P5N-E vmods

[zeeke]

Originally Posted by couppi

After doing the mod, I'm getting 1.404v idle and 1.395v load with 1.4v selected in BIOS. Great success!

With 1.5v set in BIOS, I'm getting 1.498v idle, 1.486 load. The BIOS voltage reporting is waaay off. Asus probe tells me 1.44v idle, 1.33v load.

Hi couppi, two questions:

1. Is it possible to adjust the trimpot to get even lower drop?

2. Are you still experiencing that the drop gets even worse with higher voltage, have you tried with 1.6V?

[Lestat]

and what mod are you talking about, considering all talk of this mod has been nothing but speculation, what did you mod and please show pictures if possible.

[couppi]

I did the Vdroop mod. Before the mod I was getting .04~.06v drops at load. Now I only get .01v drop at load. Just connect these two points with a 1 mohm resistor. I guess you could do this mod by pencil if you penciled the resistance down to 105~110 kohms.

Originally Posted by zeeke

Hi couppi, two questions:

1. Is it possible to adjust the trimpot to get even lower drop?

2. Are you still experiencing that the drop gets even worse with higher voltage, have you tried with 1.6V?

1. .01v drop was the best it would get. When I lowered the resistance any further, it raised both the idle and the load voltage.

2. I haven't tried that yet, but I will.

[celemine1Gig]

Originally Posted by couppi

I did the Vdroop mod. Before the mod I was getting .04~.06v drops at load. Now I only get .01v drop at load. Just connect these two points with a 1 mohm resistor. I guess you could do this mod by pencil if you penciled the resistance down to 105~110 kohms.

[/img]http://img145.imageshack.us/img145/1305/30fh0281848617xs1.jpg[/img]



[i]1. .01v drop was the best it would get. When I lowered the resistance any further, it raised both the idle and the load voltage. [/i[

2. I haven't tried that yet, but I will.

I guess that the 1M resistor is too flaky in the lower regions (the trimmer's not linear and the low resistance range changes qickly). Try this: Connect a 500k trimmer in series with a 100k trimmer and use that. This way you get a total of 600k and 500k in 15-20 turns + 100k in the same number of turns. This increases precision dramatically. You could also just add a smaller trimmer in series to the 1M trimmer. Let's say a 200, 100 or even 50k and use that for fine tuning. You should be able to adjust it easier/finer then.

[dlxmax]

Thanks for being the first to test and confirm, Couppi!

My new multimeter was delivered this morning, but I still have to trek back to the electronics shop to get the right trimmer(s).


EDIT:
Ok, so the possibility of a pencil mod to fix the vdroop was mentioned, and I figured, "what the heck..."

I measured initial resistance, with the power off, of the resistor shown in the mod photo and I also got ~130k ohms.

So now I'm setting about to do the pencil mod...

I've not been able to measure the vcore directly from the motherboard because I'm not clear on how to do it. It's been 14 years since high school electronics, and even then we never got to tinker with computers. Looking at the VR-Zone page I see the mosfet leg to measure, but where should I ground out?

[zeeke]

you can use any groundpoint you like on motherboard, easiest to locate is the mounting holes you use to fasten it in the chassis. Or you can use the third pin of a fan connector, where the black cable goes.

[dlxmax]

Thanks, Zeeke. Oh, and thanks a million Celemine for coming back to this thread and solidifying the vdroop mod for us!

My board measured different inital resistance compared to Couppi. I got half the original resistance he measured! I came in at 65k ohm. My meter is new so I tested it on another known resistance and it was correct. Soooo....

I went ahead and did a pencil mod on the resistor. Pre-mod, my board-measured vcore was 1.46 at idle, 1.40 at load. I dropped the resistance in stages, and found that 28k ohm was nearly ideal. Pre-mod, I set my bios at "1.38125v" to get those above voltages. But at 28k ohm, my idle voltage rose up to 1.49v. So, I dropped the bios setting down to "1.300000v" which gives me slightly more than 1.40v at idle and about 1.40v at load.

ORTHOS testing now, I was right at the minimum stable voltage for my clock rate before. I'm running an E4300 at 3256mhz (stable). However, pre-mod it would boot up at 3600mhz with lots of voltage, but ORTHOS would fail within seconds. Maybe it's time to try that again!

[celemine1Gig]

Originally Posted by dlxmax

Thanks, Zeeke. Oh, and thanks a million Celemine for coming back to this thread and solidifying the vdroop mod for us!

My board measured different inital resistance compared to Couppi. I got half the original resistance he measured! I came in at 65k ohm. My meter is new so I tested it on another known resistance and it was correct. Soooo....

I went ahead and did a pencil mod on the resistor. Pre-mod, my board-measured vcore was 1.46 at idle, 1.40 at load. I dropped the resistance in stages, and found that 28k ohm was nearly ideal. Pre-mod, I set my bios at "1.38125v" to get those above voltages. But at 28k ohm, my idle voltage rose up to 1.49v. So, I dropped the bios setting down to "1.300000v" which gives me slightly more than 1.40v at idle and about 1.40v at load.

ORTHOS testing now, I was right at the minimum stable voltage for my clock rate before. I'm running an E4300 at 3256mhz (stable). However, pre-mod it would boot up at 3600mhz with lots of voltage, but ORTHOS would fail within seconds. Maybe it's time to try that again!

Redo the mod! The Droop-Mod is not there to increase idle volts. It's just to adjust idle to load voltage. There should be no change in idle volts compared to a unmodded board.

[Ron 61]

Where do you measure vcore for this board? Ive tried the leg shown at vr zone with mb mounting holes for ground and I get 1.15v. I have 1.46 selected in bios and asus probe shows 1.41 at idle so 1.15 is not right. Ive tried 2 known good DMM that give correct readings for my video card and psu rails. Is there another point to measure vcore. Any help would be much appreciated.

[sbinh]

Originally Posted by Ron 61

Where do you measure vcore for this board? Ive tried the leg shown at vr zone with mb mounting holes for ground and I get 1.15v. I have 1.46 selected in bios and asus probe shows 1.41 at idle so 1.15 is not right. Ive tried 2 known good DMM that give correct readings for my video card and psu rails. Is there another point to measure vcore. Any help would be much appreciated.

Use the fan connector (black wire) as ground. You'll get better result.

[Ron 61]

Thanks sbinh I tried fan connector ground but its still way off. 1.22v idle 1.17 load. I want to do the vdroop mod but I can't do it when I can't even get accurate vcore readings. Have you measured your vcore with dmm and if so what did you get?

[dlxmax]

Are you sure you're taking a reading for vCore and not the chipset or something else? Triple check your metering spot because nobody else has had a problem. If you're at the right spot, check your meter on a known voltage like a fresh AA battery which will read ~1.6v.

[Ron 61]

Im taking reading from the top leg of the mosfet shown in the vr zone guide. Ive checked meter with batteries and other known voltages and it reads exactly what it should.I don't know why I'm getting 1.22v.

[sbinh]

These are what I just got .....

1st pic ... ground directly from PSU

2nd pic.. ground from motherboard ...


vcore in BIOS: 1.5375v .... speedfan: 1.47v ......

All vcore is at idle ... 400x9
When loaded, it drops down to 1.41v (in speedfan .. maybe 1.44v with DMM)

[VASkO]

i did the vdroop mod now. tuned it so, that the difference between idle&load is minimal. 1.550V idle and 1.549V under orthos load or 1.443V idle and 1.442V under orthos with different voltage setting through bios.
now it's time to find new voltage settings for a stable OC with this mod. hope, i can drop the volts a bit

[sbinh]

After soldering a 50K for vmod......

BIOS set to 1.49375v ---



Speedfan (4.32 Beta) idle shows 1.50v ---



DMM idle shows: 1.542v ---



Speedfan (4.32 Beta) loaded shows 1.44v ---



DMM loaded shows: 1.490v ---




Edited: using wrong term .. it's should be vmod - not vdroop.

[zeeke]

why did you solder a 50k for vdroop when you are supposed to use a 1M ?

[zeeke]

Did the Vdroop mod today, got the drop pretty stable. But I can confirm that the idle voltage DOES rise at some point. Before Vmod idle volt were 1.58V, and load 1.54V, efter mod, idle 1.60V and load 1.60V-1.59V. Res is to low on my multimeter to adjust it any better.

Have to say that so far my system doesnt seem more stable thou ;( But I have to play around with it for a couple of hours before final judgement..

[ineedaname]

ok first off i have no clue what i'm doing but I did two pencil mods and now this mobo is running voltage solid as a rock. I used to have need about 1.52 to get the cpu stable at 3.3ghz now it only needs 1.42.

The first point I penciled was the point given in VR and that made the voltages i entered into the bios actually show up after i rebooted.

The second point I penciled was that point that couppi pointed out and that got rid of my vDroop completely.

I used a 4B pencil and did about 5 strokes on the 2nd point and about 30 on the first point.

[nosleep]

Originally Posted by ineedaname

I used a 4B pencil and did about 5 strokes on the 2nd point and about 30 on the first point.

Do you remember what the resistance was on the vdroop cap after the mod?

Or anyone know the resistance after the vdroop mod?

Thanks


*EDIT*
ok, my vdroop cap measures 145k stock before mod. Gonna do some pencil surgery and report back...

[sbinh]

OK .... I did play around with this board for almost 2 hrs .....
and try to vdroop using pencil (PaperMate 2 HB) ....

Decided to remove the vmod I did earlier and apply vdroop mod...
This is what I did






As you can see in my previous post, those are before vdroop (after vmod .. which is suck )

This is what I got after vdroop:

vcore set= 1.500v
vcore in BIOS monitor= 1.47v

At IDLE

vcore in speedfan=1.47v
vcore in DMM=1.501v

At LOADED
vcore in speedfan=1.44v
vcore in DMM= 1.48v


I know, it is not the best yet (compare to what Couppi did), but I am happy with it ... Used to set vcore to 1.525v to run E6600@3.5GHz .. now, running same speed at lower vcore (1.5v) .. (which can be lower at least 1 notch)

[nosleep]

Ok, big thanks to couppi for the vdroop mod.
Resistance was 145k before and 33.5k after

Here are my results:
Bios: 1.40v

Actual Idle: 1.44v
Actual Load: 1.43v

[dlxmax]

I've got a question regarding my meter readings. My vcore is incredibly stable now thanks to the mod, but I'm unsure what is my actual voltage.

With Vcore set at "1.375v" in the bios, the bios hardware monitor reports 1.39v. My meter set at "2000m" reports 1.342v, but my meter set at "20" reports 1.48v.

Testing the meter at the same settings on AA and AAA batteries doesn't show any discrepancy between the "2000m" and "20" settings. Further, all testing of the vcore was done with the same ground (fan pin header, fan set to 100%).

I also metered the Vchipset at 1.372v on "2000m" or 1.43v on "20". It's set at "1.393v" in the bios.
Vmem also couldn't meter at the "2000m" setting, but reports 2.12v at "20". It's set at "2.085v" in the bios.

Is this the result of voltage ripple? At which setting, "2000m" or "20" am I to believe the readings? Should I have spent more than $5 on a meter?

[ineedaname]

i think you should spend a bit more on the meter and get 1 that can go down to 2

[celemine1Gig]

Originally Posted by dlxmax

...Should I have spent more than $5 on a meter?

:thumbsup: