I'm not setting the bias, that is already done. It's just the same test points for reference.
I'm working on a Pioneer stereo / amp that had blown outputs. I've replaced them and have it running, but suspect it may be running too hot. One thing that I discovered was the voltage across the Left channels emitter resistor, which is also the test point for setting the bias, is about 20% higher than the Right side. Meaning: If I send it a 1v p-p sinewave and turn the volume up "half way" I will get 450mV across the right channel test point and 650mV across the left. What does that mean?
Also This emitter voltage discrepancy is with the same RMS value on dummy loads, meaning the output appears to be the same on both channels, but one has this higher voltage.
I'm not setting the bias, that is already done. It's just the same test points for reference.
I'm working on a Pioneer stereo / amp that had blown outputs. I've replaced them and have it running, but suspect it may be running too hot. One thing that I discovered was the voltage across the Left channels emitter resistor, which is also the test point for setting the bias, is about 20% higher than the Right side. Meaning: If I send it a 1v p-p sinewave and turn the volume up "half way" I will get 450mV across the right channel test point and 650mV across the left. What does that mean?
Also This emitter voltage discrepancy is with the same RMS value on dummy loads, meaning the output appears to be the same on both channels, but one has this higher voltage.
I'm not setting the bias, that is already done. It's just the same test points for reference.
Last edited:
Bias currents should be made without a sine wave applied.
The bias current is usually set with a pot.
If no pot then a resistor will set the bias point.
You need to study the circuit diagram.
The bias current is usually set with a pot.
If no pot then a resistor will set the bias point.
You need to study the circuit diagram.
So, it’s possible that the left and right emitter resistors have different values. It’s also possible that the left and right speaker loads have different values. The bias is set by setting the voltage across the emitter resistors. The ac value is determined by the ratio of the load resistor to the emitter resistor and the voltage across the load.
Are you using dummy loads?
Are you using dummy loads?
Yes, Dummy loads.
Let me check the emitter resistors value and the dummy load vales and I will report back. Thank you!
Let me check the emitter resistors value and the dummy load vales and I will report back. Thank you!
Reads like you have incorrectly set the bias, when correctly set the voltage drop across the L&R channel emitters is the same.
I think the point is the voltage changes differently as signal level increases, not that the quiescent bias is wrong. That might indicate problems with an output device or oscillation setting in. Really needs a 'scope to diagnose properly.
I'm using a scope. The sinewaves look fine to me and pretty darn close to each other. But I'm just looking at the signal across the dummy load. Something else?
Since it does not affect sound , temperature, etc., and you only noticed a small difference when scoping, looks like a non problem to me.
jm2c
jm2c
I don't see the testpoint on the schematic.
Perhaps what you are seeing is a DC output offset?
The two channels should be the same, sorry to state the obvious!
Maybe there is a problem with C17, C18 DC block on the feedback?
Maybe an emitter resistor has changed value?
Maybe a tranistor i nthe output darlingtons is a long way from matched?
Given that it blew up before, it would be good to know that everything is right.
Maybe measure the drops across all the emitter resistors and check those caps?
Perhaps what you are seeing is a DC output offset?
The two channels should be the same, sorry to state the obvious!
Maybe there is a problem with C17, C18 DC block on the feedback?
Maybe an emitter resistor has changed value?
Maybe a tranistor i nthe output darlingtons is a long way from matched?
Given that it blew up before, it would be good to know that everything is right.
Maybe measure the drops across all the emitter resistors and check those caps?
I tend to agree with the @JMFahey assessment in post 14, but if curiosity compels you, I suggest looking at the bias spreader, i.e. voltage between base of Q8 and base of Q6.
Because these are susceptible points in the amp, I suggest 10k resistors in series with each voltmeter lead, with the bodies of the resistors tacked in near the spreader to help provide some immunity to the measurement intrusion. Perhaps prepare similar resistors on the better channel for comparison--- and monitor both channels simultaneously if you have the luxury of two meters.
I'd take initial readings, and after a period of applied drive, remove drive and observe how the spreader voltage changes as the heatsinks cool. Compare with the reference channel.
Is the thermal coupling of the diodes to heatsinks similar in both channels? Seems interesting even if not worrisome.
Because these are susceptible points in the amp, I suggest 10k resistors in series with each voltmeter lead, with the bodies of the resistors tacked in near the spreader to help provide some immunity to the measurement intrusion. Perhaps prepare similar resistors on the better channel for comparison--- and monitor both channels simultaneously if you have the luxury of two meters.
I'd take initial readings, and after a period of applied drive, remove drive and observe how the spreader voltage changes as the heatsinks cool. Compare with the reference channel.
Is the thermal coupling of the diodes to heatsinks similar in both channels? Seems interesting even if not worrisome.
Still haven’t heard whether the emitter R’s and dummy load R’s are matched. 20% ac voltage mismatch could easily be explained by that matching. Till we have a response on that, I will avoid further speculation.
Low R emitter resistor values are often hard for multimeters to measure accurately.
If the amp is stable without loads, the ac signals on the emitter resistors should be small.
I agree, it’s quite likely that the ac emitter resistor voltage observation may be a non problem.
Low R emitter resistor values are often hard for multimeters to measure accurately.
If the amp is stable without loads, the ac signals on the emitter resistors should be small.
I agree, it’s quite likely that the ac emitter resistor voltage observation may be a non problem.
I am chasing an overheating issue. At least I think it is. The transistors march past 60c pretty quickly at only about 4-5 RMS across each output. It clips at 14V, which is pretty much what it's rated for. The SX-550 has a VERY rinky-dink heatsink, but still
Top right area / outside of the schematic delineation. "T.P. 34 33"
The dummy loads are pretty well matched 8.2 ohms each according to my Fluke. The emitter resistors are a little harder to know for sure, but I get between .6-ish and .7 on them.
I've replaced C17, and 18
This for sure all meaningless, if we can switch gears to assessing whether or not it is actually overheating, then it would be mooooooooot!
I've posted that question elsewhere and haven't a response.
If worried, for peace of mind I would add a 75-80C thermal switch to heatsinks, "just in case".
It will not solve your current problem , but you will be able to enjoy your amp without a chill in your heart.
Wire it to cut mains voltage, properly isolate its terminals, at least with heatshrink.
It will not solve your current problem , but you will be able to enjoy your amp without a chill in your heart.
Wire it to cut mains voltage, properly isolate its terminals, at least with heatshrink.