Asus Striker Extreme vDroop Mod

[Smooth]

A big thx here from me !
I used to have a E6850 on a P5N32-E SLI and the Vdrop was hardly noticeable ( after the Bios-flash - the initial Vdrop was extreme ) ... all went fine ....
Then I change CPU to a Q6600 (SLACR) and started oc:
first thing I noticed "hi my dear friend Vdrop, you´re back again ... how has it been ?" -
Damn it, Vdrop @ 0,03V
Vdroop @ 0,05 V
After pencil-mod
Vdrop : 0,0
Vdroop : 0,0


So I could lower my Vcore@oc - so I got some degrees back again. To be precisely = about 5°C ( that brings me joy )
Q6600 @ 3,2GHz@load @1,28V@ 39°C (CPU-Temp)

[ineedaname]

Ok thnx for the info.
I'll just do the vdroop mod then.

But what motherboard is the one in the image?

I'm not 100% sure but I think it IS a striker extreme.
I don't really remember as it was posted so long ago.
Its either a striker or the P5N32E-SLI

Can you solder the wires to the back of the board instead of the top? Seems like it would be easier that way...

I am not an expert at soldering but I wouldn't attempt to solder the back of the board instead.
The reason is because these mobo's have at least 6 layers and u don't know if those connections on the back are the same as the front.

A big thx here from me !
I used to have a E6850 on a P5N32-E SLI and the Vdrop was hardly noticeable ( after the Bios-flash - the initial Vdrop was extreme ) ... all went fine ....
Then I change CPU to a Q6600 (SLACR) and started oc:
first thing I noticed "hi my dear friend Vdrop, you´re back again ... how has it been ?" -
Damn it, Vdrop @ 0,03V
Vdroop @ 0,05 V
After pencil-mod
Vdrop : 0,0
Vdroop : 0,0


So I could lower my Vcore@oc - so I got some degrees back again. To be precisely = about 5°C ( that brings me joy )
Q6600 @ load @1,28V@ 39°C (CPU-Temp)

Nice 2 hear that some ppl are still using the mod

[Smooth]

I'm not 100% sure but I think it IS a striker extreme.
I don't really remember as it was posted so long ago.
Its either a striker or the P5N32E-SLI

The picture in the first posting seems to be of a Striker Extreme ( cause of the capacitator ).
But it works on the P5N32-E ...

Nice 2 hear that some ppl are still using the mod

What means "still using" ?
For quad-oc it´s more needed than ever - in my opinion.
edit :
Now I was able to push my Q6600 a little bit further.
9x425@1,4V
So a big, big thx !

[DonNiger]

Has anybody measured the original resistance for the vdroop. Mine shows 58,8 kohms. Somebody mentioned that he took it down by 100 kohm

I penciled it down to 29 KOhms and results are:
Bios 1,70> Probe Idle 1,71 > Probe Load 1,70 (1,71) > Multimeter Idle 1,68 > Multimeter load 1,55
So, it actually got worse.
Previous testing

Resistance > Bios > Idle > Load > MM Idle > MM load
58,8------>1,875>-1,86->1,82-->1,83----->1,80
54,4------>1,875>-1,86->1,82-->1,83----->1,80
43,2------>1,875>-1,87->1,84-->1,84----->1,81
35,4------>1,875>-1,87->1,86-->1,85----->1,83

[Stigma]

Thank you ineedname, in the pic below is this correct?

Link to the picture refered to: http://www.xtremesystems.org/forums/...1&d=1193096766

You missed one of them. I assume that the last one (in the picture its half-covered by the heatpipe) is also a valid measure-point?

I ask because I have the whole area insulated and the only one of these coils that are exposed is the backside of the one you didnt point an arrow to.

it would be a shame for me to assume something and blow my board because of it, so please clarify if you left that one out intentionally (however unlikely I think that may be )

-Stigma

[Stigma]

Sorry for the double-post but I have one more thing I want to get straight before I attempt the vdroop mod.

While turned off, I measured my resistance over the Vdroop point to be 59,5K ohms (+/- 0,1K ohms). Now, earlier in this thread Ineedaname said that we should take away roughly 50K ohms. That only leaves 9,5K ohms (~16&#37 of the original resistance, and thats seems (to a electric noob like me atleast), like its a hell of a lot (removed that is).

So, Ineedaname (or others who have completed the mod), can you confirm that this is correct, or present some ballpark number of what resistance it should be left with after the mod? FOr anyone who has done the mod, just do a multimeter measurement over the resistor and give us that number

Now obviously I won't just blindly pencil it all in one go even if I get a number to go by, but it would sure as hell make me a lot more comfortable if I had something to aim for when I start penciling, otherwise the first pencil stroke is sort of a blind leap of faith. If I had a estimated target resistance I could go half the way to the target value, then measure the actual load Vdroop change and adjust from there on out.

(Btw, who needs a vcore mod on a board that can be adjusted as high as on the SE? ... apart from getting correct readings in the BIOS? I sure won't be doing it )

Thanks!

Has anybody measured the original resistance for the vdroop. Mine shows 58,8 kohms. Somebody mentioned that he took it down by 100 kohm

I penciled it down to 29 KOhms and results are:
Bios 1,70> Probe Idle 1,71 > Probe Load 1,70 (1,71) > Multimeter Idle 1,68 > Multimeter load 1,55
So, it actually got worse.
Previous testing

Resistance > Bios > Idle > Load > MM Idle > MM load
58,8------>1,875>-1,86->1,82-->1,83----->1,80
54,4------>1,875>-1,86->1,82-->1,83----->1,80
43,2------>1,875>-1,87->1,84-->1,84----->1,81
35,4------>1,875>-1,87->1,86-->1,85----->1,83

Looks like we posted on top of eachother regarding the exact same thing. Refer to my text above and you will see my non-modded resistance is aprox equal to yours.
From your test results it seems that we indeed might have to remove a whole lot of the resistance in order to iron out the Vdroop, but I can't explain why you suddenly got such a (dramatic!) worsening of your Vcore when you went down to 29K ohms. I'm pretty sure that must be either a mistake somewhere during your testing or penciling. I would try going even lower (and make sure the grafite from the pencil dosn't get anywhere its not supposed to). All your previous data after all shows that it gets gradually better and better the lower you go.

You said someone talked about removeing 100K oms, but thats not right. That was for the Vcore mod, not the Vdroop mod. The ballpark number Ineedaname said to remove from the original resistance was 50K ohms. Granted, that is nearly all of it, leaving only about 16%, but given your data that seems to actually make sense if you extrapolate the values from your data so far (excluding the single datapoint at the end which as I said I believe is a fluke or mistake). Given the data you have posted I now have no doubt about -50K ohms being the aproximate resistance needing to be removed. The exact value probably varies a bit from board to board, but its a ballpark number atleast.

-Stigma

[DonNiger]

If I'll have time, I'll go lower. But system was already more unstable than it was before - I couldn't run my P4 more than 5 Ghz, without vdroop mod it was running at 5,3 Ghz.


Resistance > Bios > Idle > Load > MM Idle > MM load
26,9------>1,70->-1,71->1,70-->1,686----->1,675
20,8------>1,70->-1,73->1,71-->1,689----->1,682

But, OC is going worse and worse. Also, the VTT and HT volts are rising.


EDIT: It is useless mod! Tried with several CPUs and results were all the same. With that vdroop mod, the maximum OC is about 10% less.
P4 631 without mod 5,4; with mod 5,0
P4 506 wthout mod 4,9; with mod 4,4

[Jultsu]

and another happy vdroop user here ZERO droop after some penciling

[alexanderhuzar]

Great mod! - works

Before the mod, a 1.55 in BIOS would end up as 1.52 volts MAX and go down to 1.48 under load. (Multimeter confirmed.)

With the mod (vDroop mod only), I set BIOS to vCore = 1.55, and now my PC Probe and Everest show 1.55, and my SpeedFan and CPU-z show 1.47 and 1.456 respectively. The voltage readings are flatlined with the occasional spike of 0.02 volts over BIOS setting (about 1 spike every 2-5 minutes). This spike is always exactly 0.02 volts over (1.57 volt (spike) versus 1.55 (constant) set in BIOS).

Notes:
"Multimeter confirmed" = measured from the bottom side of the contact point of the vCore measure point (not the top side which always measures 0.03 volts less). I gave up using multimeter because the BIOS, PC Probe and Everest vCore readouts were identical to the multimeter.
Resistance measurements were done with my multimeter set to "2000k OHM".
Pencil used was no-name with custom writing for a construction company (my friend uses it to mark stud locations on drywall). Cylyndrical body (not hexagon), with red eraser on the end. The mark it leaves behind on paper is solid and slightly reflective if you press down lightly. 100% solid and reflective if you press down medium-hard. I guess this is a more graphite-rich model compared to your regular pencil.


Stock resistance of the resistor = 57 kOHM.
1 pencil stroke = 9 kOHM - vDroop still exists, but I found it really surprising just what a fenominal amount of resistance was reduced by a single pencil stroke (pressing down with light pressure).
6-8 pencil strokes = 5 kOHM (too much strokes lol )(vDroop is gone, but now there's occasional spikes of excess voltage).


Can you guys please share what you measured on your multimeter after you were satisfied with vDroop (or lack thereof)?