Had an odd situation today:
Customer sent in his Fisher 400 due to 'distortion from the left speaker'.
A typical problem with these amps is that the output tubes will get weak over time and sometimes one channel will degrade more than the other to the point where one channel is clipping at 6W while the other can do 20 or more unclipped.
So I fed the AUX inputs and measured the amp. 20 wpc with both driven without clipping. I call the customer and tell him that the output stage is working fine and that it may be the speaker or wiring to the speak after all.
He tells me he's using the phono input, so I go back and test the phono input. I discover that I can't go past 12 o'clock on the volume before the amp starts to oscillate at a high frequency at first, followed by a subsonic rate so the output looks like a gated oscillator.
It does this even with the output of the UPL turned off (no signal).
Odder still, with the volume at 11 o'clock, a RIAA sweep reveals a relatively flat right channel, but a large rise in response near 20Hz and a huge -14dB dip around 100-200hz, then gradually rising again to a large peak at 5000Hz. At 20Hz, the L and R channels are out of phase, but in phase at higher frequencies.
I measured the phono stage again at the Tape Out jacks and got the expected nearly flat RIAA (compensated) curve (+/- 1dB or less deviation.)
I note that if I disconnect the phono input from the UPL, the amplifier does not oscillate even at max volume setting. This got me thinking to try a separate generator, so I used an Tektronics oscillator to drive the preamp and that worked without any instability.
So now I'm wondering what about the Fisher and the UPL that causes some sort of positive feedback loop? Is it because the Fisher output isn't ground referenced? But I've measured bridged amplifiers without this sort of instability. And it's only through the phono inputs that it oscillates.
I've come to the conclusion that nothing is wrong with the Fisher. It is a quirk in the UPL that makes the Fisher unstable. Any thoughts on this from other UPL owners?
Customer sent in his Fisher 400 due to 'distortion from the left speaker'.
A typical problem with these amps is that the output tubes will get weak over time and sometimes one channel will degrade more than the other to the point where one channel is clipping at 6W while the other can do 20 or more unclipped.
So I fed the AUX inputs and measured the amp. 20 wpc with both driven without clipping. I call the customer and tell him that the output stage is working fine and that it may be the speaker or wiring to the speak after all.
He tells me he's using the phono input, so I go back and test the phono input. I discover that I can't go past 12 o'clock on the volume before the amp starts to oscillate at a high frequency at first, followed by a subsonic rate so the output looks like a gated oscillator.
It does this even with the output of the UPL turned off (no signal).
Odder still, with the volume at 11 o'clock, a RIAA sweep reveals a relatively flat right channel, but a large rise in response near 20Hz and a huge -14dB dip around 100-200hz, then gradually rising again to a large peak at 5000Hz. At 20Hz, the L and R channels are out of phase, but in phase at higher frequencies.
I measured the phono stage again at the Tape Out jacks and got the expected nearly flat RIAA (compensated) curve (+/- 1dB or less deviation.)
I note that if I disconnect the phono input from the UPL, the amplifier does not oscillate even at max volume setting. This got me thinking to try a separate generator, so I used an Tektronics oscillator to drive the preamp and that worked without any instability.
So now I'm wondering what about the Fisher and the UPL that causes some sort of positive feedback loop? Is it because the Fisher output isn't ground referenced? But I've measured bridged amplifiers without this sort of instability. And it's only through the phono inputs that it oscillates.
I've come to the conclusion that nothing is wrong with the Fisher. It is a quirk in the UPL that makes the Fisher unstable. Any thoughts on this from other UPL owners?
Check that the low side of the speaker connections are ground or not. Some amps of that period did stuff on the low side of the amp output.
There are several potential feedback gotchas on those older receivers where the grounds are shared but not everywhere. The RCA input's low side may also be a culprit. If the tonearm shield were connected to ground in more than one place you could have issues. Lots of gain from phono to speaker in those guys, especially internally on the plates of the out;put tubes.
There are several potential feedback gotchas on those older receivers where the grounds are shared but not everywhere. The RCA input's low side may also be a culprit. If the tonearm shield were connected to ground in more than one place you could have issues. Lots of gain from phono to speaker in those guys, especially internally on the plates of the out;put tubes.
At first, I thought that it was like some Fisher amps where the 4Ω tap is ground and the 0 tap is the low side of speaker out, but this is not the case here. This amplifier has speaker out referenced to chassis. The phono section is insulated and probably tied to the power supply ground somewhere.
I tried setting the generator to ground referenced and then lift, but neither had any effect.
I noticed that the threshold of oscillation was lower if the generator was set to 600Ω and much higher if set to 5Ω.
Pretty peculiar issue. I thought maybe the ground of the generator might be connected to the ground of the analyzer, forming a secondary path.
One thing I checked just now is to see if grounds between left and right are separate. Apparently my dummy load has the ground side of each resistor tied together. Maybe that is an issue, but given that speaker grounds are chassis common on the Fisher, I can't see how.
I tried setting the generator to ground referenced and then lift, but neither had any effect.
I noticed that the threshold of oscillation was lower if the generator was set to 600Ω and much higher if set to 5Ω.
Pretty peculiar issue. I thought maybe the ground of the generator might be connected to the ground of the analyzer, forming a secondary path.
One thing I checked just now is to see if grounds between left and right are separate. Apparently my dummy load has the ground side of each resistor tied together. Maybe that is an issue, but given that speaker grounds are chassis common on the Fisher, I can't see how.
Any issues with the 10 ohm resistors on the input selector switch? You probably know already that the Fisher 400 phono inputs aren't directly connected to ground.
Using the strategy of fixing it until it's broken:
1) Cartridge loading seems just a tad under-damped. The paralleled value of R3||(R5+R7)|| R9 is ~52k, proper value would be 47k. If you change R3 (and R4) to 150k the damping should be adequate.
2) The RIAA network values on PC1 and PC2 are just a tad off. One could fashion a new SMT circuit card with the 270k value resistor changed to 243k.
WRT the UPL, I will sometimes encounter issues using the AP SYS2722 in floating mode.
Using the strategy of fixing it until it's broken:
1) Cartridge loading seems just a tad under-damped. The paralleled value of R3||(R5+R7)|| R9 is ~52k, proper value would be 47k. If you change R3 (and R4) to 150k the damping should be adequate.
2) The RIAA network values on PC1 and PC2 are just a tad off. One could fashion a new SMT circuit card with the 270k value resistor changed to 243k.
WRT the UPL, I will sometimes encounter issues using the AP SYS2722 in floating mode.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.