Vmem mod for Iwill Rdram m/b....P4R533

[Ace-a-Rue]

WARNING! YOU CAN FRY YOUR BOARD IF YOU DO THIS INCORRECTLY OR YOU SLIP WHILE MEASURING THE VMEM VOLTAGE!!

YOU ARE WARNED!!


Ok guys, I am happy to announce that I found the Vmem regulator (HIP6521) for the Iwill P4R533-N!. It is located just below the North/South rimm bank (furthest to the right side of the board) as the board is situated in a vertical ATX case.

I used pin 3 on the dimple side of the regulator. I used a 47K Radio Shack pot with the SMD Grabber.http://www.elexp.com/clp_5243.htm I set mine up where I would turn the pot counterclockwise to lower the resistance which in turn raises the Vmem voltage. You can set the pot up either way for your own preference.

HIP6521 PIC

Here comes the tricky part; measuring the Vmem voltage. There are 2 small Mosfets located in the upper right hand (Northeast corner) of the board. You use the second (South) Mosfet and measure with your VM using the most inboard leg adjacent to the rimm and capacitors. This is where you have to be extremely careful to not slip with the probe (or angle it) by touching the middle snub leg (barely visible) and shorting the Mosfet.

MOSFET PIC

I set mine for 2.92V and the Vcore to 1.93V. The funny thing about my board is that I cannot cold post anything above 148 FSB at 4X setting for the memory in the bios. I can cold post at 156 (max setting) using 3X, but not 4X. To get to 156 FSB, I boot at 148 4X and when I am in Windows, I use CPUCOOL to raise the FSB to 156. It is really slick since I could never get to 156 at 4X. My Sandra Memory is 3950 at 156 FSB. I am using 2-256mb sticks of Kingston PC1066, 32ns.

I ran Sandra burn in for 60+ times using the cpu and memory and then I ran 3d2001se for one hour continous looping with no kickouts, bsod or freezes.

 

[AUDSCOTT]

Do I smell something burning???

Nice work Ace!!

You are the modder extrodinaire!

 

[Ace-a-Rue]

Thanks my friend! It is really fun when you happen to discover something like this! Of course, without the posts that revealed the mods on the Abit boards, I wouldn't have figured it out. And, thanks to Iwill engineers in keeping communally in voltage regulators. hehehe

 

[Ace-a-Rue]

Just wanted to update on how to measure the Vmem voltage at the Mosfet.

I thought about how tricky it could be by angling the VM lead in the tight spaces and then slipping possibly shorting the Mosfet and the frying the board.

Here is a better way to check the voltage w/o moving any hardware out of the way (after the initial installation). Simply use another SMD grabber with a wire attached. Attach the grabber to that inboard leg and then run the wire where you can easily access it with the VM lead. When finished, put either an electrical twist cap or electrical tape to protect from shorting. This way you can easily access it if you have to make any adjustments later on.

I want to thank DOC who put me on to the SMD Grabber trick back when we were trying to pump the AK31.

 

[Ace-a-Rue]

I discovered a tiny kink in doing this Vmem mod. It is not to alarming. I had to change out the 47K pot to a RS 10K pot. The 47K pot didn't provide enough of a fine adjustment in hundreds of voltage points. It would jump in large increments of voltage as you got to the lower resistance value below 10K (some where between 6-7K will give you 2.92). the 10K allows a more refined adjustment.

The board is still functioning very nicely. Very stable for Windows operation using XP.

 

[Ace-a-Rue]

top 5! hehehe

 

[Ace-a-Rue]

Just to give you a link to my 'How To'http://www.thenakedreview.com/index.php?p=showarticle&id=80 at TheNakedReview