Marshal JMPl cathode cap question

[Randall]

70's JMP Master Volume Lead 50 watt has a 330uF cathode cap for the Low side of V1. It is drawn as not electrolytic, but it sure looks like one. The writing is facing down, so I can't read anything. I did manage to get a dentist mirror in there to confirm the value. It reads around 960uF in circuit, so I am suspicious of it. Forget about flipping the board to measure it out of circuit, too many connections, and if I do end up replacing it, I will snip and tack it in.

So, is this an electrolytic or not, and it shouldn't measure that high, should it?

https://www.thetubestore.com/lib/the...-Schematic.pdf

[The Dude]

Yes. It's an electrolytic. You probably won't get an accurate in circuit capacitance reading with the parallel resistor. Is there some symptom/circuit problem that leads you to think it needs replacing? Maybe you can get an iron under there to unsolder one side of that 820 ohm resistor, since it's on the edge of the board? You should then be able to get an accurate capacitance reading.

[Enzo]

Drawn as not electrolytic? You mean without a polarity +? Look at the rest of the schematic, NONE of the electrolytics have a polarity sign. Schematic notes say all caps in microfarads.

A 330uf cap is going to be either electrolytic or film. Can you imagine how huge a 330uf film cap would be? Look how big the 0.68uf right next to it is.

Yes, it is electrolytic.

No it wouldn't normally measure that high. But there is that 820 ohm resistor in parallel. You have another 330? or even a 220 or 470uf? Wrap an 820 ohm resistor (or 1k, something remotely close) in parallel with it and measure the result with your meter. That Will demonstrate that the meter can ignore the resistor or that the resistor confuses the meter.

Aha: Dude answered while I was typing and took a potty break.

[Randall]

"Drawn as not electrolytic? You mean without a polarity +? Look at the rest of the schematic, NONE of the electrolytics have a polarity sign."

Good point.

I realize it is in parallel with the resistor, and that will skew the measurement, but I didn't think it would skew so high. So, I did Enzo's test with various caps across an 820 ohm resistor, and sure enough some were very high indeed. I can't get to the resistor because it is right next to the chassis. I was going to replace it if it is an electrolytic because it is 40+ years old, and the customer wants the amp gone thru. There is no specific complaint.

On an anecdotal note, when I opened it up, I found someone had totally messed up the grounding scheme by running wires from the filter cap grounds to speaker, input and control buss grounds. THEN, they soldered in mid air, the AC safety ground, AND an old electrolytic cap with one side connected to nothing, just hanging next to the power switch. Evidently, someone was trying to fix a ground noise, and this was their "solution". Turns out it would come and go by pressing the first volume pot, so I hit those connections, and problem solved. Geesh.

[Enzo]

For reference: The way most little cap meters work is this: they send a small charging current out. the larger a cap is, the slower that current charges it. The meter detects the RATE of charge, and interprets that as a capacitance value. Slower charging means larger cap. WHen you put a resistor in parallel with the cap, the meter has no way of knowing that. but that resistor slows down the charging. So the meter assumes the cap is larger than it is. Does that make sense?

[Randall]

Indeed it does.

Customer opted to replace the filter cans, which means lifting the board, so that cathode cap is getting replaced while I'm in there. Be interesting to see what it measures out of circuit, but now I'm guessing it's probably not too far off.

[Chuck H]

Because it's a "maintenance" job I would probably replace the bias supply caps too. Just 'cause.

[Randall]

Someone already did that.

[mozz]

I'd lift one end, if it measures 300-360uf, leave it be.

[Helmholtz]

I never saw an ecap that actually increased its capacitance over time ( measured with a professional grade dedicated C-meter). Rather drying up of the electrolyte tends to lower capacitance.
Actually capacitance loss (typically 50%) is an end-of-life criterion for ecaps. Databooks charts show how capacitance decreases with operating hours and temperature.

[Randall]

"I'd lift one end, if it measures 300-360uf, leave it be. "

Well, of course so would I if I had reasonable access to it. Not wanting to disconnect umpteen connections to measure a cap, but if the board is coming up anyway, it's getting replaced.

[g1]

I never saw an ecap that actually increased its capacitance over time ( measured with a professional grade dedicated C-meter). Rather drying up of the electrolyte tends to lower capacitance.

I think the less professional cap checkers (like found on DMM) will show leakage as increased capacitance. So it is actually a measurement error. The principle is the same as Enzo outlined in post #5, just no resistor but cap leakage instead.
So, a good reason to be suspicious of caps that measure high when using a DMM cap function.

[Enzo]

On the other hand, this caap faces really low voltages, and unless there is some evidence of its failure, why tear the amp apart?

Are the other e-caps in this amp the same blue make and model? Or does it look like it was a replacement at some point?

[Randall]

"Are the other e-caps in this amp the same blue make and model? Or does it look like it was a replacement at some point?"

Aside from the filter cans, there are only 2 other e-caps in this amp, the two bias caps, which are newer Nichicons. The blue one is original I think.

[Chuck H]

"Are the other e-caps in this amp the same blue make and model? Or does it look like it was a replacement at some point?"

Aside from the filter cans, there are only 2 other e-caps in this amp, the two bias caps, which are newer Nichicons. The blue one is original I think.

It was a long time ago, but I had a similar era Marshall that was all original and I believe that cap looks VERY familiar.

But, as stated, it's a very low voltage circuit. I've noticed that bypass caps can live virtually forever and almost never read about them going south from age in amps that are used regularly. But then, considering that nearly all vintage amps used full bypass for most such circuits (the Marshall 680n being the exception) and the cathode resistor usually well exceeds a level of DC leakage we simply may not have noticed.?.