Fire in the hole!

[kc8adu]

this would have been nice in the normal places a wallwart usually lives.
like behind a sofa or curtains.
in this case it was contained inside a tight enclosure for a paint mixing scale interface.
looks like the larger cap leaked out its bung and started an arc path that grew.it was a panasonic fc 470/35

[TechGeek]

Re: Fire in the hole!

Epic fail.... is the scale dead?

[Pentium4]

Re: Fire in the hole!

I wish all bricks had screws like that to make them easy to take apart.

Is the big bulged cap Samxon?

[kc8adu]

Re: Fire in the hole!

its already fixed.
had to wash the motherboard from all the cap juice and soot.
and to reduce the odor as much as possible.not good to have a caddy dealership stink of electrical fire.
didnt look at the primary cap.took it to the dumpster asap.

[stj]

Re: Fire in the hole!

how much pcb was left after you ground away the carbon??

[ben7]

Re: Fire in the hole!

Damn, scary! Glad it was caught before it started a severe fire.

Is it just me, or is the little cap bulged? Maybe that is the one that leaked?

If the panny did leak and do that, damn!
Also, was the panny original, or did it have signs of being a replaced part? Interesting that it is in there, with a Samxon, and another unknown cap (looks to me like a chinese one ... JunFu maybe?).

Argh! Samxon lets us down yet again!

Also, that PCB in the background looks interesting. What the heck was it from? Looks like something industrial or automotive. (Or maybe medical?)

-Ben

[stj]

Re: Fire in the hole!

i have seen FC's leak, and MBZ's leak too - it happens eventually.
for some reason i'v not worked out, it's always around the negative lead.

[ben7]

Re: Fire in the hole!

i have seen FC's leak, and MBZ's leak too - it happens eventually.
for some reason i'v not worked out, it's always around the negative lead.

Ah, well, I guess all good stuff has to come to an end eventually.

Interesting that it leaks only around the negative. Wish I could say why, but I'm not a expert on chemistry. Though I'd say it definitely might be some sort of electro-chemical effect.

[kc8adu]

Re: Fire in the hole!

the board is of the unit.it doses paint and pigments for color matching automotive paint.
i replaced the power supply.

[momaka]

Re: Fire in the hole!

Wow. Just barely 30W, this adapter, and it nearly became a firestarter.

i have seen FC's leak, and MBZ's leak too - it happens eventually.

I've never seen MBZ's leak, though I have seen Panasonic FL's leak easily even from the slightest heat during soldering. It's like they have some kind of very soft rubber bungs that melt easily at high temperature. Pulled a few from Asus P4SD motherboards, and nearly half of them got damaged leads. Compare that to some of the P4SD motherboards that had Rubycon MCZ - none of them pulled a lead out or leaked, even when I heated them for much longer and at higher temperature than the Panasonic FLs. Of course, MCZ has other problems (like not being able to tolerate heat well).
No brand is perfect, I guess.

[ben7]

Re: Fire in the hole!

Wow. Just barely 30W, this adapter, and it nearly became a firestarter.

It's possible to start a fire with even less energy than that! Scary isn't it?!

[Wester547]

Re: Fire in the hole!

I've never seen MBZ's leak, though I have seen Panasonic FL's leak easily even from the slightest heat during soldering. It's like they have some kind of very soft rubber bungs that melt easily at high temperature. Pulled a few from Asus P4SD motherboards, and nearly half of them got damaged leads. Compare that to some of the P4SD motherboards that had Rubycon MCZ - none of them pulled a lead out or leaked, even when I heated them for much longer and at higher temperature than the Panasonic FLs. Of course, MCZ has other problems (like not being able to tolerate heat well).
No brand is perfect, I guess.

See this thread.... https://www.badcaps.net/forum/showthread.php?t=46076/

Rubycon MBZ bulged in storage (4 years after being pulled from a GX260) so even MBZ isn't impervious to failure. On the other hand, you had Nichicon HZs bulge after one year in storage, after pulling them from a 360, so no aqueous capacitor is perfect. Are those the only Panasonic FLs you ever had do that? I'm sure you've pulled 1500uF 16V Panasonic FLs from 360 boards and none leaked? Those do use different bungs, though, compared to the 8mm ones, so perhaps tougher bungs.

My guess is that the Panasonic FC leaked because of a bad run of bungs, physically damaged leads whilst they were installed, or that it simply ran too hot for too long and the bungs decayed first.

[stj]

Re: Fire in the hole!

no, just age.
i'v seen everything from 6mm to 12.5mm and higher leak from the negative lead in different devices including an SGI 850w psu with a big delta 120mm fan on it.

if a cap leaks after more than 10years of service, i'm not going to complain - no matter what the make.

[momaka]

Re: Fire in the hole!

It's possible to start a fire with even less energy than that! Scary isn't it?!

Yes, it is . If you think about it, a 25W soldering iron can also start a fire. So no surprise. But they way the board in this thread is charred just looks really bad.

On the other hand, you had Nichicon HZs bulge after one year in storage, after pulling them from a 360, so no aqueous capacitor is perfect.

Indeed.
Though, I will also mention that I had this happen only with my 2005-datecode 6.3V 2200uF HZ Nichicons (H0544, in particular). Never with the 16V 1500 uF or 6.3V 2200 uF HNs.

Are those the only Panasonic FLs you ever had do that?

Yes. Probably 5 to 7 computers of those P4SD motherboards had them around the CPU. Rest were with bulged MCZ.

I'm sure you've pulled 1500uF 16V Panasonic FLs from 360 boards and none leaked? Those do use different bungs, though, compared to the 8mm ones, so perhaps tougher bungs.

You are 100% correct here.
I've desoldered hundreds of the 1500 uF 16V FLs (which are all 10 mm dia.), and I think I slightly pulled the leads on only one or two when I used a crappy iron for a few boards. They didn't leak, though. Moreover, once I pushed back their leads in, they still appeared functional. But I didn't trust them, so they are used as audio coupling capacitors on my breadboard. Hey, they work great for that application . Occasionally, I also use them as filter caps on my breadboard.

if a cap leaks after more than 10years of service, i'm not going to complain - no matter what the make.

I don't know. 10 years is a bit low in my opinion. Maybe in a hot environment, that's okay. But considering I have CRT monitors that are 15-20 years old now, run rather warm, and yet have only 85C caps that are still fine is a lot more like what I would expect from a capacitor.

[Wester547]

Re: Fire in the hole!

Indeed.
Though, I will also mention that I had this happen only with my 2005-datecode 6.3V 2200uF HZ Nichicons (H0544, in particular). Never with the 16V 1500 uF or 6.3V 2200 uF HNs.

You mean 2200uF 6.3V HZs? It really depends on how long either console was ran and whether either of them were previously reflowed. Those Xenons get VERY hot around the CPU heatsink and GPU heatsink so I'm not taken aback to find bloated HZs and MFZs there, eventually.... sometimes even bloated Fujitsu RE polymers (the yellow ones). As for 1500uF 16V HN, not saying much.... never even seen a 1500uF 16V KZJ go bad in those units unless the fans seized or they were reflowed in a ghetto manner. Those capacitors are just treated very well as they are either right by the fans or very close to the vents. IMO, 2005 HMs/HNs/HZs are still bad. I'd wager that modern HNs and HZs (2006+) are about the same as MCZs and MFZs in reliability. Modern HMs (2008+... some bad batches were still in circulation as of 2006 and 2007), maybe about the same as MBZ and WG, not quite as resistant to heat as FJ however.

Yes. Probably 5 to 7 computers of those P4SD motherboards had them around the CPU. Rest were with bulged MCZ.

(like not being able to tolerate heat well).

IMO, MCZ/MFZ and KZG/KZJ/KZV actually fail under the same amount of heat, but since KZG and lower grade ESR capacitors from NCC have major chemistry issues they tend to fail whether you use them or not so MCZ/MFZ will still outlast them, at length.

You are 100% correct here.
I've desoldered hundreds of the 1500 uF 16V FLs (which are all 10 mm dia.), and I think I slightly pulled the leads on only one or two when I used a crappy iron for a few boards. They didn't leak, though. Moreover, once I pushed back their leads in, they still appeared functional. But I didn't trust them, so they are used as audio coupling capacitors on my breadboard. Hey, they work great for that application . Occasionally, I also use them as filter caps on my breadboard.

Did you try the spark test on them? That might validate their functionality.

I don't know. 10 years is a bit low in my opinion. Maybe in a hot environment, that's okay. But considering I have CRT monitors that are 15-20 years old now, run rather warm, and yet have only 85C caps that are still fine is a lot more like what I would expect from a capacitor.

Agreed. I think 85C capacitors are very underrated... even 105C highly aqueous capacitors have a higher failure rate in the wrong conditions! 85C capacitors do not have "1/4th" the life that equivalent 105*C capacitors do. That's just marketing. Usually, 85C capacitors are formed of organic solvent for the electrolyte, such as Ethylene Glycol, Boric Acid, Sodium Borate, etc... these are very low in water content compared to the more aqueous capacitors so they dry up slowly by comparison. Not quite as good as the weak non-aqueous acids found in non-aqueous capacitors like GBL, DMF, and DMA, such as Nichicon PW and Chemi-con LXZ (GBL), but definitely not bad.

10 years? Try 20 years. That seems to be the time it takes for those old Nichicon PRs and PFs to leak from the bungs, and those old NCC LXFs. And I've seen 30 year old UCC SMs still doing fine! So I think it's safe to say that the elongated and very fast thermal cycles in those old automobile ECUs that the PRs and PFs tended to leak in had something to do with the decomposition of the bungs (the bungs harden as they cool down and soften as they heat up - not good).

CRT monitors... not trying to hijack this thread but I think one of my P792s is starting to fail. It just started automatically "degaussing" itself in a "shorter" manner, it seems, at random, during operation, at 1024x768 @ 60 Hz. I'd say it roughly has 20,000 hours on it and 10,000 power cycles, just estimating. That is lower than one of my other P792s (30,000 hours, 7,000 power cycles, just guessing) which hasn't had that problem... in a long time. I said a long time because one summer, ten years ago, it started doing that incessantly, so I put it aside for a while and the issues stopped. I think that it has to do with the thermal cycles wearing and tearing the tube, as I believe there is an issue with these CRTs that causes the focus line to break away from the neck of the tube over time. Not sure if that's what's happening with mine, though, because again, the other P792 started doing that ten years ago (twenty thousand hours and thousands of power cycles earlier) and hasn't done it since (it was doing it during a game of Half-Life at 1024x768 @ 85Hz). As far as I know, CRTs degauss to remove unwanted magnetism in part.

I think the capacitors are good as they both have Delta power supplies in them, made in 2001 (the 30,000 hour one has older datecodes). UCC 330uF 400V SMH on the input (4A bridge without a heatsink too and no airflow, and 35 size transformers). Rubycon YXF, MHE, YXA, ZL, and Elna RE2 on the output of flybacks (yes I think these monitors have APFC).

[momaka]

Re: Fire in the hole!

You mean 2200uF 6.3V HZs?

Nope.
Some (like very few) Xbox 360s had 6.3V 2200 uF Nichicon HNs rather than HZs. Very rare though, just like the ones with the green Sanyos.

As for 1500uF 16V HN, not saying much.... never even seen a 1500uF 16V KZJ go bad in those units unless the fans seized or they were reflowed in a ghetto manner. Those capacitors are just treated very well as they are either right by the fans or very close to the vents.

I noticed the same thing. I use the KZJs for the least important recapping tasks on my own hardware.

Did you try the spark test on them? That might validate their functionality.

Yes. Charged them up to 15V and shorted the terminals. Got even bigger spark than what I usually get with the Nichies and Rubys.

I also regularly spark-test my caps now, because most are a few years already. Before charging them directly, though, first I use a 10k resistor in series with the cap for 5 minutes to reform the oxide layer. Then leave them disconnected but charged for another 5-10 while doing other stuff. Reforms them fine IMO. Finally, I discharge and do the spark test, and put them in circuit.

CRT monitors... not trying to hijack this thread but I think one of my P792s is starting to fail. It just started automatically "degaussing" itself in a "shorter" manner, it seems, at random, during operation, at 1024x768 @ 60 Hz.

From research on the web, that's a 17" Trinitron monitor, correct?
If yes, does the image quickly shrink and come back while the monitor makes a "tik" or "zap" sound at the same time (like similar to when you have static on you and you touch something metal)?
If that's a yes also, then I think yours might be having the same problem as some of my Sony CRTs. I haven't been able to figure it out, but my D1626HT does it a lot more often now - like once an hour or more sometimes. My E540 just started doing it, but very rarely and only after a long day's of ON time. The only one that only did it once and never afterwards was my D1025HTX - that one is a tank. I used it in my rental place when the temperatures were 100F inside the apartment, as that was the only monitor that would run mildly warm. Everything else was "on-fire" hot .

(yes I think these monitors have APFC).

Nope, no APFC. Their PSUs are just a wide-range flyback topology. Primary cap charges to whatever the line voltage is (times SQRT of 2, of course).

[Wester547]

Re: Fire in the hole!

Nope.
Some (like very few) Xbox 360s had 6.3V 2200 uF Nichicon HNs rather than HZs. Very rare though, just like the ones with the green Sanyos.

IIRC they used 1500uF 16V Sanyo WFs on the VRM input. Never seen one blown in these units either, but they were used very rarely. Seen a few blown in massive volume on some motherboards, though. Otherwise they used 2700uF 6.3V MFZ on the VRM output in the Xenons, and MFZ should be equal to HZ.

I noticed the same thing. I use the KZJs for the least important recapping tasks on my own hardware.

I wouldn't ever use KZG or KZJ for anything. Just my opinion, because these capacitors will bulge on the shelf, still retaining their full leads. The 3300uF 6.3V KZGs, in particular, are the worst when it comes to that. I've seen tons of different Intel 845 and 850 series boards that were never used and all (either two or four) the 3300uF 6.3V KZGs in the VRM output, mixed with polymers, were badly bulged. Even the bad date code HMs (and sometimes, the Teapo SCs used in the VRM input as well as the SKs for the smaller capacitors) were still okay on the unused boards, but KZGs couldn't even survive in storage.

From research on the web, that's a 17" Trinitron monitor, correct?

It's a 17" flat CRT Trinitron ("Aperture" grille technology) monitor by Sony/Dell with a Delta PSU inside ("Sony" is silkscreened on the PCB but the transformers are Delta models).

If yes, does the image quickly shrink and come back while the monitor makes a "tik" or "zap" sound at the same time (like similar to when you have static on you and you touch something metal)?

No, the monitor just looks like it's trying to degauss (the image goes out of whack for a second as it readjusts itself but doesn't shrink at all), and makes a bit of a ticking noise, but it's the same ticking noise it makes when it finishes degaussing when you first turn it on.

If that's a yes also, then I think yours might be having the same problem as some of my Sony CRTs. I haven't been able to figure it out, but my D1626HT does it a lot more often now - like once an hour or more sometimes. My E540 just started doing it, but very rarely and only after a long day's of ON time. The only one that only did it once and never afterwards was my D1025HTX - that one is a tank. I used it in my rental place when the temperatures were 100F inside the apartment, as that was the only monitor that would run mildly warm. Everything else was "on-fire" hot .

My image doesn't shrink so I think it's another issue on my behalf. My other P792 stopped doing it ten years ago, as aforementioned, so I'm not sure what to make of it. However, my second P792 did it within 10 minutes of on time when it did do it. And both of my P792s run surprisingly cool.

Nope, no APFC. Their PSUs are just a wide-range flyback topology. Primary cap charges to whatever the line voltage is (times SQRT of 2, of course).

That's a relief. No worrying about extra APFC transistors blowing or coils burning, especially since these PSUs aren't the most efficient.

[momaka]

Re: Fire in the hole!

I wouldn't ever use KZG or KZJ for anything. Just my opinion, because these capacitors will bulge on the shelf, still retaining their full leads.

Well, the few KZJs I used were for more of an "experimental" fix (like Photonicinduction - in the name of science ).
IIRC, I put 1500 uF 16 V Rubycon MCZ and the same specs Chemicon KZJ (both from Xbox 360s) in a fairly-hot LCD monitor. The MCZ was overheated once during a reflow and bulged (but no pressure under the vent after the cap cooled). The KZJ... well it's a KZJ, let's see how it does with the heat
Both are in parallel, so we will see which one gives first.

The other place I used KZJs was in a crappy CWT-built PSU - did the entire main PS output (minus 5VSB and -12V rail) with them. Original caps were much smaller and Fuhjyuu anyways.

The 3300uF 6.3V KZGs, in particular, are the worst when it comes to that. I've seen tons of different Intel 845 and 850 series boards that were never used and all (either two or four) the 3300uF 6.3V KZGs in the VRM output, mixed with polymers, were badly bulged.

Yes, those are terrible, I agree. I have only one that did not bulge, but it's sitting in my "failed junk caps" bin rather than the "not-failed but junk caps" bin

No, the monitor just looks like it's trying to degauss (the image goes out of whack for a second as it readjusts itself but doesn't shrink at all), and makes a bit of a ticking noise, but it's the same ticking noise it makes when it finishes degaussing when you first turn it on.

If you don't hear any 100/120 Hz line hum when that happens, then it is not degaussing.

My image doesn't shrink so I think it's another issue on my behalf.

Well, my image doesn't always shrink. Sometimes it shakes, sometimes it scrolls, sometimes it completely disappears for a fraction of a second.

And both of my P792s run surprisingly cool.

Try them in a room temperature of 25C (~78F) or upwards.

[Wester547]

Re: Fire in the hole!

Yes, those are terrible, I agree. I have only one that did not bulge, but it's sitting in my "failed junk caps" bin rather than the "not-failed but junk caps" bin

I've seen several of those actually fail without bulging before (and never bulged to this day), but they were almost completely shorted, so can't call them good...

Well, my image doesn't always shrink. Sometimes it shakes, sometimes it scrolls, sometimes it completely disappears for a fraction of a second.

I don't hear the low frequency line hum, but the noise it makes is exactly the same as the noise it makes after the low frequency line hum, that's why I was perplexed. Yes, my image shakes. I guess it is the same problem. Again, though, happened to my other P792 ten years ago and all these years later, after all that usage, hasn't happened since. Seems that it is a very common issue with these monitors but that only some manage to survive it....

Try them in a room temperature of 25C (~78F) or upwards.

They don't even get hot (at least at the exhaust vents) at 30C room temperature.

But the issue doesn't happen at any one temperature. It's happened at 22*C to 26*C, so...

I wonder if it's a cap issue, if one has silently leaked like in this thread...

[momaka]

Re: Fire in the hole!

I wonder if it's a cap issue, if one has silently leaked like in this thread...

From what I was told, it is likely a cap issue... inside the flyback (i.e. you can't replace it without replacing the flyback). But the good news is it takes a while for it to break down fully. See this thread:
https://www.badcaps.net/forum/showth...hlight=D1626HT