[RESOLVED] Abit KA7-100 Raid burning up capacitors?!?!

[XateEsix]

This is a copy of a letter I recently sent to Abit.

Hello,

I have an Abit KA7-100 RAID motherboard with JPCon TM series 1500uF and 2200uF 6.3v 105ºC capacitors which have failed. I need 6 2200uF and 15 1500uF to repair this apparently failed product. If there have been revisions or upgrades to this product due to these failures could you please inform me of what you recommend for replacement.

Thank you


If anyone has heard of this please email me.

Thanks
XateEsix

[Darren Wilson]

Ruslan would be the person to ask on here regarding these types of query. Maybe a PM to him notifying him of this thread will help?

------------------
I am my own Judge , Jury & Executioner. Although I am availabel for Weddings, Birthdays, Bahmitzvahs, etc,etc......

[Antimatter]

Generally capicitor failure is caused by to much heat from the voltage regulators due to heat being transferred through the mobo from hot transistors that lack proper heatsinks. I've noticed this in several boards where the voltage regulator transitors are attached directly to the mainboard close to capacitors.
Lately I've been gluing heatsinks to the transistors to reduce the heat and so far it seems to be doing a good job.
If I'm wrong about this Ruslan is most likely able to correct me. http://forums.windrivers.com/cgi-bin/forum1/noncgi/smile.gif

[This message has been edited by Antimatter (edited October 11, 2001).]

[XateEsix]

Thank you both for responding. I sent a message to Ruslan as well. It is the capacitors on the voltage regulation circuit right next to the cpu. The heat sinks on the power transitors is probably a good idea. I am also curious if there is a better brand of capacitor to use than the JPCon brand that are on it from the factory. Its also possible the transistors are shot but I doubt it. The system went intermitantly with the final boot attempts initializing then failing. It really wants to run.

John

[Ruslan]

I have already answerred on that PM,but couldn't find this topic... http://forums.windrivers.com/cgi-bin/forum1/noncgi/biggrin.gif LoL!

Agree with Antimatter http://forums.windrivers.com/cgi-bin/forum1/noncgi/cwmsmilies/cwm16.gif
Those capacitors are located too close to heavily loaded power MOSFET transistors... Exactly same problem is now being discussed at IXBT forums here: http://forum.ixbt.com/ubb/Forum9/HTML/027610.html (Who want to learn Russian - come on! http://forums.windrivers.com/cgi-bin/forum1/noncgi/cwmsmilies/cwm16.gif ) So, that problem is international...

Capacitors are also burning up due to overloading - usual average temperature of those capacitors is about 50-55C... And depends on CPU's load...
And those caps cann't handle with that high current (especially working at high frequency - about 150-200 KHz). Obvious engineer's mistake...

One of recommended solutions is using instead of original capacitors similar capacitors, but designed for higher voltage (about 16V),but with same as original capacity.
Another solution is using capacitors, designed for same voltage,but with bigger capacity (about 2500 - 4700 Mkf).
Anyways,using capacitors with larger size IMHO, can do the trick. You see,capacitors with larger size can dissipate more power (these capacitors works like reactive resistors).

I've replaced those capacitors quite a few times in ACORP and ABIT motherboards and have had good luck so far. I used capacitors with higher voltage. Jamicon (Kaimei) capacitors are considered to be good enough for that purpose...
Anyway, check also power MOSFET transistors using multimeter - they are at least should not be shorted... Usually these transistors remains good, but who knows...
Good luck!

[Ruslan]

Yeah, and better to use capacitors,designed for higher max. working temperature (105C instead of usual 85C)...