FX Loop Buffer Circuit Hum (Not a Ground Loop) (1989 Laney AOR 50 Series II)

[neptoess]

I didn't write down DC resistance across C30, but it was over 100 kOhm and rising when I measured it in circuit (I'm aware that I need to remove it to get a really accurate estimate, but it still rules out dead short, 1k leakage resistance, etc)
I also didn't lift the chassis connection (because I didn't check this before going into the amp), but wouldn't that just give me no signal? The signal coming into the return would be unreferenced.

And yeah I'm beginning to think I have some inaccurate measurements as well. I was using insulated clips, as that seemed to give me a lot cleaner readings. Either way, I'm thinking I'm lacking the tools to effectively diagnose this issue. Particularly, an oscilloscope.

So, I've decided I'm just taking the amp to the shop. I'll make sure to update the thread, but I've just sunk way too many hours into playing with this stupid thing. If it didn't have any sentimental value to me, I would have been much further ahead driving to Samash and trading it in on a 2203.

[Helmholtz]

Well, troubleshooting often requires some "Sitzfleisch" (patience, stamina, endurance).
I think we could have done without a scope. And we have some excellent "repair shop technicians" here.

BTW, a Marshall 2203 while providing awesome distorted sound (clean not so much), doesn't have an FX loop.

[neptoess]

I'm open to hearing more suggestions. When I had it open, I did try to dig further into the source of the hum, but it was starting to make my head spin.
I took these measurements multiple times, with the black probe inserted into the chassis-mounted ground jack on my amp and an insulated clip on the red probe, and the amp humming.

1. Junction of C32 and R44 (measured by clipping R44) measures 0 VAC (fluctuates a bit, up to 0.2 mVAC, but I'm going to call it 0). This tells me IC3B is giving me no output, which is good. That's correct.
2. Junction of C32 and R45 (measured by clipping R45) measures 3 mVAC. This is weird. Mega weird. With nothing plugged into the return, and the amp not humming, I only measure 1.2 mVAC here.
3. Junction of R45, R46, and TS6 measures 2.5 mVAC (measured by clipping R45, then again by clipping R46).
4. The grounded leg of R46 measures 1 mVAC. Pretty weird. With power removed, I measure the resistance from this point to the chassis ground as 0.8 ohm. I see 0.3 ohm from the sleeve of the return jack to chassis ground. This is a difference, but it's only 0.5 ohm. I can't imagine this is the issue, particularly because I get no noise with a pass-through cable inserted between the send and return jacks.

I did start scratching my head about TS6, but it looks like all it's there for is to ground any preamp output when the power switch is off.

Also, Helmholtz , I was talking about a new, reissue 2203, which does have an effects loop. If I buy an old one, it will be a JMP, and those aren't the easiest to find.

EDIT: I should also point out that I'm not using a Fluke. I'm using a cheap Radioshack meter. It's repeatable, but measurements this small probably have a good deal of error in them.

[Helmholtz]

ACV meter readings below maybe 5mV are typically not reliable. Short the tips of your meter leads, what reading do you get with leads laid across the chassis? Typically the AC field of the the PT will induce some mV.

Why don't you measure hum at the amp's output terminals (OT secondary, speaker jack)?

Do you have around -15VDC at the output and both inputs of IC3B? Be sure to use a good grounding point for the black/negative lead of your meter, e. g. the grounded side of R43.

[Helmholtz]

Also meter tips are typically sharply pointed, because it may be necessary to scratch/poke through an insulating layer of oxide or solder flux to get good contact.

[neptoess]

Yeah the very low values of the readings make it hard to for me to convince myself I'm actually measuring the problem and not being tricked by field hum.
I was measuring in the preamp to try and pinpoint _where_ the issue originates. I can plainly hear the hum from the speaker. Wouldn't measuring the OT secondary / speaker jack just validate that?
I do have 14.4-14.6 VDC on the op amp input (both inverting and non-inverting) and output pins. I measure 0 VDC on pin 8 (Vcc+) and -29.5 VDC on pin 1 (Vcc-)

EDIT: Yeah I know all about having to scratch around. That's why I like my insulated alligator clips. They bite pretty hard, so I don't have to put as much effort into getting the probe to make contact.
EDIT 2: Also, I'm positive the field hum could affect my meter. Just waving the positive probe around inside the amp chassis with the negative probe grounded, I can easily find spots that will read >100 mVAC without the positive probe touching anything.

[Helmholtz]

Wouldn't measuring the OT secondary / speaker jack just validate that?

Yes, but it may be helpfult to have some quantitative reference.

I do have 14.4-14.6 VDC on the op amp input (both inverting and non-inverting) and output pins. I measure 0 VDC on pin 8 (Vcc+) and -29.5 VDC on pin 1 (Vcc-)

Good. Is this with and without the shorting plug?

Also, I'm positive the field hum could affect my meter. Just waving the positive probe around inside the amp chassis with the negative probe grounded, I can easily find spots that will read >100 mVAC without the positive probe touching anything.

Open meter leads is a somewhat unrealistic condition, because it doesn't allow for induced currents. Rather you're measuring capacitively coupled electrical (as opposed to magnetic) interference.

[neptoess]

Good. Is this with and without the shorting plug?.

Yes. My previous 0 V measurement here was wrong.

[Helmholtz]

I somehow missed that you found the hum frequency to be 120Hz not 60Hz. That more or less rules out the opamp supply (-30V/ -15V) as that would have 60Hz ripple (half-wave rectification).

You also indicated that the previous owner did some modification/repair?

[nickb]

I somehow missed that you found the hum frequency to be 120Hz not 60Hz. That more or less rules out the opamp supply (-30V/ -15V) as that would have 60Hz ripple (half-wave rectification).

You also indicated that the previous owner did some modification/repair?

Without seeing the specturm for ourselves see the I really don't think we know what the fundamental. Be careful not to go off into the woods.

[pdf64]

Without seeing the specturm for ourselves see the I really don't think we know what the fundamental. Be careful not to go off into the woods.

The sound clip in post #41 sounds a lot more like 120 than 60Hz, though I guess the cab will be filtering it.

...I also didn't lift the chassis connection (because I didn't check this before going into the amp), but wouldn't that just give me no signal? The signal coming into the return would be unreferenced.....

There's a screened cable on the jacks which is going back to the pcb, so it should continue to get its 0V reference.

[Pedro Vecino]

Some time ago I built some loops with LND150 transistors for a series of 18 watt amplifiers. This loop is powered with high voltage taken from the screen grids node with a series resistor to reduce it to 300V and an electrolytic capacitor behind it.
I connected the module's ground line to the star ground.
I tested it by interrupting the signal in the return (the amp stopped sounding) and connecting a cable between send and return recovering the sound. I adjusted it to sound identical activated and bypassed. Zero noise. I thought it was perfect.
The next week I plugged in a processor to test and a nasty hum appeared. Something completely unforeseen.
The solution I took was to connect the general ground line of the module next to the socket and connect the negative of the filter capacitor (separate from the above) to the star ground. The hum disappeared completely.



It is certain that in this amplifier the ground of the jacks is connected to two different places. The one that leads the shielded cable from the board and the direct connection to the chassis (the one with the rivet).
I think it could be related to my problem and with the ground distribution.

[nickb]

The sound clip in post #41 sounds a lot more like 120 than 60Hz, though I guess the cab will be filtering it.

I'd need to see spectrum made with a decent tools. The cabinet, speakers and mp3 don't make the grade by a long way.

[neptoess]

The sound clip in post #41 sounds a lot more like 120 than 60Hz, though I guess the cab will be filtering it.


There's a screened cable on the jacks which is going back to the pcb, so it should continue to get its 0V reference.

The cables only have the shield connected at the jacks. See the two red arrows in the bottom right of this picture
https://i.ibb.co/4sqmkyF/loop.png

[neptoess]

Some time ago I built some loops with LND150 transistors for a series of 18 watt amplifiers. This loop is powered with high voltage taken from the screen grids node with a series resistor to reduce it to 300V and an electrolytic capacitor behind it.
I connected the module's ground line to the star ground.
I tested it by interrupting the signal in the return (the amp stopped sounding) and connecting a cable between send and return recovering the sound. I adjusted it to sound identical activated and bypassed. Zero noise. I thought it was perfect.
The next week I plugged in a processor to test and a nasty hum appeared. Something completely unforeseen.
The solution I took was to connect the general ground line of the module next to the socket and connect the negative of the filter capacitor (separate from the above) to the star ground. The hum disappeared completely.



It is certain that in this amplifier the ground of the jacks is connected to two different places. The one that leads the shielded cable from the board and the direct connection to the chassis (the one with the rivet).
I think it could be related to my problem and with the ground distribution.

See red arrows in bottom right of https://i.ibb.co/4sqmkyF/loop.png. The shield is only connected to the ground at the jacks, not at the PCB as well.