Loud hum, and high AC at idle on one output tube.

[bendyha]

A problem I can't figure out with a pair of EL84 in a fixed bias P-P amp. Class AB.
It is a guitar amp, but the problem is in the output stage, and could apply to any tube amp.


The initial state of the amp was:
One EL84 tube blown, along with its burnt-out screen grid resistor- resistor and both tubes replaced.
One Control grid to PI coupling caps was dodgy - now both are new and check OK.


Quick post repair test: Loud 60Hz hum.


More tests where everything seems OK
OT winding DC resistances all seems correct.
PT winding DC resistances all seems correct. All voltages tally with the schematic.
Filter caps all check OK.


Resistance measurements on both tube sockets without the Power tubes inserted:
Both tube sockets measure identically in all tests:
Continuity and resistance to ground from control grid, cathode, and the filaments all correct in respect to the ratings of the resistors in their circuit paths.
Plate to control grid - gives a sweeping reading from 220K to around 1M
Filament to plate – around 4.5M


Voltage measurements without the power tubes inserted:
Both tube sockets measure identically in all tests:
Filament voltage normal
Bias voltages correct.
Plate and screen voltages are as expected – correct, and (practically) free from AC.




Now here is where it gets strange:


Voltage measurements with one power tubes inserted in the previously unaffected socket
V5- Tubes Plate: 336VDC & 75AC
V4- Empty sockets plate supply: 336VDC & 350VAC


Plug both tubes in:
V5- Tubes Plate: 336VDC & 3.3VAC
V4- Tubes Plate: 336VDC & 538VAC … and the tube is noticeably hotter.


When applying the multimeter probes to the plates for measuring, the loud humming that is present gets noticeably louder when measuring V4, but disappears when measuring V5.




Any suggestions or thoughts would be welcome.

[baudouin0]

Do you have a schematic?

[petertub]

Originally Posted by bendyha

A problem I can't figure out with a pair of EL84 in a fixed bias P-P amp. Class AB.
It is a guitar amp, but the problem is in the output stage, and could apply to any tube amp.


The initial state of the amp was:
One EL84 tube blown, along with its burnt-out screen grid resistor- resistor and both tubes replaced.
One Control grid to PI coupling caps was dodgy - now both are new and check OK.


Quick post repair test: Loud 60Hz hum.


More tests where everything seems OK
OT winding DC resistances all seems correct.
PT winding DC resistances all seems correct. All voltages tally with the schematic.
Filter caps all check OK.


Resistance measurements on both tube sockets without the Power tubes inserted:
Both tube sockets measure identically in all tests:
Continuity and resistance to ground from control grid, cathode, and the filaments all correct in respect to the ratings of the resistors in their circuit paths.
Plate to control grid - gives a sweeping reading from 220K to around 1M
Filament to plate – around 4.5M


Voltage measurements without the power tubes inserted:
Both tube sockets measure identically in all tests:
Filament voltage normal
Bias voltages correct.
Plate and screen voltages are as expected – correct, and (practically) free from AC.




Now here is where it gets strange:


Voltage measurements with one power tubes inserted in the previously unaffected socket
V5- Tubes Plate: 336VDC & 75AC
V4- Empty sockets plate supply: 336VDC & 350VAC


Plug both tubes in:
V5- Tubes Plate: 336VDC & 3.3VAC
V4- Tubes Plate: 336VDC & 538VAC … and the tube is noticeably hotter.


When applying the multimeter probes to the plates for measuring, the loud humming that is present gets noticeably louder when measuring V4, but disappears when measuring V5.




Any suggestions or thoughts would be welcome.

Is the tube oscillating ? Install grid and screen stoppers!

Get a matched pair !

Randomly selected tubes may differ wild

[baudouin0]

I tend to agree given just probing the plate of V5 seems to stop the oscillation.

[bendyha]

Originally Posted by petertub

Is the tube oscillating ? Install grid and screen stoppers!

Get a matched pair !

Randomly selected tubes may differ wild

Your thoughts are the same as my first ideas....
Control grid resistors of 1K5 are present and correct.
Screen grid resistors of 100 Ohm are installed, but this is maybe a bit small to block all oscillation senarios....I will try adding 500 Ohm onto the tube socket.
Tubes are new, and are a standard matched pair.

[petertub]

Originally Posted by bendyha

Your thoughts are the same as my first ideas....
Control grid resistors of 1K5 are present and correct.
Screen grid resistors of 100 Ohm are installed, but this is maybe a bit small to block all oscillation senarios....I will try adding 500 Ohm onto the tube socket.
Tubes are new, and are a standard matched pair.

Not all tubes sold as "matched" are anywhere close to matched.

Try another pair from another vendor, someone like mcshane that does real matching.

[bendyha]

Originally Posted by petertub

Not all tubes sold as "matched" are anywhere close to matched.

Try another pair from another vendor, someone like mcshane that does real matching.

As may be...but the described problem would not be induced by a pair of tubes not quite up to cork sniffer standards....vin ordinaire is palatable enough to do the job.


Some of the best sounding guitar amps I have built, sound at their best with totally mismatched output tubes.....but the above described problem has never resulted from that scenario.

[wiseoldtech]

What are the cathode DC volts when at idle?

[baudouin0]

Do you have a schematic even if its hand drawn - its difficult to help without one.


When you say V4- Tubes Plate: 336VDC & 538VAC what did you measure the VAC with - its huge. If its a multimeter they tend to roll of at high frequencies. Do you think its 120Hz or much higher. The EL84 should not oscillate with the resistor values you have.

[bendyha]

Originally Posted by wiseoldtech

What are the cathode DC volts when at idle?

Cathodes at ground potential.

Originally Posted by baudouin0

Do you have a schematic even if its hand drawn - its difficult to help without one.

Originally Posted by baudouin0

When you say V4- Tubes Plate: 336VDC & 538VAC what did you measure the VAC with - its huge. If its a multimeter they tend to roll of at high frequencies. Do you think its 120Hz or much higher. The EL84 should not oscillate with the resistor values you have.

A Fluke multimeter. Works fine...but just to be sure, other multimeters were also used....same results.

Yes, the AC voltage is huge.
The hum sounds like mains AC 60Hz...or maybe 120Hz.
No..it shouldn't be doing any of the wierd things that it is.....I admit to being stummed on this one...what am I missing?