Help repairing a crossover

[CoolbeaN]

I had this project drop in my lap. A friend picked up a pair of 30+ year old M&K speakers at a garage sale for free because they were "broken" The peerless tweets in them alone are running for $50 a piece used online so already he made a $200 profit. He hooked them up to his system and said they worked fine. I hooked them up to my system and they played for about 5 minutes then I lost the mid range in one of them. I opened it up and found a resister had burned hot enough to singe and bubble up the board. I was able to find a suitable resister and replaced it, but I still had no continuity between the posts. Then I found a capacitor that appears to have failed. I did a little searching around and can't find an exact replacement. Here are all the numbers I have found on it.
1Si-7925
PXRC8JK50L
8uF-50VAC
AC

I did some testing using a multimeter and compared it to the other board and using the same test points everything else checks out so I'm hopeful this will work if I can find a suitable replacement.
If not I'll pull out the old crossovers (a hard task as they are glued in) and replace them with new.

[wolf_teeth]

Originally Posted by CoolbeaN

8uF-50VAC

This the value that matters. Higher VAC will be okay.

Now- did you cheack the continuity acrossed that component to decide that it needs replaced? If so- it will appear as an open connection with a standard DMM. If you use a C/L tester, it should read at value.

The reason- Capacitors are virtually infinite resistance at DC, which is what your DMM measures at. It might be okay, and not need replaced at all. Most poly caps at 8uF are rather inexpensive, so you wouldn't be out much trying it.

You can try out a cople of these:
Dayton DMPC-8.2 8.2uF 250V Polypropylene Capacitor | Parts-Express.com

Later,
Wolf

[CoolbeaN]

Thanks for the response.
I assumed failure because the top of the cap is domed and I think it might be leaking.. Either that or the installer got sloppy with the glue.
I just went to measure it after it has been sitting all night and it read an infinite resistance and 50V. touched the poles and now the resistance dropped to 1MOhm and is slowly creeping back up. Volts seems to creep up when not being measure, but if I have the tester on it, they creep down.

[BobEllis]

Low resistance creeping up is normal cap behavior if you connected the leads the opposite way around last time as the charge bleeds off through your ohmmeter and recharges the opposite direction.

Are you reading 50 VDC across the cap? That is odd - if so there must be a battery biasing it and any ohmmeter readings are incorrect. You'll need to disconnect at least one end of the cap from the circuit to test.

Such a low voltage rating makes me think it is a bi-polar electrolytic cap. How big is it? replacing an electrolytic with a poly cap may alter the sound if the crossover was relying on the electrolytic's ESR as a response shaper. Sometimes it works well sometimes not.

they played for about 5 minutes ... but I still had no continuity between the posts

Do you mean the main binding posts? Reheat all solder connections on the crossover. If there are crimped connections with insulated ends it is possible that over 30 years the wire has broken inside the terminal due to stress. The problem described sounds more like a loose/intermittent connection than a broken part. It may take pulling the crossovers, but if reheating a few joints (especially in the area of the burnt resistor) solves the issue you're way ahead)

Measure the DCR of your mid out of the circuit (at least one terminal disconnected). it should be somewhere under 10 ohms, really just looking for continuity. if open, look for a break in the tinsel wires leading to the voice coil. You may be able to repair a break there carefully soldering. Repeat for the woofer.

Pictures would be helpful if these measures don't pan out.

[CoolbeaN]

Those reading were with the cap disconnected from the circuit.
As it sits on my desk this second, it is reading .10 DCV and 3MOhm, I must have touched both poles when I picked it up.
If I let it sit for a while and test again, both the volts and ohms will be higher.

Yes, I had no continuity on the main binding posts.
I would love to pull the whole crossover out... but that is proving difficult... SOOOO much glue..... I didn't test the drivers.. that is a thought.

[BobEllis]

OK, that voltage is residual from the meter's testing. You shouldn't get higher than a couple volts.

Looks like a bi-polar electrolytic cap. You're not likely to find a poly cap to replace it at that size.

Check the leads from binding posts to crossover, too. Does wiggling them change anything? Is there a fuse on crossover?

[CoolbeaN]

I tested the drivers, both read 7Ohm. I got the crossover out by heating the glue with a hair dryer. I put the cap back in and went back over the board with my multimeter. I reflowed a few suspect connections, especially on one of the inductors. Put the drivers back on and got a good reading on the binding posts. Plugged it in for a quick test, everything appeared to be working fine. Put it all back together and it sounds great!
Thanks for the help!

[BobEllis]

Glad you're back up and running. What a deal!

Looks like they've been used hard - the one inductor really heated up it's bobbin.