Recapping of Asus P5B

[omega]

Hi,
I would like to report my experience with thorough recapping of one of about twenty motherboards Asus P5B (aka P5B vanilla) which I have been using in machines for my present work.
Over the last years, I have been recapping few (unfamous) 820 uF 6.3 V caps, which had been getting bulged, with (almost) constant success.
So encouraged by my own experience and by readings of a number of threads here in BADCAPS, I realised that the TK cross semi-cut caps (Figure 1) had to be replaced, and I done it in a complete fashion with 19 original Panasonic FR series identical value caps (Figure 2).
However, most of them simply did not stand the intense working conditions of my research work and as you can also see in Figure 3, one also get bulged as short as 66 hours service (these mobos have been generally working 24/7 for years with minor issues), where some others have been blowing on bottom.
So I have been getting somewhat disappointed by this last experience and I keep wondering where I have failed. Any explanations/suggestions of yours?
By the way, what kind of caps are those in Figure 4? They seem to be hard-to-find ones.
Thanks for attentions

[goodpsusearch]

Re: Recapping of Asus P5B

ASUS marks the polarity of the caps opposite than the other board makers.

WHITE is +, NOT GROUND.

As clearly seen in the pics you uploaded you have soldered the cap in reverse polarity..

[omega]

Re: Recapping of Asus P5B

dear goodpsusearch,
you are definitely right, if was a mistake of mine, thanks for the observation!
Though it may happen sometimes (it's my first time in years, I'm getting old), it is unbelievable that I did not notice myself. Shame on me! Anyway, I still don't explain why some others (correctly soldered) have got their bottom "stopper" moved in so short a time.
Any comment as regards the caps in Fig. 4? Are they OS-CON or simple through hole electrolytic? I still remember that in one out of 20 mobos I examined so far, I noticed this zone capped by simple 680uF, 4V caps under CPU heatsink. Of course, most of them have been bulging!
Thanks again

[linuxguru]

Re: Recapping of Asus P5B

At first glance, they looked like polymers of either Nichicon or some lesser-known off-brand. On second glance, it looks like it could be Panasonic FL (electrolytic) in clear lacquer finish. If so, it is a 2000-hr series and perhaps not as durable as polymers.

[ChaosLegionnaire]

Re: Recapping of Asus P5B

i noticed cap silkscreens printed on mobo pcbs all have the cap polarity oriented in one way for horizontal (left-right) hole spacing alignment and all in another but same way for vertical (up-down) hole spacing alignment. i dont understand why its possible to get the polarity wrong when they all have to face the same way. please pay more attention next time when soldering caps. use the orientation of the other caps on the board as a guide.

I noticed this zone capped by simple 680uF, 4V caps under CPU heatsink.

those are the cpu vrm output filtering caps. they have to be low profile and not more than 9mm tall as specified by intel in their platform design guidelines so as to not interfere with the lga775 heatsink (processor keepout zone). they also have to be placed as close as possible to the cpu (or load) for the fastest response to transient loads. the disadvantage is that the heat from the heatsink and/or the cpu has a high tendency to kill the caps.

u can replace them with sanyo oscon sepc but they are costly and cost about US$1.50 each. there are 9 of them so u'll end up spending at least US$13.50 w/o shipping on them and they are not immune to heat either. u'll also need to get a better heatsink like a low-profile tower cooler which doesnt regurgitate hot air over the caps and killing them.