Repair active subwoofers - Weird cap value and size

[JJanus]

Hello everybody, I just joined this forum as it is now of my interest to learn more about electronics and especially audio gear.
I'm also active in some other forum on DIY synth making.

I have a bit of experience (empirical) in repairing different stuff, like synths, drum machines, tape recorders…
I underwent the dismantling of a set of two subs: the Nova IN 15 SUB.
It is composed of an active sub with an amplifying module inside and I started to open this module after the sub started to me unresponsive to incoming signal.

What I discovered is not so unusual: a bunch of pretty bad, blown caps.

Now I wanted to start this repair by switching these bad caps to new caps but I stumbled upon some weird values and size.

They are big 25mmx50mm 2200uF 125V electrolytic caps.
I found out that this value of caps could not be found in 125V (really odd rating) and the only ones available in 200V are 10mm larger so would not fit on the board.

I have not investigated yet what their role is on the board, if it is power supply related or audio filtering related so I thought some of you guys would have a better guess than me.

I could theoretically solder bigger caps and glue the crap out of them on the board, but i'd rather not do that on such vibrating and sensitive equipment.

I emailed the manufacturers but they refused to send the schematics or a service manuel to me and just proposed that I ship the unit (expensive process i believe).

So, do some of you already stumbled across this kind of odd value?
Could these caps be replaced with other more common value?

Thanks a lot for you help and your time,

best,

JJanus




 

[schmitt trigger]

The available space is very tight. They may have chosen this odd ball value in order to fit a certain amount of microfarads in a can size that would fit within the space constraints.

Measure the height and diameter, and see if you can find something that will fit with a higher voltage rating.

[JJanus]

Hello schmitt trigger, thanks for your answer!

But that's the problem, I measured them and it was impossible for me to find this specific size (25mm width x 50 mm height) anywhere on my regular component suppliers as well as on Ebay or, auction sites.
The only caps with the right value and a 200V rating I could find were snap-in caps that are 10mm larger in width and thus will not fit the board.

[Twoflower]

On the second picture the rightmost cap has some wired looking material on the PCB. Is that glue to fix the cap or is that also a defective cap?

And I wonder why the four caps vented so violently. I wouldn't be too surprised if new caps will vent the magic smoke immediately.

[JJanus]

@Twoflower
I believe it is residue of glue, but I'll plan to change these caps too anyway. These are less hard to find value and size anyway.

I was also surprised the blew up like this but could not find anything else suspicious on the board. I beleive these subs just stopped working.
I must state that they were in a concert place that have sometimes 4 to 5 concerts a week so they were under quite a lot of stress. Could the capacitor just turn bad because of aging and then blow up because of heavy use?

[Twoflower]

I would suspect another defect. Some suspects are those four big diodes close by the big-blown-caps (in the second picture just below them). It looks like a rectifier bridge. If that fails the caps see AC and start to boil.

I would not expect such violently failure caused by age/reasonable overheat. To me it looks like that the partly melted plastic cover indicates that the electrolyte was very hot while venting. Which would indicate a rapid event.

[wraper]

Have you tried measuring actual voltage on those caps? If it's <90V, I would just use 100V caps instead.

[TheMG]

I would suspect another defect. Some suspects are those four big diodes close by the big-blown-caps (in the second picture just below them). It looks like a rectifier bridge. If that fails the caps see AC and start to boil.

I've also seen capacitors fail rather violently from power supplies going open-loop (voltage negative feedback loop fails, causing the power supply to go uncontrolled full-smoke-ahead and output way more voltage than the caps are rated for).

But what looks like melting could also be chemical damage from the venting electrolyte slowly reacting with the plastic over time (disclaimer: I'm not a chemist so this assumption could be completely wrong).

Advice: keep a close eye on output voltages when powering the power supply up after replacing the capacitors, just in case.

[picburner]

If the subwoofer is 1000W 8 Ohm the 125V could may be needed.
Probably the capacitor manufacturer built the capacitors on NOVA specification (capacity/size/voltage), however it was a risky idea to use 85°C capacitors in a switching power supply.

[JJanus]

Hello all and thanks a lot for you replies.
@Twoflower I'll look into the rectifier bridge and see if any of those diodes is broken.
@Wrapper At the moment, I do not have the power supply chord (A Neutrik PowerCON). As soon as i have retreived it, i'll try to probe around!
@TheMG I'll monitor the outptu voltage even though, without the service manual, it is hard to guess what kind of value I should get.
@picburner Ah! I did not think about the fact that some brand makes factories manufacture custom caps! It kind of make sense, the brand of the caps is Junzl, it appeared it is an Indian manufacturer.
Also, the unit is indeed 8 Ohm and 1.440W @ 4 Ohm!

[picburner]

If you are satisfied with a slightly smaller capacity I found this in a 25x50mm case:
United Chemi-Con (UCC) EKMR161VSN182MQ50S 1800uF 160Volts 105°C
The quality of the capacitor is certainly better and the supported voltage is higher.
Before replacing you must check that there are no other failures of course!

[JJanus]

@picburner
Thanks! I'm not sure I can guess what the consequence would be if I lowered the capacitance on the whole block of caps in this circuit though…

[EHT]

Can you not find 2200uF 150v 105C the same size on DigiKey or Mouser? I would also recommend getting a low ESR, high reliability type for SMPS. An alternative is to put cheap 2200uF >125V that might be too tall first, then continue testing. You may destroy the first set anyhow.

When you power it up, as others have said, there is quite a high likelihood it will blow again. I would recommend testing all the major components you can for shorts and physical damage before you power it up, i.e. all diodes/bridges and the power mosfets. One or two caps blown, maybe the cause but all 4 looks like a symptom.

When you power it up you could either use a variac or put an incandescent lamp in series with it. I haven't tried either technique with an SMPS but should be better than no precaution. It shouldn't draw much current if there is no fault and no input signal, so in that case the bulb will not light and not attenuate the voltage much. Maybe others have some comment about the efficacy of these methods on an SMPS.

[Yansi]

If the subwoofer is 1000W 8 Ohm the 125V could may be needed.
Probably the capacitor manufacturer built the capacitors on NOVA specification (capacity/size/voltage), however it was a risky idea to use 85°C capacitors in a switching power supply.

Why it is risky, please explain.

[wraper]

Probably the capacitor manufacturer built the capacitors on NOVA specification (capacity/size/voltage), however it was a risky idea to use 85°C capacitors in a switching power supply.

It is not risky as such. Done so in most high quality computer PSUs for high voltage PFC capacitors. Big snap-in capacitors still usually come in 85°C flavor. In this particular case they failed most likely because they are made by some crappy Chinese manufacturer.

Can you not find 2200uF 150v 105C the same size on DigiKey or Mouser? I would also recommend getting a low ESR, high reliability type for SMPS. An alternative is to put cheap 2200uF >125V that might be too tall first, then continue testing. You may destroy the first set anyhow.

For high voltage capacitors low ESR is not a thing. Even those which are available will certainly not fit as they are larger for same ratings while size already is an issue even for regular caps. Also 85°C capacitors are available in smaller sizes for same ratings.

[perdrix]

either use a variac or put an incandescent lamp in series with it.

Dim bulb incandescent lamps (in this case quite a few in parallel if this beast is rated to 1kW) is fine, but it's generally not a good idea to bring most SMPS up on a variac - they tend to draw way to much current at low voltages which causes lots of problems (normally of the magic smoke escape type).

David

[mcovington]

If you are satisfied with a slightly smaller capacity I found this in a 25x50mm case:
United Chemi-Con (UCC) EKMR161VSN182MQ50S 1800uF 160Volts 105°C
The quality of the capacitor is certainly better and the supported voltage is higher.
Before replacing you must check that there are no other failures of course!

Good suggestion.  I would say that 1800 is barely distinguishable from 2200, especially if the 1800-µF capacitor is inherently better quality.  That's what I would use.

[JJanus]

Hey guys, thanks for everything! I have enough elements to keep probing around and do some components order, I will keep you updated if I succed in the repair!
Cheers!