Hi,
I’ve had a Rotel RB-870BX since new that I’ve been very pleased with. Recently I bought a second one, but unfortunately this one is slightly noisier and has a hum in the left channel.
Compared the inside of both amplifiers and they are absolutely identical and there are no signs in the “new” one that it has ever been fixed, repaired or modified in any way.
I guess the large 10000uF power supply capacitors for the left channel could be likely suspects?
But actually any suggestions on where to start tracking down and hopefully fixing the hum issue would be greatly appreciated.
Link to the service manual: http://bwgroupsupport.com/downloads/techmanuals/rotel/RB-870BX-TM.pdf
I’ve had a Rotel RB-870BX since new that I’ve been very pleased with. Recently I bought a second one, but unfortunately this one is slightly noisier and has a hum in the left channel.
Compared the inside of both amplifiers and they are absolutely identical and there are no signs in the “new” one that it has ever been fixed, repaired or modified in any way.
I guess the large 10000uF power supply capacitors for the left channel could be likely suspects?
But actually any suggestions on where to start tracking down and hopefully fixing the hum issue would be greatly appreciated.
Link to the service manual: http://bwgroupsupport.com/downloads/techmanuals/rotel/RB-870BX-TM.pdf
This unit has separate bridge rectifiers and smoothing caps for both channels. I'd check the rectifiers for shorted diodes, if those are fine a bad smoothing cap does seem reasonably likely.
The unit being more noisy is interesting, as noise of these should be negligible vs. most preamps. Maybe the bridge mode switch and/or relay have contact issues and would appreciate a bit of a workout.
Of course you might've gotten unlucky and the unit contracted a hairline crack in the PCB during shipping, which in turn could cause all kinds of funny issues - definitely worth checking. How well was it packaged?
The unit being more noisy is interesting, as noise of these should be negligible vs. most preamps. Maybe the bridge mode switch and/or relay have contact issues and would appreciate a bit of a workout.
Of course you might've gotten unlucky and the unit contracted a hairline crack in the PCB during shipping, which in turn could cause all kinds of funny issues - definitely worth checking. How well was it packaged?
Last edited:
100HZ means smoothing issues. Possibly the smoothing caps on the affected channel. If you bridge the large caps with some 4k7u caps and it improves, I would replace the old capacitors.
Aluminium Capacitors | RS Components
Aluminium Capacitors | RS Components
Thanks for the good suggestions!
It was shipped in the original box and the box wasn't that badly bruised in transport, so I doubt there are any cracks in the PCBs or broken solder joints. The difference in noise level is really,really minor, so that no big deal. But the hum can be heard 1 meter from the speaker when the room is silent.
I finally got some time to investigate the issue some more, and the rectifier is OK and the capacitors appear OK too.
The ripple voltage on the positive supply is around 12mV p-p and 10mV p-p on the negative supply, guess that is OK so the problem probably is somewhere else?
I dug up my old analog oscilloscope from storage and powered it on for the first time in 15 years to check the output signal to the speaker and this is the result, (sorry for the lousy picture).
The time is 5ms/div and 5mV/div which is on the edge of what this scope can detect.
Any ideas what might cause this "error" in the output signal? And/or where to start a deeper investigation?
It was shipped in the original box and the box wasn't that badly bruised in transport, so I doubt there are any cracks in the PCBs or broken solder joints. The difference in noise level is really,really minor, so that no big deal. But the hum can be heard 1 meter from the speaker when the room is silent.
I finally got some time to investigate the issue some more, and the rectifier is OK and the capacitors appear OK too.
The ripple voltage on the positive supply is around 12mV p-p and 10mV p-p on the negative supply, guess that is OK so the problem probably is somewhere else?
I dug up my old analog oscilloscope from storage and powered it on for the first time in 15 years to check the output signal to the speaker and this is the result, (sorry for the lousy picture).
The time is 5ms/div and 5mV/div which is on the edge of what this scope can detect.
Any ideas what might cause this "error" in the output signal? And/or where to start a deeper investigation?
Attachments
That looks like the kind of problem caused by an imperfect grounding scheme. Those pulses are the reservoir cap charging currents entering the audio path.
If its like this with the inputs shorted and nothing else connected then its probably as good as its going to get without major diagnostics to look at where the problem is... problem being down to the design rather than a faulty part.
If its like this with the inputs shorted and nothing else connected then its probably as good as its going to get without major diagnostics to look at where the problem is... problem being down to the design rather than a faulty part.
Since this was -at the time- Rotels top of the line amplifier I would be a bit surprised if the issue was because of bad design.
The amplifier has no mains ground (/safety earth) connection and both of the amplifier PCBs are connected to the chassis at one single point.
I will remove all connections to the left amplifier circuit board except the cables to the power transformer and the speaker terminals and short the input pins on the PCB. If that doesn't have any effect on the hum at least I know that the issue is somewhere on the PCB.
The amplifier has no mains ground (/safety earth) connection and both of the amplifier PCBs are connected to the chassis at one single point.
I will remove all connections to the left amplifier circuit board except the cables to the power transformer and the speaker terminals and short the input pins on the PCB. If that doesn't have any effect on the hum at least I know that the issue is somewhere on the PCB.
It happens
It might be worth loading the rails one at a time with a mains bulb to pull some current. If the grounding scheme is correct then the increase in ripple should not affect the residual hum at the speaker output.
You mention in post #1 of having another identical amplifier. Is that one truly silent with no residual at the speaker terminals ?
I'm starting to realize just how right you where Mooly.
It's a grounding issue and it wont be easy fix completely.
One thing I found is that there is a 1K resistor on each input that is not in the service manual schematic, so even when I thought I had grounded the inputs there was actually a 1K ohm resistor between signal and ground. Explains why shorting the input pin on the PCB worked so much better than shorting the RCA-connector.
What isn't obvious in the service manual is that the grounding between the 3 BCPs and the chassis is a single point where the grounding cables from each PCB connects to each other and the chassis, which is good.
But there is also a second ground connection between each of the amplifier PCBs and the input PCB through the signal cable shielding. (Two small ground loops there).
By testing all the different combinations these 5 ground paths could create I found that I could make the issue worse by only connecting the right amplifier to the chassis and on the other end force the hum to the right channel(!) by only connecting the the right channel to the input PCB.
But connecting the left amplifier only to the input PCB reduced the hum to a level I can live with, (at the cost of a slightly more noise), so I will probably leave it there.
Thanks everyone for all helpful replies!
It's a lot worse in the barrel below.
It's a grounding issue and it wont be easy fix completely.
One thing I found is that there is a 1K resistor on each input that is not in the service manual schematic, so even when I thought I had grounded the inputs there was actually a 1K ohm resistor between signal and ground. Explains why shorting the input pin on the PCB worked so much better than shorting the RCA-connector.
What isn't obvious in the service manual is that the grounding between the 3 BCPs and the chassis is a single point where the grounding cables from each PCB connects to each other and the chassis, which is good.
But there is also a second ground connection between each of the amplifier PCBs and the input PCB through the signal cable shielding. (Two small ground loops there).
By testing all the different combinations these 5 ground paths could create I found that I could make the issue worse by only connecting the right amplifier to the chassis and on the other end force the hum to the right channel(!) by only connecting the the right channel to the input PCB.
But connecting the left amplifier only to the input PCB reduced the hum to a level I can live with, (at the cost of a slightly more noise), so I will probably leave it there.
Thanks everyone for all helpful replies!
You forgot "high end" before barrel.
It's a lot worse in the barrel below.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.