trick to get more stable vdimm on the dfi nf4!

[Lastviking]

One more thing vmem in smartguardian are 3.39v then its real 3.46v tested with multimeter..so if you have 3.33v in bios its more like ~3.38v so you need atleast +3.3 @3.5v+

[saaya]

what happens, if I set 3,3V Vdimm with 3,36V on the 3,3V rail?
can something blow up, if the 3,3V rail can't support the needed vdimm?

I believe you would just have an unstable vdimm around 3.3volts. That's not good to have, but shouldn't blow anything up in the short term.

exactly

has anyone tried this trick without a psu with adjustable rails...
what happens if you choose like 3.3 rails in the bios would you get 3.3v cause thats the max for the 3 rail, right. or would you still get 3.2.
what happens if you pick 3.4?

just interested i guess....

read the thread! :P

thx for the infos lastviking!

[[G.N.U.]Fragman]

any wise comments on a Tagan 480W guy....has it got the adjusteble pots inside the psu or should i just live with 3.2 for now...aint ready to do physicallly changes as pot modding to psu or board atm.

[Malves]

saaya any modd to get more than 4.0v? , i get this board soon and my bh5 memmorys need to feel some power

http://www.xtremesystems.org/forums/...ad.php?t=66624

[MonkSP]

Which other components use the 3.3 rail? My 3.3 rail is set @ 3.54 to have a true 3.3 rail. Its working perfectly. My PSU is a OCZ Powerstream and can delivers 3.8 in that line. But how much can I deliver without hurting the components?

[undefeatedone]

Few PSUs i wouldn't try this on granted it is same as vcap mod.
1) PC&P 510 Dlx.
2) Enermax (pre noisetaker models)
I've had friend who would consistantly kill A64 memory controller doing it on above 2 PSU. He moved on to OCZ Powerstream and had no more dead FX-55 (well at least on memory controller). Zippy is another one that worked fine for months... where is my zippy now... hmmmm -_-;

I have a PCP&C 510W Dlx SLI. I'm thinking about doing the 5V vdimm mod to get 3.5V with my PC4000 VX Gold. Does he mean I shouldn't do the 3.3V mod, the 5V mod, or both?

[saaya]

Which other components use the 3.3 rail? My 3.3 rail is set @ 3.54 to have a true 3.3 rail. Its working perfectly. My PSU is a OCZ Powerstream and can delivers 3.8 in that line. But how much can I deliver without hurting the components?

sata hdds afaik and pci devices, other than that only the mainboard to generate vdimm and sometimes the chipset voltage.
hard to tell what will hurt components. ive been running 3.8v for almost a year and 3.5v for half a year now without any problems.
well one of my hdds died and two had corruption problems, but tha has nothing to do with the 3.3v rail since the dont even use that rail.

I have a PCP&C 510W Dlx SLI. I'm thinking about doing the 5V vdimm mod to get 3.5V with my PC4000 VX Gold. Does he mean I shouldn't do the 3.3V mod, the 5V mod, or both?

id suggest you use the 5v feed option to get 3.5v vdimm. so switch the jumperblock on the dfi mainboard to 5v feed, and in case your using a 20pin psu, get a 20 to 24pin psu adapter

[Flib]

Current SATA HDDs aren't using the 3.3V rail!
The 3.3V wire of the SATA connector isn't in the current spec and not used yet. Maybe in a few months/years it will be used.

My PSU has no 3,3V wire on the SATA connector(native!). There are also Molex-> SATA adapters around and it works without 3,3V.

3,3V is only used by mem, pci cards and maybe some onboard components.
So you shouldn't get HDD corrupction because of too high 3,3V rail!

[funkflix]

2 days ago, i placed a temp.-sensor on the Vdimm mosfet and today i started memtest 1.60 and the temps of the mosfet reached 80°C easy! I already had installed the single jumper on the jumper-table, bcs. i had this mod installed before(but i switched back to the +5V bcs. of testing with higher vdimm), so i switched back to the +3.3. Now look at the temp. difference.. i was really impressed!

Before -

http://213.202.211.105/gravedigga/Temp/before.AVI

After -

http://213.202.211.105/gravedigga/Temp/after.AVI

Really nice mod!

[Spyrus]

A question from me:
When setting the jumber to 3.3V i cannot have vdimm over 3.08 Volts.
I need 3.12 volts & the only way that i can get it is by putting the jumber to 4V option.

So this tip here will help or my PSU is responsible for its crappy 3.3 rail?
ANTEC True Power 430W-20pin is my PSU by the way.

I know that it is not recommended to run at 4V jumber when someone wants less than 3.2Vdimm...am i right?
I use a 120mm that blows to that area,have no high temps but i am still concerned with low vdimm(3.12) with 4v jumber.

[funkflix]

U need a +3.3V railmod on ur PSU then.

[h3x0r_1984]

I know it's been a long time since people haven't posted here,but i have a problem with the trick on my board.I have a very profi Vsense on my Enermax Noisetaker.One pot adjusts 3.3V between 3.3V and a Max value...and the other pot...adjusts the Max value.
Now i have set the the 3.3V rail at 3.56V measured on a profi multimeter.I set the jumpers in the correct position....and if i set VDIMM at 3.4V..all i get is 3.3V rock stable(read in the BIOS).If i set VDIMM at 3.3V i get the same thing.With or without +0.03 Enable.What's the problem?Why can't i get 3.4V VDIMM with a 3.56V on the rail ? The difference between VDIMM and 3.3V rail is too small ?? The first posts said some thing about a 0.25 difference to a have a proper regulation and a rock stable VDIMM.
But i saw that JasonXXX had a 3.6V on the rail..and a 3.5V stable on VDIMM,on a PC&C.Or was he saiyng bollocks?

I should mention that the voltages are rock stable..and i have only a 0.05 Drop on the 12V rail between load and idle.
PS:Where can i measure the VDIMM on the board?

[Magnj]

Old thread Resurrection!
I think you need to be measuring off the board to be sure of what voltages you are getting. You can measure how much Voltage your RAM is getting by using a multi and grounding one lead and touching the other to a small circle near the DIMM, half will be while half will be black. Touch either side one will just be -, and that should tell you how much voltage our rams are getting

[Magnj]

Is there a backup of that original picture..It would be quite useful, I could host it if needed.

And while I'm at it I have a question.
I have this jumper trick ( I think, I haven't looked at this stuff in a while But I'm pretty sure I have this jumper trick done unless there is another Jumper trick ) anyway, I have my 3.3v Rail at 3.55 and I have this jumper trick, bit in the DFI 406 BIOS I'm not being allowed to select more than 3.2V o_O whats goin on?

[funkflix]

[Magnj]

Awesome, does anyone mind If I mirror this on Flickr?

And could anyone tell me why I would not be seeing more than 3.2V with this trick?
I have the jumpers set right still from last year but when I flashed my BIOS to 4-06 I get no option to choose more than 3.2...

[funkflix]

Host it wherever u want.

U can only choose > 3,2V if u are not on Auto-Vcore and some other things i already forgot about..

[Magnj]

Yea I know about that. I have a feeling I have to switch back to standard, choose a non auto voltage, and then go back to the trick.

[halcy]

Resurrecting again. Does anyone have a diagram of where to measure the vdimm on an SLI-DR with a multimeter?

I am having a bizarre problem, I have a pair of 1GB Ballistix DDR sticks running on this motherboard in a media center pc setup. The machine is on 24/7, either fully on or in standby. And in the last 8 months I have had to RMA the ram twice. What happens is the machine is fine when its on, itll come in and out of standby no problem for weeks or months on end, however when I actually have to physically power the machine off for more than a few minutes it will then cease to boot, giving me the dreaded memory beeps. The only thing I can think to check is the voltage during standby/full operation, does anyone else have any suggestions on what could be wrong?

Is standby harder on ram than running normally 24/7? I can't imagine how it could be but I don't design the boards. I am also not overclocking these chips at all, running them at a very loose 200 3-4-4-8, stock 2.6V.

Thanks-

[halcy]

Resurrecting again. Does anyone have a diagram of where to measure the vdimm on an SLI-DR with a multimeter?

I am having a bizarre problem, I have a pair of 1GB Ballistix DDR sticks running on this motherboard in a media center pc setup. The machine is on 24/7, either fully on or in standby. And in the last 8 months I have had to RMA the ram twice. What happens is the machine is fine when its on, itll come in and out of standby no problem for weeks or months on end, however when I actually have to physically power the machine off for more than a few minutes it will then cease to boot, giving me the dreaded memory beeps. The only thing I can think to check is the voltage during standby/full operation, does anyone else have any suggestions on what could be wrong?

Is standby harder on ram than running normally 24/7? I can't imagine how it could be but I don't design the boards. I am also not overclocking these chips at all, running them at a very loose 200 3-4-4-8, stock 2.6V.

Thanks-

Found an answer on where to measure vdimm:
http://www.diy-street.com/forum/showpost.php?p=64740

I measured my Vdimm during standby and it went from 2.62-2.63V down to 2.35V.. can prolonged undervolting cause dimms to fail?