bias of -84V with Vak=250 gives 50mA for each triode, 100mA total. BUT if the bias drops to -75V then each triode will draw 100mA! The 6336A is a big powerful valve and anode current is VERY sensitive to cathode bias.
For the 6080: With Vak of 250V the bias would need to be -140V to hold anode current at 50mA.
If the valves are OK, then you could test the components in the bias circuit? Is it working well?
As I recall, Luxman were one of the few manufacturers to use the 6336A in audio and they put a lot of effort into getting it right. They had the only commercial designs I know of that used 6336A AND fixed bias. Kees Huevelman and Wim de Haan (Glass Audio, May 2005) found the 6336A is hard to use in fixed bias - I agree!!
thanks for help...Now, the bias voltage for 6080 is 125 volts, not 80. But it stays at 50 milliamps for 10 seconds, and then it walks and rises. I also saw the 120 milliamps, so I turned it off quickly. The triodes are the same, the tubes are new and good. I bought about 50 to 6080 before. It is true that the bias fix is not good for these lamps. But it is clear that this device has been working for years. What should we do now?
bondini:
For the 6336A: Bias of -84V with Vak=250 gives 50mA for each triode, 100mA total. BUT if the bias drops to -75V then each triode will draw 100mA! The 6336A is a big powerful valve and anode current is VERY sensitive to cathode bias.😱 😱 😱 😱 😱
None of the voltages changes...only the cathode voltage of the tube....the bias is completely fixed.
It's very difficult to measure actual bias voltage (at the grid), because of the high impedances involved. Voltage at the other end of the grid leak resistor from supply may be quite different, varying with grid current. We mostly assume that grid current is negligible, but that's not always true, and grid current from space charge leakage is in the direction that reduces bias voltage, tending towards a runaway condition. Without an actual servo control, only keeping grid leak resistance to low values can contain a (even slightly) gassy valve.
All good fortune,
Chris
All good fortune,
Chris
Are we still referring to the schematics in the very first posting? If so, what happens if you pull the 12BH7 cathode follower tubes? If the 6080/6336 cathode currents remain stable then, what happens if you reinsert the 12BH7 and desolder the 22 nF couplling capacitors to it's grids?
Best regards!
Best regards!
1: check with scope for oszillation.
2: check the OPT, disconnect the OPT anode windings from PS and check isolation at least the resistance between anode and all grounded windings.
2: check the OPT, disconnect the OPT anode windings from PS and check isolation at least the resistance between anode and all grounded windings.
Just a thought: what if you disconnect the anodes from the output transformers and connect the B+ directly (or via a resistor with the same value as the OPT DC resistance) to the anodes of the 6336?
See if the current stays stable in that case.
See if the current stays stable in that case.
It seems the problems come from the sockets. Possibly the contact of the tube pins with the sockets is bad either because they are dirty inside or they are worn out.
You can check this by turning a thin wire around the tube g1 pins (1 and 4), then put the tube in its socket and measure the resistance between these wires and the socket pins under the chassis.
You can check this by turning a thin wire around the tube g1 pins (1 and 4), then put the tube in its socket and measure the resistance between these wires and the socket pins under the chassis.
My summary of the most likely culprits.
Temp sensitive coupling cap.
Bad grid bias resistor.
Bad connections involving grids.
Unstable bias supply.
Gassy tube.
Just because the tubes are new one cannot rule out the gassy tube likelihood. In fact, a new manufacture tube is likely to have poorer quality control than the well established vintage manufacturers. Could be an entire lot of tubes are gassy. Symptoms are consistent with trapped gass in the tube that is liberated with heat. Measure the grid current. A larger grid stopper resistor should exasperate the issue and make it easier to measure the current in the grid.
Temp sensitive coupling cap.
Bad grid bias resistor.
Bad connections involving grids.
Unstable bias supply.
Gassy tube.
Just because the tubes are new one cannot rule out the gassy tube likelihood. In fact, a new manufacture tube is likely to have poorer quality control than the well established vintage manufacturers. Could be an entire lot of tubes are gassy. Symptoms are consistent with trapped gass in the tube that is liberated with heat. Measure the grid current. A larger grid stopper resistor should exasperate the issue and make it easier to measure the current in the grid.
Cathode resistors can make a better controllable situation as experimental arrangement. 470 ohm (a lot of watts) for each cathode. It can prevent the thermal runaway or gives more chance for to measure an abnormal function.
After #27 , of course.
After #27 , of course.
Good looking amp, world class iron... approaching 50 years of age.
If it were me, I'd do a total rebuild: from the pics on the net, some resistors appear to be carbon composition that may have drifted, the bias pots may not be up to par (multi-turn recommended), the capacitors could use an upgrade, the connectors, and, of course, the electrolytics.
If it were me, I'd do a total rebuild: from the pics on the net, some resistors appear to be carbon composition that may have drifted, the bias pots may not be up to par (multi-turn recommended), the capacitors could use an upgrade, the connectors, and, of course, the electrolytics.