9800GX2 voltmods thread

[NightRaven]

You got a 1.5 card? Is it leadtek? I know most are able to use only 1.35 including myself.

nah nah i was benching 1.35V too. read above posts

but the card would not blank out till 1.5V at IDLE

however, above 1.35V, it would blank out mostly at GT2 for 06. so i figured it was OCP

[rintamarotta]

So this is SLI for non-SLI motherboards , right ? Nvidias way of getting into the Intel powered motherboard sector ? Perfect partner for my Maximus Formula .

Its single card that have internal sli, so yes it will work fine on any motherboard that have PCI-Express 16x or better.

[T_M]

GX2 #2

This time all vmods and readpoint are on the backsides of the boards.
Also added some solid caps to vcore.









Next ill add some inductors because the vcore OCP is still a problem.

[RLM]

thanks for posting that T_M. off to get some VRs today. Waterblock I''ve been told is being sent out today as well.

I'm guessing on future cards we will finally get control over the voltages without mods.(Not as much fun I know).



RLM

[sidewindercomp]

GX2 #2

This time all vmods and readpoint are on the backsides of the boards.
Also added some solid caps to vcore.

That is a thing of beauty.

[RLM]

Edited by RLM.

RLM

[RLM]

Thanks to you guys for posting your pics, really helpful.

Mods gone ahead just fine. Followed a bit of Vince's and a bit of T_M's. Didn't even need to take the card apart.

Just waiting for the waterblock now. Hopefully today.....

Cheers

RLM

[Deanzo]

Hey guys, a little help please.

First off, I'm very new to doing mods myself. I've only pick up an iron a few times. But I'm trying

I'm doing the vgpu mods for the GX2 and I just can't get one side to work, and am looking for ideas to try.

This side works just fine and was easy to do:


But this side does not want to play nice at all. I also tried the vcore point in red in post 20 of this thread with no luck.
I'm open to any ideas that may help.

[k|ngp|n]

Hey deanzo, I dont see a vcore mod done on the bad side?
Am I blind? Looks like you did the mem mod on one side and the vcore mod on the other...

[k|ngp|n]

I marked in green where the other vcore mod should be that I am not seeing. I use resistors tied in with same ciruit as the solder pads that everyone uses.

[T_M]

Yep spot on Vince, thats not the vcore mod you've done Deanzo, you have actually connected the vMEM read point to ground via the VR

[Deanzo]

Hey deanzo, I dont see a vcore mod done on the bad side?
Am I blind? Looks like you did the mem mod on one side and the vcore mod on the other...

Yep spot on Vince, thats not the vcore mod you've done Deanzo, you have actually connected the vMEM read point to ground via the VR

Thanks alot guys I see where I've gone wrong now.



When I was looking at T_M's picture, I thought the vr are went to the point I've just marked green. But in fact it needs to go to one of the two points I've marked in red.


Hehe, you could say I got my wires crossed looking at the picture

[T_M]

hehe, at least its resolved easily

[Deanzo]

After somehow killing the vr

I've changed that and soldered the wire to the point Vince pointed out, and it all works now .



Thanks again Vince & T_M

[T_M]

Hopefully this will help with OCP limits on LN2:

[T_M]

Nope... didnt help at all

[Alexxx650]

What disconnect OCP? The Card can work under 1.355, but anymore put. 1.375 brings about Black screen.....
With has done so


All work, but more 1.355 not to expose. That else possible to do?

[Alexxx650]

http://service.futuremark.com/compare?3dm05=4335100

Cold water, pci-e@125, vGPU 1.360, Caps mod

[RLM]

Bit early to tell, but I think the IC resistor on my top card (The one Vince used) is failing. My voltages have been fluctuating quite a bit.

Set initially around 1.2v, throughout the day I'd notice my voltage increasing , sometimes to the point that the OVP would kick in.

if I run vantage this would happen almost immediately 1.2v straight up to 1.34v. Sometimes the reverse would happen and the voltage would drop down to 1.1v.

I've re done the solder a few times now, and I'm now on my 3rd Variable resistor, but my conclusion is that the VR's are fine. So after no joy, I've tried taking the voltmod off, and fortunately no fluctuation.

Anyway I've now moved my solder point from the IC resistor to TM's point and fingers crossed this seems to be working just fine. Steady at 1.21v.

As I say I can only presume that the IC resistor is failing, but I really don't know. Anyone here, have any similar experiences or have any info on this?

Cheers

RLM