capacitor multi can question

[Randall]

I am recapping an old Scott stereo tube amp (21 tubes!). I came across this one 4 x 100 uF @ 75vdc that reads "can com pos". I have no schematic. Does this mean these caps are wired in reverse of what I think they are?

[Bruce / Mission Amps]

I am recapping an old Scott stereo tube amp (21 tubes!). I came across this one 4 x 100 ohm @ 75vdc that reads "can com pos". I have no schematic. Does this mean these caps are wired in reverse of what I think they are?

Picture please... Ohms and Capacitance are apples and oranges!!!
And I guess, it is possible that it is wired in reverse to be used in the bias supply or who know what else....

[Enzo]

Yeah, MAYBE for bias supplies. And I agree with Bruce, ohms and capacitor doesn't make sense. But come on Randall, tell us the amp model and we might just have schematics.

[olddawg]

Shouldn't be too difficult to figure out with an ohm meter. I would just use 4 100uf caps @ 75vdc or better. They won't be very large nowadays. You cold probably stuff them in the old can. That is assuming that it is uf not ohms of course.

[Randall]

Whoops. I meant 4 x 100 uF @ 100vdc, sorry about that. It's a Scott Stereomaster 380, and it is a bias supply. I must have done something wrong because it's pulling over 2 amps vairac'd up to around 90vac. I have it on a bulb limiter, and the way I have it wired at his point there are negative voltages on those four caps, from about -4v to -16v.

I found the schematic: http://hhscott.com/pdf/380.pdf

The 4 x 100 is actually now a 2 x 100 can with two 100uF/100v caps wired in. I have triple checked my wiring. The can gets really hot and blows the fuse. I found a leaky rectifier diode on the negative side of the bridge, so I replaced them all with fresh, but it still gets really hot. What I am not getting here is why the positive (right) side of the bridge is connected with a black wire that goes to all the other can case lugs, and is chassis grounded. The can I have now has a floating neg. lug, which is where the neg side of the other two caps are as well as the cathode of the two bridge diodes, and the black wire going to the other can lugs. That is where the wire was connected on the original can.

Is that Right?

[Enzo]

If the can gets hot, I bet it is backwards. Why is the + end of the bridge grounded? Because it is a negative supply. A power supply can be either polarity depending on how you reference it to ground. Imagine a 9v battery. Ground the minus terminal and the Other terminal is +9v. But if you ground the positive terminal then the other one is -9v.

The old cap said right on it can common positive, so all the caps have their + ends together on the can. Perfect for a negative supply, because you will WANT the + ends grounded, and the can is a simple way to do that.

Now look at your schematic, there are four caps in that negative supply. The positive end of each is grounded. That means the + ends are all wired together and to ground - not separate. The negative ends are the hot ends - they are NOT wired together. You have described wiring all four negative ends together. And then connecting them to ground. yes the old can was grounded, because it was the positive end of the caps. Your ground wire needs to connect to the + ends of all four caps. Unless you can find a new can with positive ends common and can positive, you are not going to be able to use a can. Well, unless you hollow the old one out.

I think you are assuming the can needs to be grounded no matter what, but the old cap was specially made for this.

[Randall]

Now look at your schematic, there are four caps in that negative supply. The positive end of each is grounded. That means the + ends are all wired together and to ground - not separate. The negative ends are the hot ends - they are NOT wired together.

Ding! That, of course is it. I learned two things the hard way today. First, I have never seen a reverse polarity can before, nor heard of one. This is what prompted my post, I smelled a rat, and I found one! Second, I was not able to determine polarity of any can caps from the schematic until after Enzo's lesson. Then I deduced that the positive side is marked by the geometric symbols on the can, ie square, triangle, D. That explains a lot. Just as well that I can't use that can after the abuse I put it through. I'll fit in a couple of more 100 uF caps in there and I should be good to go.

Thank you very much Mr. Enzo, and everyone else for making this forum such a rich source of wisdom, and for sharing it so unselfishly.

[Enzo]

Actually, the geometric symbols indicate the terminals in the center of the can end. MOST of the time, the can is the negative common and those are the positives, but your bias supply can was backwards from that. Look at the can end. See the terminals? They poke up from inside the can. Now look closer. One will be just a plain terminal poking up through a slot, but the others will have a bit of hole next to the terminal, and the hole will be a triangle or a square or a half-round. In the drawing, the polarity is inferred by the circuit. I admit those symbols can be vague, usually there is a + sign or the cap has one straight line and one curved line.

The symbols indicate the hot lead of the cap, the ones on the terminals. If the can is wired positive to can, then they will be negative. IN most cases they will be positive because the can is usually negative.

[Randall]

In the drawing, the polarity is inferred by the circuit. I admit those symbols can be vague, usually there is a + sign or the cap has one straight line and one curved line.

The symbols indicate the hot lead of the cap, the ones on the terminals. If the can is wired positive to can, then they will be negative. IN most cases they will be positive because the can is usually negative.

I do realize the symbols identify the can terminals, I just always assumed they identified the positive side. I must say I am not impressed by those polarized electrolytics being drawn with two straight lines and no + or - signs. The only distinction I can see in that drawing is where those geometric symbols are placed, which to me seems like they are indicating the positive side of the caps, based on how the main power supply section is drawn. SOOO.. on the cap in question the symbols identify the terminals on the physical can, but not the terminal side on the schematic. A little vague perhaps.

[Enzo]

Ah, I see what you mean. Yes, the shapes ought to have been on the lower side of those four caps. They ALMOST always identify the positive side because the vast majority of caps are made that way. Your cap is very specially made for this amp.

Just simply from years of experience, I often don't even see things like that because it is intuitive by now which terminals go where to me. SO, good eye, Randall.

[Randall]

Tricky, hard and potentially expensive lesson learned. But it's up and working now, the customer is thrilled, and once again wisdom is passed down here from the willing to the eager. Thank you.

[Steve A.]

Enzo: Is there a way to test the polarity of an electrolytic cap without blowing it up half of the time? For example like feeding it a very small voltage and observing readings on your DMM...

Steve Ahola

[Enzo]

I never thought about it. Cans are most always marked as are most parts.

The engineer guys like RG or Loudthud or Steve Connor might have some ideas.

[Steve Conner]

I think it's pretty hard to measure the polarity of an electrolytic cap without damaging it. That's why they are always clearly marked.

99% of the time, the can is the negative terminal. Even in capacitors that don't use the can as a connection. The only exception I know of is 600V Sprague Atoms, I believe they are two caps in series internally, and the can floats at about half voltage.

And now I know about this "can com pos" thing too.

PS: Does anyone know what the can of a non-polar electrolytic is connected to? I think the last one I saw had rubber bungs in both ends like a 600V Atom.

[Mick Bailey]

Just did a quick search on testing the polarity of caps. A few interesting methods using discharge curve timing and oscillators.

I noticed that my capacitance meter won't give a fixed reading if the cap is backwards.

The NP caps I have are insulated from the can, but I had a crossover for repair last week and the NP caps looked just like regular ones, with one lead welded to the can.