Unloading bridge rectifiers for testing where load and bridge are on the same PCB

[BenJBX]

Hi everyone, I'm working on a Sony TA-88 hifi amp which has pretty severe ripple on the DC supply rails. The ripple amplitude pulsates at about 1/30th the frequency of the ripple itself i.e. the pulsation waveform period is 700-800ms ... and I apologize I didn't grab a scope trace of this because my PC wasn't hooked up to the scope at the time (but just taking a pic on my phone would still have been more useful than trying to describe it )

Anyway I have two theories.. 1/ components in the bridge rectifier are not functioning properly or 2/ given that ripple is a function of load current, the load current draw is pulsating and creating the ripple pattern I'm seeing on the scope.

To test this I'd like to disconnect the load from the bridge so I can put the scope across it's unloaded output rails. The problem is the bridge is on the same PCB as the output stage of the amplifier circuit. I need to create an open circuit on the PCB, but I can't think how to do this without either desoldering several components, or deliberately damaging the PCB track which would be quicker but seems unnecessarily destructive.

Any tips/tricks for quickly and ideally non-destructively creating an open circuit to unload PS circuits for testing would be very gratefully received. P14-16 of the attached service manual has the relevant schematic and PCB layout although it doesn't show the PCB tracks unfortunately.

 

[Ike Zimbel]

The diode test function on a DMM usually works in circuit, so you could test the bridge diodes that way. And then, I would have a long look at C301 and C302, the main filter caps, as in a unit of that age, cap failure is a real possibility.

 

[JohnRoberts]

Of course make sure power is turned off and reservoir caps are discharged before measuring rectifier diodes with VOM.

JR

 

[ccaudle]

When the power supply impedance is too high you can get feedback from the output stage to the input stage, which results in low frequency oscillation often called "motorboating" because it sounds a little like a single cylinder engine idling.
Just start by replacing all the 50 year old electrolytic capacitors, that will likely cure the ripple and the low frequency noise.

 

[BenJBX]

Thanks for the replies, I've already replaced all the electrolytics because that was my first thought too. I will check the new ones I put in though in case there's a duff one in there.

I'll check the diodes with a DMM, good shout. I also think I can easily power down everything upstream of the output driver stage by lifting one side of R306, it's not fully unloading the bridge but it's a start. I have another issue to diagnose which is constant static sound on the left channel, but the volume fader attenuates it so it must be in preamplifier circuit somewhere. If these two are related, powering down the preamplifier circuit may improve the ripple on the bridge output. I'm away now til the weekend but I'll try that when I get back.

 

[BenJBX]

When the power supply impedance is too high you can get feedback from the output stage to the input stage, which results in low frequency oscillation often called "motorboating" because it sounds a little like a single cylinder engine idling.
Just start by replacing all the 50 year old electrolytic capacitors, that will likely cure the ripple and the low frequency noise.

Interesting, I'll read up on this. The modulation I'm seeing is too LF to hear but the descriptions sounds similar to what I can see on the scope.