Droop mod question - Hey, macci!

[Malves]

I did mine a couple of weeks ago and it's working fine. But what about removing the resistor from the board and soldering the trimmer there. Is it possible? Would it work even better? If so what resistance should I start with on the trimmer?

[Holst]

I dont see why you cant do it that way.

If you take the resistor off the board and meaure it then you need a pot thats a tad bigger so you can set it the same as the resistor... then tweak it either way.

[Malves]

Humm...but do you think it would be better?

[Holst]

better... probalby not..

just a different way of doing the same mod.

[Malves]

I checked the Vcore with a MM and found out that setting 1.525v in the BIOS, w/ the trimmer set @ ~34k, it reads 1.58v. What's interesting here is when I run P95 (large FTT's and small FTT's), putting loand on the CPU, my MM reads 1.61v.
I have no idea why, but I'm sure not complaining.

I was using Asus probe to check my voltages, and it was reading close to this. But AP shows more and bigger variations than the MM does.

This still with the same mod.

[macci]

Its not a resistor its a capacitor. And there is usually a reason why they solder those on mobos

If its working nicely I see no reason why you should remove that cap.

[Malves]

Thanx, bro. I'll leave that thing alone. It sure is working nice.

[Holst]

Oh... I couldnt see it was a cap..

If it is a cap then you shoudlnt remove it.

Do you have some more info on this mod... I assume the cap is in paralell with another resistor or set of resistors.

[PIRATA!]

I checked the Vcore with a MM and found out that setting 1.525v in the BIOS, w/ the trimmer set @ ~34k, it reads 1.58v. What's interesting here is when I run P95 (large FTT's and small FTT's), putting loand on the CPU, my MM reads 1.61v.
I have no idea why, but I'm sure not complaining.

I was using Asus probe to check my voltages, and it was reading close to this. But AP shows more and bigger variations than the MM does.

This still with the same mod.

I have the same thing with my trimmer set to 35KOhm.
I still have some low hits on VCore wile in full load testing with Prime95.
I am thinking of getting down the trimmer to see if those low hits on VCore goes away.
I would leave the mod like this, but I can get stable without a perfectly stable VCore.
Any suggestions??

[eulercdc]

There's any spot on the mobo that i can read the Vcore???
Can i trust the Vcore from the Bios???
I've made the drop mod here with the trim set to 28K and when i set the BIos Vcore to 1,6V it's read 1,66V... Is that overvoltage normal???

Thanks!!

[PIRATA!]

I have the same problem.
I just can't find out the correct point where to check the VCore with the multimeter.
Please tell us!!

[Aronier]

Hi,

i have do malves droop mod on my P4C800 Dlx. Rev. 2.0 w/ the trimmer set @ ~28k. It sure is working fine but i checked the Vcore with a MM and found out that setting 1.525v in the BIOS it reads 1.62v.

BIOS = MM reads
1,55v = 1,64v
1,575 = 1,67v
1,60v = 1,69v
But if i setting 1,625v or more my MM reads 1,82v and more!
I don`t have any vcore between 1,7v-1,8v, what`s wrong here?

When I run P95 (large FTT's and small FTT's), putting loand on the CPU, my MM reads 0,03 higher voltages?

I can turn the trimmer higher or lower but the reaction is 0,03v more or less?

What`s here wrong, please help me.

[PIRATA!]

You have to just hope you have not made a bad soldering because in that case you could have damaged your mobo.
I had a similar problem with my last P4C800-Delux (Not E) Rev.2.00 that I have changed in RMA to my store because I losted Voltages between 1.5V and 1.6V soldering on the alternative blue spots instead that directly on the smd.

On my new P4C800-Delux (Not E) Rev.2.00 I have made the mod again with success soldering directly on the smd. I used a wire with a 50KOhm trimmer with a 2-way plug in the middle so that I can unplug the resistor and misure with MM the real resistance withount the infuence of the mobos circuits (that alter the real trimmer resistance reading).
I have tried the following resistances with the following resaults:

38 KOhm:
bios 1.525 -> tester idle: 1.580 | full: 1.660

35 KOhm
bios 1.525 -> tester idle: 1.550 | full: 1.525
bios 1.535 -> tester idle: 1.565 | full: 1.550
bios 1.550 -> tester idle: 1.580 | full: 1.565

33 KOhm
bios 1.525 -> tester idle: 1.560 | full: 1.565
bios 1.535 -> tester idle: 1.565 | full: 1.575 (*)
bios 1.550 -> tester idle: 1.585 | full: 1.580

30 KOhm
bios 1.525 -> tester idle: 1.540/1.550 | full: 1.590/1.60 <<-- (my actual setting)
bios 1.535 -> tester idle: 1.565 | full: 1.610
bios 1.550 -> tester idle: 1.580 | full: 1.620


Here is a screenshot: http://img60.exs.cx/img60/6254/stabile.jpg

I started with 35KOhm and the I downgraded the trimmer resistance ONLY shutting down the computer, unplugging the trimmer and reading and changing the resistance with the MM reading.

I hope this could help.

[Aronier]

Hi,

i don`t think that i made a bad soldering. I had soldering on the alternative blue spots instead that directly on the smd. Do you think it help to do the mod directly on the smd?

[PIRATA!]

I have read on some forums that the alternative blue dots does not always function well.
In other words, most of the people that had this bad VCore values did the mod on these alternative blue dots.
Try to eliminate the trimmer from those spots and hope to have your mobo working back like before the mod because in most cases those bad VCore values just remain like that.
The only way to fix it then is to change mobo with another one and do the mod on the smd with very good and fast soldering.
P.S.: use a low watt iron.

[inferno_gr]

I assume you take the mm read from the mosfets, the reason for the elevated reading in mm is the voltage drop across the copper path from the fets to the cpu core.
Copper has a resistance even a very small one and that causes a small voltage drop wich depend on the cpu load.
The voltage regulator gets its feeback from the cpu pins, so u try read ur vcore as close to cpu pins as possible.
In my system, with the vdrop mod, a change of vcore setting in bios from 1,600 to 1,625 will result in a big increase in vcore, 0,1v.
It seems like some p4c800 models have this problem.