Can anyone say what the residual mains hum should be in a correctly working Quad 405 power amplifier, measured across a specified load (say 8 Ohms, RMS or Peak voltage) with the input open circuit? I have not found this information anywhere. It would help to know when to stop trying to fix it!
There are many threads referring to C5 as the usual culprit when hum becomes noticeable, but of course other components around Tr1 can be defective.
There are many threads referring to C5 as the usual culprit when hum becomes noticeable, but of course other components around Tr1 can be defective.
I've never seen official figures in that form but whatever the value is, as a user the noise should be pretty much inaudible.
Always use shorting plugs on any inputs when checking for hum/noise issues. Open or floating inputs are always prone to stray pickup.
The amp should be silent with such plugs fitted and only speakers connected. If you are using ultra sensitive speakers such as horns then what is a non problem in many amps can become more noticeable.
Always use shorting plugs on any inputs when checking for hum/noise issues. Open or floating inputs are always prone to stray pickup.
The amp should be silent with such plugs fitted and only speakers connected. If you are using ultra sensitive speakers such as horns then what is a non problem in many amps can become more noticeable.
With the input terminated with 1k (as per Quad service manual, p.5, #10) I get the same result: 5.5mV (RMS) across an 8.2 Ohm resister on the output. replacing the electrolytic capacitors on the amp board (C5 and C10 were below par) brought this to 2.5mV.
Sorry for the vague response but I'm sure there is a tantalum cap on the pcb that's failure (or just sub par performance from ageing) can be responsible for hum on the output. Apologies for the lack of clarity but its 15 years since I had my grubby paws in a 405.
It is possible this could be caused by deterioration of the main reservoir caps. I have the 405 set up as a simulation (never looked at ripple rejection before) and it seems that as little as a couple of volts ripple will generate the kind of levels you are seeing.
The injection mechanism seems to be via the positive rail rather than negative.
This is with 2 volts peak of ripple on each rail (making the negative rail pure doesn't change the result).
The injection mechanism seems to be via the positive rail rather than negative.
This is with 2 volts peak of ripple on each rail (making the negative rail pure doesn't change the result).
Attachments
Puggie: The tantalum capacitor you are thinking of is C2 at the op amp input. It is C5 which usually gives trouble which is a normal electrolytic, 100µF.
Mooly: I had indeed expected deterioration of the main reservoir caps, which is partly why I asked the question. Thanks for the simulation.
ScottJoplin: You are right to insist on a termination for the amp input. Quad calls for a 1k resistor across the input tags on the amplifier PCB. Since this requires 2.8 x 0.6mm connectors and soldering on a resistor I didn't think it practical to ask any Quad owner to go to this trouble, which is why I suggested an open circuit input.
After screwing on the top cover I find that I have a much lower hum and noise figure than I previously reported. These are the measurements with the top cover screwed on:
1. input s/c at input socket: 0.57mV RMS (all RH channel only)
2. input open circuit at input socket: 0.43mV RMS (hand capacitance effects at 12cm from socket)
3. Quad-specified 1k resistor across PCB inputs of power amp: 0.6mV RMS
Measurements 10Hz-100kHz BW with Levell AC microvoltmeter TM3B,
There is no typo above: the s/c measurement is higher than the o/c, and repeatable.
I think this amp is fixed and would meet Quad's -93dB unweighted below 100W output into 8 Ohms. Thank you gentlemen for thinking this through with me.
Mooly: I had indeed expected deterioration of the main reservoir caps, which is partly why I asked the question. Thanks for the simulation.
ScottJoplin: You are right to insist on a termination for the amp input. Quad calls for a 1k resistor across the input tags on the amplifier PCB. Since this requires 2.8 x 0.6mm connectors and soldering on a resistor I didn't think it practical to ask any Quad owner to go to this trouble, which is why I suggested an open circuit input.
After screwing on the top cover I find that I have a much lower hum and noise figure than I previously reported. These are the measurements with the top cover screwed on:
1. input s/c at input socket: 0.57mV RMS (all RH channel only)
2. input open circuit at input socket: 0.43mV RMS (hand capacitance effects at 12cm from socket)
3. Quad-specified 1k resistor across PCB inputs of power amp: 0.6mV RMS
Measurements 10Hz-100kHz BW with Levell AC microvoltmeter TM3B,
There is no typo above: the s/c measurement is higher than the o/c, and repeatable.
I think this amp is fixed and would meet Quad's -93dB unweighted below 100W output into 8 Ohms. Thank you gentlemen for thinking this through with me.
> no typo above: the s/c measurement is higher than the o/c
Shorts are usually quieter than opens.
If there is any input hiss current, short shorts it out, open lets it all convert to input voltage hiss.
And any "open" connection tends to absorb all the crap in the room and often inside the case. Again, short shorts this out.
Shorts are usually quieter than opens.
If there is any input hiss current, short shorts it out, open lets it all convert to input voltage hiss.
And any "open" connection tends to absorb all the crap in the room and often inside the case. Again, short shorts this out.
- Status
- Not open for further replies.