• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Fault finding - loud pop and 120hz noise instead of music

Jump to Latest
Klimon

Klimon

Member
Joined 2005
#1
I've enjoyed this amp for quite some time now, but it developed a problem I'm trying to track down: https://www.diyaudio.com/community/threads/6v6-pp-ul-bias-point.360543/post-6359252

Symptoms? Loud pop and low level 120hz hum (no more music) in one channel. Happened the first time when I was playing white noise for speaker testing. Then another trouble free month. It now does it half of the time, the other half playing flawlessly. Also when powering up I experienced a few times the pop + a complex noise getting very loud while the tubes are heating, leading me to power off before killing my speaker.

I measured most of the components (power off, seems fine), swapped tubes (no cigar) and measured with power on (probably in bad-mode at that time: voltages were way off in one channel).

Before I do a systematic check of voltages in the bad channel, out of curiosity: is this oscillation? A bad solder contact? What baffles me is that it either works, either not, but the latter is steadily taking over.

Cheers
Simon
 
#2
Input stage/phase inverter tube is bad. Significant difference in g1 leakage current between the two halves. That can cause all the bias to shift to one of the pair and latch up, leaving the stage effectively cut off. It may be ok at normal quiescent, but have trouble once it reaches significant voltage swing - hence the problem occurring with a large test signal. I had one do this due to one of the triodes having too much h-k leakage. Since your cathodes are near ground, I’d suspect g1 leakage current. Result is the same - latch-up. Make the g1 resistances equal to minimize the possibility of it, but if one triode is leaky it really ought to go.
 
#4
wg_ski said:
Input stage/phase inverter tube is bad. Significant difference in g1 leakage current between the two halves. That can cause all the bias to shift to one of the pair and latch up, leaving the stage effectively cut off. It may be ok at normal quiescent, but have trouble once it reaches significant voltage swing - hence the problem occurring with a large test signal. I had one do this due to one of the triodes having too much h-k leakage. Since your cathodes are near ground, I’d suspect g1 leakage current. Result is the same - latch-up. Make the g1 resistances equal to minimize the possibility of it, but if one triode is leaky it really ought to go.
Click to expand...
I just popped in another pair, playing fine so far. Touch wood: such a precise diagnosis & easy fix are a godsend.

I'll install a 6.8k grid stopper (R25) on the other triode halve as well once I stop listening.

Many thanks
Simon
 
#5
Klimon said:
I just popped in another pair, playing fine so far. Touch wood: such a precise diagnosis & easy fix are a godsend.

I'll install a 6.8k grid stopper (R25) on the other triode halve as well once I stop listening.

Many thanks
Simon
Click to expand...
No luck yet, after an extended listening yesterday evening, the problem reoccurred this morning. I'm ruling out the tube hypothesis. Meanwhile I swapped tube position several times and it's always the same channel that fails.

Will start measuring under power with special focus on the phase splitter circuit.
 
#9
wg_ski said:
It still sounds like a latch-up. Check the voltages across the plate loads of the phase splitter when it happens to be sure we’re looking in the right place. Other things besides a bad tube can cause it.
Click to expand...
Your analysis is correct: I measured around the input stage and it presented three 'modes' the last hour:

  • everything ok = 155v across each 150k plate load & 0.9mV across R26 (100R between both cathodes)
  • pop followed by light hum, no music = 168v (R24) and 147v (R23)
  • ? (no speakers connected), probably the loud hash I experienced a couple of times = R24 (300mV), R23 (180mV), R26 (2V), R28=lower cathode to ground (15V)

FWIW I'm running the amplifier in triode, without feedback.
  • R28 is connected to ground (no C3), but on the other side of the pcb so no measurements yet.
  • the C4 and R2 loop is absent.
But that shouldn't matter since the other channel is identical and works fine.

I'll get some isopropyl tomorrow to clean the pcb and rule that out as well.

Thanks a lot for your assistance
Simon
 
You must log in or register to reply here.
Top Bottom