Hello everyone,

I have a noob question, about (potentially) a PSU in a 5.1 audio / video receiver (from 2002 or so): Thomson DPL660HT (find technical manual here)


A few days ago, as strong summer started here, a hum started coming out of the front left speaker. Switching the cable (front left for front right) made the hum move to the other speaker, so that eliminated cable or speaker issue.

After opening the unit, I noticed a monster capacitor bulged. It was a 8200uf 63v one (I assume it belongs to the PSU as it's next to the animal transformer), on the main board. There was this and another twin 8200uf cap. The other one looked ok (although now I regret not checking it better).

After searching for a replacement and failing, I decided to try my luck and bought a 10000uf one (once again, I regretted right after, but regardless, nobody sells here 8200uf ones so it was that or nothing). See bulged cap in the photos below. Narration continues after:










After replacing it (worst nightmare ever given the burger-town design of this machine with multiple levels of boards and cables connecting everything to everything (literally, seriously, even though some boards are slotted to each other) the fix was done.
First day: no hum, all speakers good. Happiness.
Second day: gentle hum for a few seconds, then no more hum.
Forth day (or so): switch on immediately caused a switch off ("hello" .... "goodbye" shown in the display) with a not nice loud bang coming out of at least one speaker. This "little" issue happened 3 or 4 times until I quickly repeated the switch on action a few times and then it stayed on and came back to normal. The bang is randomly happening on one or another speaker, or not at all. The hum starts low, and gradually rises till an obvious level. After some 10 / 20 seconds it may abruptly disappear. It may gradually come back, then disappear. After a minute or so, it's totally gone.

Now my questions:

- Can the issue be related to me placing a 10000uf capacitor instead of a 8200uf? (voltage is the same, 63v). I'm particularly scared of the auto off situation.
- The unit has always been very hot in summer: can this be the root cause? Inside it has 2 very small fans that spin at not too high speed (I checked the caps in the fan board and the 3 of them seem to be good; maybe I should replace a resistor that controls the speed... Maybe not...!) Nevertheless I was planning on adding a big PC fan (whether internally or outside the case, to suck the hot air out through the vents) with an external PSU (don't wanna mess anything further).
- A crystal in the radio board (see 1st pic, middle-left area, floating board) was totally goodbye (see its black contents under it) and came off when accidentally touching it: can this missing component be causing a short or something? (I don't use the radio function, so it doesn't really bother me otherwise).

Thank you all very much for your time and help, any idea is greatly appreciated!

 

Well the first thing that strikes me is that you replaced just one capacitor rather than both, so by being different you might have an imbalance in some way. Also those large caps are C120 and C121 yet they do not appear in the schematic, only in the physical layout, so I have no idea of their role.

 

First rule of capacitor fight club, the weak die first and the next weakling in the chain will die next if the weak are replaced with the strong.

So, yes you should have replaced both at the very least and IMO all of the power supply section filter caps.

 

Oh, thank you very much both of you, it makes sense.
I'll try to go tomorrow or the next day and buy the other one. I'll check the ones around them and if they have them in the store, I'll buy them as well.

Do you think it can die if I use it once or twice as it is, before replacing the caps?

Thanks a lot for your help!

 

I defer to someone more expert than me on that overall safety question, I'd probably stop using it myself until the caps are replaced at least then you've eliminated them, for all you know something else is misbehaving too and might lead to greater damage.

The caps are likely 20% tolerance and 10,000 is actually about 22% away from 8,200 which isn't too bad I suppose. I'd try to get the 8,200 though, they don't seem rare or anything, where are you situated?

 

I am refraining from using it then, until I replace the cap(s).

Well I'm situated in Hungary, Budapest; that may very much narrow the possibilities! I have just found the 8200 in sketchy webshops, at crazy prices, or otherwise I've found a decent one but unfortunately it's massive: 350V 91x144.5. That would by no means fit, specially considering I need 2
Another option would be to bring them from China, but that would take about 1.5 months. Maybe I can try getting the other 10000uf one (so they are both equal) and later on order from China both 8200uf depending on how the child behaves.

Thank you very much for your great help!

 

I am refraining from using it then, until I replace the cap(s).

Well I'm situated in Hungary, Budapest; that may very much narrow the possibilities! I have just found the 8200 in sketchy webshops, at crazy prices, or otherwise I've found a decent one but unfortunately it's massive: 350V 91x144.5. That would by no means fit, specially considering I need 2
Another option would be to bring them from China, but that would take about 1.5 months. Maybe I can try getting the other 10000uf one (so they are both equal) and later on order from China both 8200uf depending on how the child behaves.

Thank you very much for your great help!

OK that's tough, so I suggest you note which capacitors need replacing, take their stated values and compute the +20% and -20% values for each of them, and then try to find suitably sized replacements that fall within those limits and have at least the same max wkg voltage.

Also the tolerance might be stated on the labels, and is not always + and - the same amount.

Oh and if you have a capacitance mode on your multimeter, it would be good to measure the values of each one you remove after its removed, be interesting to see if any of them really are acting up, do you still have the one you already removed, the one that was bulging?

 

OK that's tough, so I suggest you note which capacitors need replacing, take their stated values and compute the +20% and -20% values for each of them, and then try to find suitably sized replacements that fall within those limits and have at least the same max wkg voltage.

Also the tolerance might be stated on the labels, and is not always + and - the same amount.

Oh and if you have a capacitance mode on your multimeter, it would be good to measure the values of each one you remove after its removed, be interesting to see if any of them really are acting up, do you still have the one you already removed, the one that was bulging?

Today I bought all the capacitors with the same capacitance & voltage except the two fat ones (got another 10k uf). Replaced them all.

Unfortunately the situation is worse: when switching on, it immediately goes switches off, no chance for anything, no "goodbye" after the "hello". I'm not sure if there was no little smoke, although I try to be positive and think it was dust, or something else (I was working at night, with a headlight, and saw "something" in the air, but hopefully it's not...). I didn't smell anything, I think. The machine is full of fuses and I didn't see any blown.
What I managed to see afterwards (yes, I switched it ON again once or twice, because "now that we are here"...) is an "OVERLOAD" message on the screen that flashed for less than half second.

I'm assuming the two 10k uf capacitors instead of 8200uf are too much. Someone tell me this is the problem please...
I wanna cry like a child, damn it.

 

Today I bought all the capacitors with the same capacitance & voltage except the two fat ones (got another 10k uf). Replaced them all.

Unfortunately the situation is worse: when switching on, it immediately goes switches off, no chance for anything, no "goodbye" after the "hello". I'm not sure if there was no little smoke, although I try to be positive and think it was dust, or something else (I was working at night, with a headlight, and saw "something" in the air, but hopefully it's not...). I didn't smell anything, I think. The machine is full of fuses and I didn't see any blown.
What I managed to see afterwards (yes, I switched it ON again once or twice, because "now that we are here"...) is an "OVERLOAD" message on the screen that flashed for less than half second.

I'm assuming the two 10k uf capacitors instead of 8200uf are too much. Someone tell me this is the problem please...
I wanna cry like a child, damn it.

I don't think the 10,000 uF capacitors are a huge concern here, they only slightly outside of the 20% tolerance range. My next step would be to examine the re-soldering I did, checking for proper solder joints and verifying that the capacitors are inserted the right way around, I hope you understand that those capacitors MUST be in the right way, that's why they always have a mark showing the +ve or -ve pin, so make sure that none of them are wrong.

If any were then they are likely destroyed, they are not designed to operate in reverse, this conceivably might explain what you're now seeing.

 

 

What is the voltage rating of the replacement capacitor?? That is more critical than the capacitance value.

 

I did it good.
Luckily I took photos from similar angles (before and after):


If there's any bad joint, it can be on the 2 huge caps, as I had to extend their legs (given the caps are bigger and wouldn't fit). Although the legs were well soldered. But I can try to check that. Space is minimal, so I'll have to use a wire, needle, or something to reach the legs above and below the board. Also I'll have to dismount everything again...
Maybe in the meantime I can try to find 8200uf caps anyhow; I don't know.

 

What is the voltage rating of the replacement capacitor?? That is more critical than the capacitance value.

Thanks for your answer; all new caps voltages are matching exactly the old ones, being the big ones 63v and the others 35v, 25v and 16v respectively. All are 105º (higher than the original) and "high frequency" (no idea if that affects at all for this purpose. I know for audio yes, but for the task here I don't know).

 

I'm assuming the two 10k uf capacitors instead of 8200uf are too much. Someone tell me this is the problem please...
I wanna cry like a child, damn it.

It depends on where the capacitors are. If they're on power rails, they may be overloading the power supply. In some cases, higher or lower values won't make an appreciable difference. In others, it might.

A 20% tolerance 8200uF cap could be as high as around 9800uF, but a 20% 10,000uF cap could be as high as 12,000uF. Unless you think know more than the designer, you should stick with the values in the original design.

If the tolerance on the original cap was -20/+80%, substituting a larger cap might be okay. Even more likely to be okay if the replacement was a tighter tolerance.

 

It depends on where the capacitors are. If they're on power rails, they may be overloading the power supply. In some cases, higher or lower values won't make an appreciable difference. In others, it might.

A 20% tolerance 8200uF cap could be as high as around 9800uF, but a 20% 10,000uF cap could be as high as 12,000uF. Unless you think know more than the designer, you should stick with the values in the original design.

If the tolerance on the original cap was -20/+80%, substituting a larger cap might be okay. Even more likely to be okay if the replacement was a tighter tolerance.

Thanks Dennis, I understand.
If this may cause an overload, and the system reported "overload", then I think we may be having an answer here.
I definitely don't know more than the designer, of course. As I mentioned before, I tried the closest replacement I could find in my location. I'll try to expand my search, hopefully I can find the 8200uf ones!

Thanks everyone once again, I really appreciate it.

 

I recently read an article about the effect of the input capacitor on peak currents in power supplies. That discussion was relative to power transformer heating,, but it did talk about how it affected also the maximum power. Certainly that package is arranged in an unusual manner. I can see that it would be a serious challenge all around, just to get into it. Probably not designed to ever be serviced.

 

I recently read an article about the effect of the input capacitor on peak currents in power supplies. That discussion was relative to power transformer heating,, but it did talk about how it affected also the maximum power. Certainly that package is arranged in an unusual manner. I can see that it would be a serious challenge all around, just to get into it. Probably not designed to ever be serviced.

Yes, I believe this is already too much to invest on. Each time I need to test something, I have to unmount / mount back the full thing, and I'm not making good progress.

I put back the 2 original caps. Got fireworks and smoke. The original big cap that was partially bulged is now fully inflated. Maybe the sparks came from this one, but I don't know. It doesn't make any sense, as I just put the 2 original pieces as they were. Only changes were the other caps around that I previously replaced. But why would it blow? Seriously, why, theoretically? Can it be that the other caps around were having lower capacitance already, the were all in a (bad) "harmony", and having them replaced made the big one blow? For the curious ones: I didn't mess the polarity and I plugged all the cables good (painted marks on each and every cable with different colors).

I think I'm giving up. Each attempt takes about 3 hours, meaning I've spent several full days with this. Even if the 2 big caps are the reason, I would need to order them from China, pay a nice price, wait minimum a month, and maybe they are blown on the spot. I'll try to find another second hand system, and give this one for free to electronics workers. I know whatever from these times is probably gonna have a similar design (by the back connectors' arrangement I can already tell), but there's not much more that I can do.

Thanks guys for the support.

 

A failed capacitor bulges because the failure leads to excess heat, which expands almost everything. And large capacitor failure is usually a cascading process that is not going to reverse.

Replacement capacitors with dimensions similar to the originals should be available without ordering from China.
In addition, a request for service information posted on this website may eventually bring a better result than "check on-line" which will often bring ten thousand hits that bear no relation to the original search.

 

A failed capacitor bulges because the failure leads to excess heat, which expands almost everything. And large capacitor failure is usually a cascading process that is not going to reverse.

Replacement capacitors with dimensions similar to the originals should be available without ordering from China.
In addition, a request for service information posted on this website may eventually bring a better result than "check on-line" which will often bring ten thousand hits that bear no relation to the original search.

"And large capacitor failure is usually a cascading process that is not going to reverse." - Can you please be more specific? You mean that other components are also failing? Or this will lead to others components to fail? Or all together? In short: it will be a pain to try to fix it? Just give me straight directions here please, at this point I'm not so much into hypothesis. I was thinking of a last test: 1 original healthy 8200uf big cap + one 10k uf replacement one to see if it even starts, but I'm afraid of it blowing my new 10k uf cap. What do you say? I try it or I don't bother?

"Replacement capacitors with dimensions similar to the originals should be available without ordering from China." - I agree, there should be. But I live in a small country (Hungary) where these things are not manufactured, just imported. The cheapest and more straightforward is China. USA is too far, and it makes it too expensive. We don't have Euro here, which makes things complicated to import from Germany and others. And if 2 capacitors = half the price for a second hand system, then I guess they don't care to import them.

 

What does "overload" being displayed mean? what does manual say it means? I won't mean that some capacitor is misbehaving or too much current is being pulled, it likely means an amplifier input is overdriven, but can you tell us?

If any of the fans have failed, the system will shut off automatically so check all fans.