Announcement

**Wester547** · 07-10-2017, 02:14 PM

Re: ASUS A8V-MX cap diagram and some experiments

So the bulged capacitor in question was FJ and not FJS?

I remember you first made mention of the failure over three years ago. That might be the slightest bulge I've ever seen a capacitor exhibit - a neophyte could easily miss that. It didn't vent enough to leak at the top, anyway. Capacitors with truly volatile electrolyte seem to vent either quite noticeably or with a vengeance when they go, both at the top and bottom (think Sacon and KZG/KZJ). Of course I'm not saying Panasonic FJ/FJS is impervious to sudden failure - premature failures aren't non-existent.

I noticed at the end you deemed it a good motherboard, and I wouldn't dissent to that. However, I'm still wondering if, after all these years, the stigma still persists on the forum that ASUS motherboards should be avoided at all costs due to the dreaded "sudden death syndrome".

Motherboards don't just quite working for no reason even when troubleshooting methods don't yield any results. Perhaps "ASDS" is limited to certain models. But the notion that all ASUS motherboards (made after a certain point) are more-or-less doomed to failure seems silly, nonsensical, unfounded, and comes across as slightly phobic and biased (I'm not applying that comment to anyone, just making a point). Of course I realize your thread isn't about that at all so I'll leave it at that.

Now let's hope that HZ doesn't decide to bulge anytime soon, since the FJ before it left such an inspiring status quo to live up to.

**stj** · 07-10-2017, 02:29 PM

Re: ASUS A8V-MX cap diagram and some experiments

enable the hidden options in the bios.

**pfrcom** · 07-10-2017, 07:40 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by Wester547 View Post

Motherboards don't just quite working for no reason

I disagree

https://www.badcaps.net/forum/showth...ghlight=p2-99b

**Wester547** · 07-10-2017, 07:52 PM

Re: ASUS A8V-MX cap diagram and some experiments

*quit working

I recall Topcat said in another thread that he's seen motherboards of that era fall victim to bad caps. Ironic, considering those boards had interchangeably used Chemi-con LXZ, Chemi-con TM (not TMZ / TMV / TMJ), Rubycon YXG, and Sanyo (Suncon) CG / DX, IIRC (in your case YXG), all good series from good brands AFAIK (but then again, unexpected failures can happen such as the exemplified case of FJs which dried up just sitting in storage). I don't remember the brands of the smaller caps as of the moment but I doubt that they would be causing your issues. Due to rubber bung degradation, any capacitor (without a hermetic seal) can eventually go out-of-spec after 15+ years. Still, the issues you were describing probably point to something else.

**momaka** · 07-12-2017, 04:25 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by Wester547 View Post

So the bulged capacitor in question was FJ and not FJS?

I remember you first made mention of the failure over three years ago.

Yup, FJ.
I probably had a memory lapse when I posted in that other thread as the capacitors being FJS. Or probably also the fact that I had pulled some FJS capacitors out of another ASUS motherboard for reuse.

Originally posted by Wester547 View Post

I noticed at the end you deemed it a good motherboard, and I wouldn't dissent to that. However, I'm still wondering if, after all these years, the stigma still persists on the forum that ASUS motherboards should be avoided at all costs due to the dreaded "sudden death syndrome".

Well, ASUS motherboards did seem to die a lot more "for no reason" than other brands, especially when considering they typically used good Japanese caps. But I would say this was more relevant only for the socket 462/939/AM2 and 478/775 boards. The general notion (or rumor, if you will) was that they always cut corners on ESD protective devices for the ports, which made the motherboards more susceptible to ESD damage, and thus the main cause behind "ASDS". Looking at this board, I can't say the thinking behind that notion is unfounded - after all, look at how many capacitors and other components are missing on the board.

On the other hand, I think what also needs to be considered is if the board in question was made for an OEM PC maker, such as HP, eMachines, Gateway... etc. I say this, because it seems that a lot of cheaper OEM PC makers tend to instruct ASUS to cut corners in order to build cheaper boards, hence the possible lapse in quality there. You can see the same holds true for the MSI MS-7191 and MS-7184 motherboards I posted.

Also worth mentioning is that faulty chipsets could have swayed some people to believe that ASUS ASDS is worse than other boards - after all, the M2N variants were very popular. Having bumpgate-affected chipsets surely did not help their reputation. Meanwhile, the ASUS P5Q motherboards were pretty rock-solid.

Either way, I never really had any particular bias against any one board manufacturer (not even ECS!

) It seems that they all had their strengths and weaknesses.

Originally posted by Wester547 View Post

Now let's hope that HZ doesn't decide to bulge anytime soon, since the FJ before it left such an inspiring status quo to live up to.

That HZ was actually bulged before - it came from an Xbox 360 that someone tried to reflow on the wrong profile and half of the Nichicon caps on it were slightly domed due to the high heat. But what's interesting is that after those caps cooled down, I could push their tops back down with minimal force, implying there was no high pressure inside. Hence, the bulging was likely due to contents expanding under high temperature rather than electrolyte failure / Hydrogen formation. All of them measured fine on ESR and capacitance back then (2012) and still did a few years afterwards when I tested them prior to reforming about a year and half ago. Also not the first time I used bulged Nichicons from that motherboard for a repair. So far, it seems all of them have been holding up fine.

Lol, now where's that "playing with fire" smiley?

I guess I will use this one instead:

Originally posted by stj View Post

enable the hidden options in the bios.

How?

Some online research points to BIOS reflashing (which I am not a big fan of, unless I need it to resolve a problem), while another website suggested pressing "Ctrl" + "F1" or "Alt" + "F1" or "Shift" + "F1" while in BIOS.

Any clues?

Originally posted by pfrcom View Post

I disagree

https://www.badcaps.net/forum/showth...ghlight=p2-99b

Interesting failure there. Not sure how I misted that thread.

Looks like it might be a bad CMOS chip (not to be confused with a bad BIOS chip!)

Also, old boards like that almost never failed due to BGA-related issues, so a reflow should have never been done.

But I suppose it's all a moot point now.

**Wester547** · 07-12-2017, 06:00 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by momaka View Post

Well, ASUS motherboards did seem to die a lot more "for no reason" than other brands, especially when considering they typically used good Japanese caps. But I would say this was more relevant only for the socket 462/939/AM2 and 478/775 boards. The general notion (or rumor, if you will) was that they always cut corners on ESD protective devices for the ports, which made the motherboards more susceptible to ESD damage, and thus the main cause behind "ASDS". Looking at this board, I can't say the thinking behind that notion is unfounded - after all, look at how many capacitors and other components are missing on the board.

Intel allowed many 865-series Foxconn motherboards to die a similar death with the dreaded ICH5 southbridge which lacked ESD protection.

On the other hand, I think what also needs to be considered is if the board in question was made for an OEM PC maker, such as HP, eMachines, Gateway... etc. I say this, because it seems that a lot of cheaper OEM PC makers tend to instruct ASUS to cut corners in order to build cheaper boards, hence the possible lapse in quality there. You can see the same holds true for the MSI MS-7191 and MS-7184 motherboards I posted.

Yes, particularly eMachines/Gateway, which used the cheapest components possible when building to contract.

Also worth mentioning is that faulty chipsets could have swayed some people to believe that ASUS ASDS is worse than other boards - after all, the M2N variants were very popular. Having bumpgate-affected chipsets surely did not help their reputation. Meanwhile, the ASUS P5Q motherboards were pretty rock-solid.

MSI and ECS also made several boards whose northbridges used chipsets that were prone to "bumpgate" failure.

That HZ was actually bulged before - it came from an Xbox 360 that someone tried to reflow on the wrong profile and half of the Nichicon caps on it were slightly domed due to the high heat. But what's interesting is that after those caps cooled down, I could push their tops back down with minimal force, implying there was no high pressure inside. Hence, the bulging was likely due to contents expanding under high temperature rather than electrolyte failure / Hydrogen formation. All of them measured fine on ESR and capacitance back then (2012) and still did a few years afterwards when I tested them prior to reforming about a year and half ago. Also not the first time I used bulged Nichicons from that motherboard for a repair. So far, it seems all of them have been holding up fine.

What an irony that the one which bulged right away soldered (yes, pun intended

) on, but those that bulged later on in storage did not survive. The depolarizers were weak with those ones.

**momaka** · 07-12-2017, 06:10 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by Wester547 View Post

What an irony that the one which bulged right away soldered (yes, pun intended

) on, but those that bulged later on in storage did not survive. The depolarizers were weak with those ones.

Perhaps I should clarify: the HZs that I had bulge in storage (mind you, only 3 or 4) had a 2005 date code (week 40-something, IIRC), whereas the overheating-induced bulged HZ above has a 2006 date code. So I think that's what makes the difference here.

In my stash of caps, I also have one Rubycon MCZ 16V, 1500 uF (again, out of an Xbox 360) that I accidentally put in a circuit in reverse polarity. The cap bulged. But due to low currents in the circuit where I had it, it did not vent at all - not until I punctured it. Then, after releasing the excess gas, I pushed the vent back down. The cap still retained its specs even a year after that. Then I decided to hot-glue the vent to seal it a little better. Did that maybe about a year ago. Guess what? - That Rubycon MCZ cap is still in spec in terms of ESR and capacitance.

And it is not fooling my meter either: I can charge it to 15V and discharge it rapidly on a metal surface with considerable pops and sparks for its capacity (when compared to an equivalent 1500 uF cap at the same voltage).

Now that's a cap that deserves my respect.

**Wester547** · 07-12-2017, 06:42 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by momaka View Post

Perhaps I should clarify: the HZs that I had bulge in storage (mind you, only 3 or 4) had a 2005 date code (week 40-something, IIRC), whereas the overheating-induced bulged HZ above has a 2006 date code. So I think that's what makes the difference here.

Not sure about that... this eBay listing shows some vented and leaking HZ. Although it's very hard to tell from the picture (I could be wrong, one may need X-RAY vision

), I think early 2006 datecodes are visible on the HZ and HN if you look very carefully. The issue may be that the oxidizers in the electrolyte are not always 100% successful in absorbing excessive hydrogen or precluding water-driven corrosion, which is why some may arbitrarily bloat a la KZG and KZJ.

In my stash of caps, I also have one Rubycon MCZ 16V, 1500 uF (again, out of an Xbox 360) that I accidentally put in a circuit in reverse polarity. The cap bulged. But due to low currents in the circuit where I had it, it did not vent at all - not until I punctured it. Then, after releasing the excess gas, I pushed the vent back down. The cap still retained its specs even a year after that. Then I decided to hot-glue the vent to seal it a little better. Did that maybe about a year ago. Guess what? - That Rubycon MCZ cap is still in spec in terms of ESR and capacitance.

And it is not fooling my meter either: I can charge it to 15V and discharge it rapidly on a metal surface with considerable pops and sparks for its capacity (when compared to an equivalent 1500 uF cap at the same voltage).

Now that's a cap that deserves my respect.

I guess this is why VRM output failures are more commonplace than VRM input failures - the low voltage, very high surge currents and high ripple current applications seem to be killer. The internal heating brought about by high amounts of ripple current is more likely to cause an electrolytic to vent than external heat such as high ambient temperatures, especially considering both reverse polarization and very high ripple current loads form the cathode foil (the latter at a slower rate) which generates hydrogen gas. That said, higher ambient temperatures will still accelerate all such failures.

MCZ, despite their sensitivity to heat, are good capacitors. One should not underestimate them.

Perhaps my memory is faltering on me, but I recall you said five years ago that one of the 1500uF 16V MCZ you reverse polarized had formed permanent gas inside - IE, it was hard as a rock and you couldn't push the top back down.

**momaka** · 07-15-2017, 02:56 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by Wester547 View Post

Not sure about that... this eBay listing shows some vented and leaking HZ. Although it's very hard to tell from the picture (I could be wrong, one may need X-RAY vision

), I think early 2006 datecodes are visible on the HZ and HN if you look very carefully. The issue may be that the oxidizers in the electrolyte are not always 100% successful in absorbing excessive hydrogen or precluding water-driven corrosion, which is why some may arbitrarily bloat a la KZG and KZJ.

Not buying it. :\ (Pun intended, too.)

Nichicon HN and HZ caps with 2006 and later date codes are definitely good and you won't see them fail on their own. You'd be hard-pressed to find a board with so many bloated like the above one. Most likely the caps on the above board were overheated due to someone trying the towel "trick" or some equally stupid idea (oven bake

). And given that the Xbox 360 board above comes without anything else, I think we can be certain that some idiot already tried to reflow these and failed.

Originally posted by Wester547 View Post

I guess this is why VRM output failures are more commonplace than VRM input failures - the low voltage, very high surge currents and high ripple current applications seem to be killer.

Well in my case, the MCZ vented due to reverse polarity. The only thing that saved it from venting violently was a 10-Ohm resistor that limited the current quite a bit. After running my circuit for a few moments, I noticed that the cap was very hot and just started to bulge. I quickly disconnected power, but the cap continued to bulge until it vented.

Originally posted by Wester547 View Post

MCZ, despite their sensitivity to heat, are good capacitors. One should not underestimate them.

Perhaps my memory is faltering on me, but I recall you said five years ago that one of the 1500uF 16V MCZ you reverse polarized had formed permanent gas inside - IE, it was hard as a rock and you couldn't push the top back down.

That was a blue Panasonic GP cap from a TV board. Popped that one in the same circuit that popped the MCZ. After sitting many years on my shelf, that Panasonic cap has finally lost the pressure inside, and if I remember correctly when I last checked it, it was reading okay on capacitance and ESR again.

**Wester547** · 07-15-2017, 04:23 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by momaka View Post

Not buying it. :\ (Pun intended, too.)

Nichicon HN and HZ caps with 2006 and later date codes are definitely good and you won't see them fail on their own. You'd be hard-pressed to find a board with so many bloated like the above one. Most likely the caps on the above board were overheated due to someone trying the towel "trick" or some equally stupid idea (oven bake

). And given that the Xbox 360 board above comes without anything else, I think we can be certain that some idiot already tried to reflow these and failed.

Well, only the HZ are bulging on that board. The rest of the capacitors, HN included, appear fine. Boards that are "oven" reflowed (as a whole) for too long usually show more signs of thermal duress, such as severely darkened electrolyte and discoloration on the SMD FET pads (from orange to brown). And there have been sightings of prematurely failed HM made after 2005:

https://www.badcaps.net/forum/showpo...69&postcount=4
https://www.badcaps.net/forum/showpo...9&postcount=20

Of course, only a handful of failures don't mean much.

Well in my case, the MCZ vented due to reverse polarity. The only thing that saved it from venting violently was a 10-Ohm resistor that limited the current quite a bit. After running my circuit for a few moments, I noticed that the cap was very hot and just started to bulge. I quickly disconnected power, but the cap continued to bulge until it vented.

In the duration of the endurance tests at the factory, so as to make certain the peak voltage does not exceed the rated voltage, capacitor manufacturers tend to use 1000Ω resistors in series with the capacitors. They also use such resistors to limit the current during reverse polarity tests (to test the functionality of the pressure-relief vent). The hydrogen gas that escapes quickly enough to rupture the can does get very hot, around 100ºC.

Here is the post you made about the 16V MCZ venting. You noted it was hard as a rock. Or was that another one?

That was a blue Panasonic GP cap from a TV board. Popped that one in the same circuit that popped the MCZ. After sitting many years on my shelf, that Panasonic cap has finally lost the pressure inside, and if I remember correctly when I last checked it, it was reading okay on capacitance and ESR again.

So it was the Panasonic M 85ºC series? Impressive that there wasn't permanent pressure build-up inside. If it had been a Teapo or CapXon, OTOH...

**ChaosLegionnaire** · 08-10-2017, 01:29 AM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by momaka View Post

The other notable thing is that this motherboard does have an option to change the Base/Host Clock. However, doing any kind of CPU over-clocking in this way (which is actually how it's done with socket 939 AMDs) makes the motherboard completely hang on POST. And the only way to fix the hanging is to remove the CMOS battery and reset CMOS. This happens even when the RAM is set to run at a slower clock and the HT link is set to a lower multiplier. I even tried setting the PCI-E clock to a “Fixed” frequency (there is an option for that as well). But still, the motherboard refuses to over-clock the CPU even by a single MHz.

Thus, this is clearly not a good motherboard for over-clocking.

we probably already mentioned this in a pm, but i'd like to state it in public for the benefit of other members and newer members.

the reason why overclocking doesnt work is because its a rev1 board from the photos. only asus' rev2 of those via chipsets had working pci/agp/pci-e locks. so even tho the pci-e clock is set to fixed in the bios, it doesnt work... if u run the pci-e clock signal pin through an oscilloscope, i bet it wont show the pci-e clock as running at 100mhz.

so if u want an asus via s939 board for overclocking, make sure its rev2 so the frequency lock works and u can overclock to your heart's satisfaction. *starts playing benny benassi's satisfaction song*

**momaka** · 08-10-2017, 02:14 PM

Re: ASUS A8V-MX cap diagram and some experiments

Originally posted by ChaosLegionnaire View Post

we probably already mentioned this in a pm, but i'd like to state it in public for the benefit of other members and newer members.

the reason why overclocking doesnt work is because its a rev1 board from the photos. only asus' rev2 of those via chipsets had working pci/agp/pci-e locks. so even tho the pci-e clock is set to fixed in the bios, it doesnt work... if u run the pci-e clock signal pin through an oscilloscope, i bet it wont show the pci-e clock as running at 100mhz.

Interesting.

I don't recall reading that in your PMs, so definitely good to have this info here.

I suppose the only thing I can do with that mobo is look for a more powerful CPU... which shouldn't be a problem. I now have two spare Athlon 64 3500+ chips and one 3700+. If voltage control works with CrystalCPUID, I may be able to get some really cool temps with these CPUs, if nothing else.

Originally posted by ChaosLegionnaire View Post

so if u want an asus via s939 board for overclocking, make sure its rev2 so the frequency lock works and u can overclock to your heart's satisfaction. *starts playing benny benassi's satisfaction song*

Ha! Been a while (maybe a year) since I heard that one. Time to play it on my headphones

Announcement

ASUS A8V-MX cap diagram and some experiments

ASUS A8V-MX cap diagram and some experiments

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment