Should I replace the caps show below with electrolytic caps or a regular metalized polyester film cap?

Shown below are photos of the crossover network I pulled from the Magnavox 252MX console I'm working on. I didn't think they were electrolytic caps, but I noticed each of the four 1.0 uF caps is banded on one end and each pair of caps appears to be wired in series. A detail of the schematic is also show below.

They're not polarized. The stripe isn't 'minus', it's what it says, "outside foil" which may matter in some applications but not in crossovers.

"Ajax Condenser Co." eh? Sounds like something Wile E. Coyote would know something aboot.

Look specifically for "crossover capacitors". I can't tell you exactly what the difference is but I can tell you there IS a difference.

The caps in the pictures appear to be just very large paper/wax capacitors not electrolytic. Makes sense because they should not be polarized in an AC application. Film capacitors should work just fine.

Here is a link to some film caps specifically for crossover applications. http://www.parts-express.com/crossover- ... -index.cfm

They appear to be Polypropylene rather than polyester but I have no idea why.

Curtis Eickerman

Thanks Curtis and Arbilab. I order the caps from parts-express. Supposedly polypropylene are the best. I didn't go for the $15 ea audiophile caps though. $60 for four caps seems a little steep especialy since I only paid $20 for the console.

These retro components tend to be price jacked. Either legitimately because they are limited production runs, or just greed because many restorers will pay it. I'm sure your choice will be fine.

Then you can go on ebay and sell "antique wax capacitors" for twice what you paid for good ones.

That's a good decision. Most of the "audiophile" caps are built about like the gigantic originals, without the wax, and tend to be microphonic and prone to failure, also just like the originals. When I was regularly working on amplifiers, various people would request that I install them, which I strongly resisted. Even when I demonstrated that they were microphonic (oh, and the capacity changed when you squeezed them or bent the leads), they would still insist. A 25 cent polypropylene, no problem and generally much cleaner sound. You REALLY don't want them in a speaker crossover, but that's where they are most popular. "Concert Hall Sound" whether it's supposed to be or not!

BTW apropos of the original question, yes, those should all be non-polar but it's not at all uncommon to see the various irrationally-priced "salon" equipment to have electrolytics in AC applications and there have been some rather famous arguments over that topic - people who have even a scintilla of knowledge about the topic VS. the "Golden Ears". This argument was instrumental in the infamous "magic caps" phenomenon of the early 80's and was the tipping point, more-or-less, from legitimate engineering to the current "magic components"/voodoo school of high fidelity you see now.

Brett

Replace those with some 2.2uf/50 volt non-polarized electrolytic caps (they're tiny) and you won't be able to tell the difference.
And call it a day.

Since we are on the subject, in the schematic, reposted below, which are the woofers and which are the horns? Most of the wires to the speakers had been cut away, and I'm new to these crossover networks. On these Magnavox speakers, the horns aren't visible as they are placed in front of the woofers, behind the speaker grill.

It looks to me as if SP1 and SP2 are the horns as the green lead shown in the image below is coming from the horn in front of the woofer as is the black lead, which had been clipped. The white lead from the center of the crossover had also been clipped. So I guess SP3 and SP4 are the woofers?

I might just do that. I tested the caps, and here are the results:

Left pair: 2.24uF, ESR=.12ohms
Right pair: 2.38uF, ESR=.14ohms

Or maybe I should just use the original and not worry about replacing it.

I don't know how ESR meters work.

I do know, that capacitors working nominally at pulsed DC can be completely ineffective at operational frequencies. And I attribute that to ESR/freq.

I'd be skeptical of xover caps with ESR readings in excess of expected driver impedance.

According to the look-up table that came with the meter, the ESR numbers are well within the good range as are the uF values. Must have been pretty good caps.

ESR stands for "equivalent series resistance."

All real-world capacitors have an equivalent series resistance. A typical ESR meter uses a bridge circuit --

http://en.wikipedia.org/wiki/Bridge_circuit

-- to measure the degree to which a capacitor has a resistive element. To do this, the circuit must use an AC test frequency high enough so that the reactance of the capacitive element is very small. The ESR meter I use at my shop has a test frequency of 100kHz.

Additionally, the ESR meter must use a very small test signal. The reason for this is because many ESR measurements are taken in-circuit. A test signal of less than 100mV P-P will not turn on junction transistors, will minimally activate field-effect transistors and will not activate (or harm) digital circuits.

* * * * *

Another useful feature of an ESR meter is that it is, for all intents and purposes, a very sensitive low-ohms meter. Thus, it works wonderfully for testing cables, verifying ground integrity, and checking tonearm wires.

Fred
owner
Classic Audio Repair

I doubt those tweeters will be horns.

More than likely they will be the 5" Magnavox cone tweeters.

Either use back to back electrolytics at twice the .mfd ot use Non polarized caps. can parallel them up to get original value. Chances are tweeters are already gummed up enough to produce nothing anyway. Sorry.

If those tweeters are the type I think they are they will still be very good.