Old caps in a 1940's Webster Electric amp

[hearforever]

I'm trying to bring back a huge old Webster Electric amp from the 40's. Its built into a wooden cabinet and has a giant 15" speaker installed on it. The chassis is in two parts preamp separated from the phase inverter/power section. Inside the power section are some wild old paper filter capacitors and funny looking coupling caps. I believe the amp spent some part of its life sitting in a puddle of water, hence the rust (and the bottom of the cabinet is pretty warped). I think I'd like to go ahead and replace all the filter caps and these coupling caps with some new ones mounted on terminal strips. I'm a bit stuck on figuring out the values of these though. Anyone have any experience with these? Thanks very much!

[Prairie Dawg]

Those are called bathtub capacitors. The military and the Scott Radio people loved them. Here's a starting point for your research. Bathtub Capacitors

[Jazz P Bass]

I'm a bit stuck on figuring out the values of these though. !

[ATTACH=CONFIG]19009[/ATTACH][ATTACH=CONFIG]19010[/ATTACH][ATTACH=CONFIG]19011[/ATTACH][ATTACH=CONFIG]19012[/ATTACH]

Picture #1 ia a 20 uf/ 150V cap.
Picture #2 is a 20 uf/ 250 volt cap.
Picture #3 there are (2) .01uf caps. Not sure of the voltage rating.

[hearforever]

Those are called bathtub capacitors. The military and the Scott Radio people loved them. Here's a starting point for your research. Bathtub Capacitors

Thanks for the tip. So the smaller electrolytic values are obvious- 20mfd 150v
The coupling caps end with '104K' which means they should be replaced with .1 uF caps
The large capacitor looks to be a 6 section electrolytic... perhaps all 20mfd
The markings are strange though... I'm assuming the outer metal is common ground. As far as voltage, should I assume typical 450+ voltage?
Thanks!

[hearforever]

Picture #1 ia a 20 uf/ 150V cap.
Picture #2 is a 20 uf/ 250 volt cap.
Picture #3 there are (2) .01uf caps. Not sure of the voltage rating.

Thanks- there look like there might be 6 sections though in Picture #2. Picture #1/#3 I think I have the values for...I actually think its .1 not .01
but Picture #2 is where I'm stumped

[Jazz P Bass]

Yeah, the 104K means 0.1uf.
Sorry.

[hearforever]

OK so I am still trying to identify the big bathtub capacitor
It appears to be a 5 section electrolytic capacitor but I can't identify all of the invidivual sections' values or the voltage rating.
It looks like the upper left is the common ground terminal and I can make out 8 and 5 for the bottom left 2 terminals. I assume maybe the total capacitance of all sections summed is 20MFD as labeled on the outside of the cap.
Do you think it would be safe to just wire individual 10MFD with a rating of 450 or 525V capacitors?
The rectifier is a 5U4 so 10 MFD would be well below the max reservoir value

[Dave Mac Farlane]

Thing is, capacitor values are notoriously innaccurate. They often have huge tolerances, and people working with RC configurations often discover a change of the same type makes a difference (Me of little faith, once bitten, twice shy - used to be in research, where we had time to look into things like that).
To varythe capacitance by even 50% will get you working, and if the voltage measures off, then reduce or increase accordingly, and I include the amplitude of hum in that statement.
Quite often theoretical values miss out on something practical. These capacitors all look like a paper type of dielectric, and they were bad for leakage and distorting, although these look like mod style numbers.
Point is - don't go too far out, but try whatever you can get that's nearest - you can put two in parallel to double the value, and in series to reduce it, same as resistor theory but opposite - (sure you knew that, taking no chances),
But the voltage must at least be higher - it's only the efficiency that's reduced then, not the practical value.
Look forward to hearing of the finished result - looks good.
Dave.

[Enzo]

My working assumption would be that any section in it would be 20uf. Perhaps 5x20uf with C being common.

[hearforever]

My working assumption would be that any section in it would be 20uf. Perhaps 5x20uf with C being common.

That's what I thought too except the weird markings are throwing me off...
I think I'll probably just wire in new 20mfd caps seeing is thats not above the reservoir max value

[Enzo]

I could be wrong, but the marking I see look to me like they are labeled C,1,2,3,4,5. So common and five caps.

I suspect that 8 in the lower left is really a 3. Maybe?

Voltage rating is 250 volts at 800Hz. At least that is how I read it. ANy chance this was military equipment? Or aviation? A 250v AC rating would be about a 350v DC rating.

Back then (looks like 1942) we didn't have Hertz as a unit. We used cycles per second or cps. Can also be abbreviated cyc or cy. Today I might say "I applied a 400Hz signal." Then I would have said "I applied a 400 cycle signal." We use 60Hz for domestic power lines. Military mostly used 400Hz for their generators in the field. We can discuss why elsewhere. Meanwhile full wave rectified 400Hz power would result in 800Hz ripple. So I could be totally off base, but that is what comes to my mind when I see your cap - a multisection filter designed for use in a 400Hz powered circuit.

[hearforever]

I could be wrong, but the marking I see look to me like they are labeled C,1,2,3,4,5. So common and five caps.

I suspect that 8 in the lower left is really a 3. Maybe?

Voltage rating is 250 volts at 800Hz. At least that is how I read it. ANy chance this was military equipment? Or aviation? A 250v AC rating would be about a 350v DC rating.

Back then (looks like 1942) we didn't have Hertz as a unit. We used cycles per second or cps. Can also be abbreviated cyc or cy. Today I might say "I applied a 400Hz signal." Then I would have said "I applied a 400 cycle signal." We use 60Hz for domestic power lines. Military mostly used 400Hz for their generators in the field. We can discuss why elsewhere. Meanwhile full wave rectified 400Hz power would result in 800Hz ripple. So I could be totally off base, but that is what comes to my mind when I see your cap - a multisection filter designed for use in a 400Hz powered circuit.

Had the same thought after your reply. The "8" could be a 3. Still the 5 digits are in a weird order.
Thanks for the info thats really helpful. I think - with 5 20mfd -450v caps is the way to go. Will let you know how it works out and will try to post a picture of what this thing looks like in its cabinet. With any luck it should make a great low power bass amp.