KX7 Vmods -> Dead Mobos!?

[Goldlocke]

Hi there! My first post here at XTremesystems. I followed OPP, MrIcee and DDTung over from Icrontic.

Though I am a newbie here, I am not in overclocking. In the last 3 years I´ve successfully voltmodded every ABIT board I owned(KT7/KT7A/KR7/KX7).

BUT: I wanted to discuss those voltmods and their issues with you. I never hesitated to do all 3 major voltmods to my ABIT boards the 2nd day I got them. I considered them absolutely safe when done carefully (which I had). But there where 2 dead ABITs which made me think.

It had been 1 KR7A-Raid and 1 KX7-333Raid. All 2 boards died the same way!
I had run these boards successfully with all 3 voltmods (VCore, Vmem, Vio) for months without trouble. Then suddenly the next day they would not boot anymore!? I unsoldered the mods - tested - changed RAMs, CPU, etc. They were dead from one day to the next!
Fortunately I could RMA them which is much more complicated here in Austria then in the US.

Hmmmm, so I wonder what was the cause for that:

1. some unusual volt-spike?
2. some faulty voltmod?
3. some bad CWT 420 PFC PSU? (=Antec 420).

What do u think? I used following max. voltages:

VCore ... 2.33V
Vmem ... 3.23V
Vio ... 3.77V

[DDTUNG]

Welcome to Xtreme!

To have used the voltages you mentioned for Vcore, Vmem,and Vio you must have been running some extreme overclocks. Those voltages were way over spec for the components on the boards, and I am not surprised if some of them failed, leading to the two dead boards that you mentioned. On the KR7 and the KX7 I have not found it necessary to use the Vio mod, which probably saved the day for me.;)

DDTUNG:cool:

[Goldlocke]

thx, DD for you're quick reply.

I have to add that the voltages I mentioned where the maximum. I usually used:

VCore ... 1.97V
Vmem ... 3.23V
Vio ... 3.70V

to Vio: It helped me sustain a better DDR-RAM overclock on my GF2 GTS. artifacts were reduced by higher Vio!

[DDTUNG]

I understand your point about Vio, but feel that it is better to try to Vmod the display card directly, or else mod the Vagp. That way you have better control and do not overspec as many other components.:)

DDTUNG:cool:

[Goldlocke]

So, DDTung, you think that the raised Vio was the cause for both deaths? :)
Not the Vmem in combination with a fried AGP controller?

MrIcee, OPP? What's your opinion here?

[MrIcee]

The VMem has only proven to ruin AGP controllers if you spike or inadvertantly prolong in excess of 3.4v DDR voltage. As far as the Vio is concerned..I concur with DDTUNG that volt modding the vid card gives much better results than modding the VIO voltage. You must remember that Abit already runs VIO much higher...than the vast majority of board makers out there.

I will agree with DDTUNG also...that by our volt modding of the boards...we ARE essentially running these boards out of specification...and running them for an extended period of time can lead to their demise as far as components are concerned. For those of us on the cutting edge its not an issue, as we turn over our boards on a 3-6 month cycle depending on the release of a next generation part, or buy from a supplier who will RMA the unit and send us a new replacement. You must remember, this sector of the hobby we participate in comes with self induced pitfalls as far as equipment is concerned, its an accepted risk we accept to have the ultimate performance for brief periods of time:)

Randi:D

[Goldlocke]

Ok then. Thx Randi and DD!

Randi, you're absolute right about "our" attitude towards hardware ;)

In the future I'll leave the Vio mod alone then!