Glass diode rectifier? Silvertone stereo console question

[nsubulysses]

Trying to fix a Silvertone stereo console for someone. One of the furniture pieces from the 60s/70s. Basically just a AM/FM stereo and record player with speakers.

It had a loud hum so I recapped the power supply and accidentally put in the -25V supply cap backwards. After replacing the fuse and re-orienting the cap I turned it on with a bulb limiter just in case. Glad I did because one of the rectifier diodes was shorted. I pulled it out and it's glass but on the parts list for D301-304 it just says "rectifier" and that's it. They're all gunked up with glue because they were right next to a power supply cap so I can't read them. I put in a 1N4007 because that's what I had on hand and wonder if that's why I might have a weird buzzing, pulsing sound. What is the function of these glass rectifier diodes vs. silicon? Should I suspect this?

I get all my proper power supply voltages -- +/-25V (C310, C308A), +22V (C308B and C) and -7V (C353). Here is the power supply and output section





Here is the power supply for the radio tuner section. I don't suspect the noise is coming from here but just in case I'm wrong here it is. C506A, B and C were replaced but that's it.



The pain of this repair is that it is on site and I am not a master

Lastly, here is the noise

Sound familiar in any way? It gets a loud static burst every few seconds on top of the buzz.

[The Dude]

Does the noise go away if you unplug the "tuner input/signal socket"?

[dwmorrin]

I would think step 1 is to divide and conquer between the noise coming from radio or power amp. Can you easily unplug the radio input? The schematic suggests there is some connector you can unplug (S301). Monitor just the power amp with no input.

The bursts of noise suggest a bad transistor. You'll need to continue the divide and conquer approach. Is it in the left, right, or both channels? You can mute individual signal transistors by alligator clipping from base to emitter (I don't do that to power transistors though).

Glass diode vs. a 1N4007? I don't know of a reason not to sub modern diodes in there... can you monitor the power supply voltages with a scope during the noise? If those rails aren't noisy, then I wouldn't suspect the power supply.

[nsubulysses]

Thank you for your replies I forgot to mention that the reason I suspect it's coming from the power amp is because I did unplug the S301 plug and the noise persists.

I have the power amp and output section in my possession and can check it further. I will try muting transistors with the base emitter clip and also checking the power supply rails with the scope. From what I remember the noise was coming from both sides.

One question held me back from doing this yet though -- the left and right speaker terminals (by R321 and R371) are transformer coupled to the speakers. Would it be bad to run this with no load since it's transformer coupled? When I pull the two speaker terminal wires the total load of the ouput section reads about 110 ohms with a DMM. Coincidentally I do have 22 ohm 5W resistors I could use in series as a load, if necessary.

[audiopete]

The glass diode was probably germanium and the new ones are silicon and much more reliable.

[dwmorrin]

When you read the terminals you are reading R321 on the L, and R371 on the R. There is a driver transformer (T301) that drives output transistors (Q303/304), and the final output is capacitively coupled to the terminals (C304). That cap will block your attempts to measuring anything looking "into" the terminals.
Also worth noting is that there is a speaker off switch (SW451) that connects 45Ω 10W dummy loads on each output.
So there you go - make a 44Ω, 10W dummy load from your 5W resistors in series if you need it, and then mute one channel while monitoring the other. Do you need to do it? I'm not sure, but it's always good to play it safe when in doubt.

[52 Bill]

When you read the terminals you are reading R321 on the L, and R371 on the R. There is a driver transformer (T301) that drives output transistors (Q303/304), and the final output is capacitively coupled to the terminals (C304). That cap will block your attempts to measuring anything looking "into" the terminals.
Also worth noting is that there is a speaker off switch (SW451) that connects 45Ω 10W dummy loads on each output.
So there you go - make a 44Ω, 10W dummy load from your 5W resistors in series if you need it, and then mute one channel while monitoring the other. Do you need to do it? I'm not sure, but it's always good to play it safe when in doubt.

Each of the speaker outputs has a 100 ohm resistor load connected, so you probably don't need to bother with any additional load resistors, unless you are running the volume up really high.

The other three glass diodes still test correctly? What voltage drop do you get across each one? The voltage reading will tell you if they are germanium or silicon. In any case, I'd suggest replacing all four with the same type.

[nsubulysses]

You guys are great

[drewl]

Old home stuff like this usually has a resistor of some sort on the output so it doesn't fry when the owner forgets to hook up a speaker.
Some old guitar amps do also

[Tom Phillips]

I don't understand why no load on a SS amp output is perceived as an issue.
Also, it seems odd that the Silvertone Hi Fi amp switched in load resistors when the speakers were turned off
Can anyone explain?

[nsubulysses]

I thought that in general an amp that is transformer coupled to the speakers needs a load whether it is solid state or tube because of inductive kickback.

This is the first solid state output section I've encountered that has an OT. Unless I am interpreting it wrong as an OT, but since it switches in load resistors I think I am right

[Tom Phillips]

I thought that in general an amp that is transformer coupled to the speakers needs a load whether it is solid state or tube because of inductive kickback.

This is the first solid state output section I've encountered that has an OT. Unless I am interpreting it wrong as an OT, but since it switches in load resistors I think I am right

The transformer is driving the output transistors. It is not an "output transformer."

[nsubulysses]

Oh I see. I was misinterpreting the way the speaker section of the schematic is drawn. I thought the inductor symbol and the speaker symbol were symbolizing two separate things.

[Tom Phillips]

I see what you describe. The schematic is showing the speaker voice coil winding.

[Mick Bailey]

Maybe the amp output needs to keep a load on it to prevent the output caps from charging and then dumping this charge when the speakers are connected. The feedback path is also taken off after those caps.

A few SS amps will become unstable without a load and go into open-loop gain, but I don't see that being an issue with this model.