How to volt mod IC7 Vmem

[DDA]

Originally posted by MnM
Thanks to you guys I have finished this mod.
I replaced the 0Ohm resistor with a 331Ohm one (IC7-G).
It works perfect. At 2.55 selected in bios it will actually give me 2.75. Didnt try any higher yet.

Funny thing is this.
I have done one motherboard first and it worked perfect.
Then I did my friend's one. He would get a blank screen and no post. We replaced the resistor (the 331Ohm one) with a new one, thinking we might have given it too much heat. Same result. No display at all. What I have done then, is I shorted the 331Ohm resistor with some twizers therefore giving it a 0Ohm value.
I got post. I removed the twizers not post.
In the end what I have done is I put another resistor (331Ohm) ON TOP of the 331Ohm already on the motherboard. I solder it in parallel. We turn the PC on - POST!!!!
Now with this mod, he gets only half the voltage from the normal mod eg. 2.55 selected in bios actually gives him 2.67 real volts.
He was happy with it as he already thought he had to spen $$ for a new motherboard. Plus in the end he still gets more than the 2.8 limit.

So 2 identical boards, two identical mods but only one of them worked. I think the components used on the boards and their tolerances play a big part in modding the vdimm. I am now sure that if I had a 400-450Ohm resistor around and would have actually solder it to my friends board will have worked giving him a bit better than just 0.1 increase over the value selected in bios.

Maybe this info can help the guys who cant get a post after the board has been modded. Try solding one to the motherboard and one on top of the resistor. At least you will get more than the defualt max voltage in the end.

2 x 331 Ohms in parallel make 165 Ohms abt ....

[Kin Hell]

Originally posted by DDA
2 x 331 Ohms in parallel make 165 Ohms abt ....

Sure, and if they were in Series 662 Ohms,...but summat aint adding up with this board and how the VDIMM is controlled.
Of all the boards I have ever modded for Voltages, I have not yet seen one give you more Voltage as you increase the Resistance. Then along came the IC7-G..........................

Then there is the anomally of the Mod not working for some of us. I've tried 3 diff VR's without success...board dead everytime.

Something isnt quite right here!

[DDA]

Originally posted by Kin Hell

Something isnt quite right here!

I agree with you ....

[NyCUndaGrounD]

Originally posted by Kin Hell
Sure, and if they were in Series 662 Ohms,...but summat aint adding up with this board and how the VDIMM is controlled.
Of all the boards I have ever modded for Voltages, I have not yet seen one give you more Voltage as you increase the Resistance. Then along came the IC7-G..........................

Then there is the anomally of the Mod not working for some of us. I've tried 3 diff VR's without success...board dead everytime.

Something isnt quite right here!

Are u sure u didnt burn the pads out of recognition. i burned the resistor next to the R50 beyond recognition and it still works!LOL.
When i see the pads disappearing i add a bit more solder onto it to keep it alive.

[Kin Hell]

Originally posted by NyCUndaGrounD
Are u sure u didnt burn the pads out of recognition. i burned the resistor next to the R50 beyond recognition and it still works!LOL.
When i see the pads disappearing i add a bit more solder onto it to keep it alive.

I wouldnt have thought so considering the "0 Ohm"resistor is in place atm & thats been resoldered back to the board 3 times now! I didnt scorch the Res. next to it either but the 0 Ohm Res. has about half of one of its original three zeros. So I guess that pup has gotten a tad warm!

A 15Watt Iron & suitable tip with the all important "steady-hand" should not have scorched the Res. next to R50. Your a tad lucky there m8!

[Kin Hell]

This is really aimed at anyone that might have done this Mod, but, having done the VDIMM Mod 3 times & get NO post on each occasion, I did wonder if the VTT might be the root cause of NO Post.

I only ask this as when I set my VDIMM @ 2.8v, Hardware Doc has alsrm bells ringing telling me my VTT is out of range (too high)

As im increasing the VDIMM with the 1K VR, I did wonder if that would exaberate the VTT issue I see & therefore render me a NO Post scenario!?

MnM discovered that adding another Res. in Parallel lessend the VDIMM available, but at the same time, would not have pushed the VTT quite as high as was prolly happening with a NO Post with just 1 x 330Ohm Res.

Whatchya think lads?

Beers all around again!

[tome]

Hi,

I just tried the mod using a 1k pot this morning. Thereby I removed the R50 resistor by accident (or using two left hands). Now the resistor is off but there is nothing left where I can solder the two wires coming from the 1k pot. Maybe I also removed successfully the contacts of the boards.

Any chance to avoid scrapping the board? Maybe using the 4th leg of the winbond and mass from the mosfet nearby (as posted earlier)?

Greetz tome

[Pita^Norf]

Originally posted by Kin Hell
This is really aimed at anyone that might have done this Mod, but, having done the VDIMM Mod 3 times & get NO post on each occasion, I did wonder if the VTT might be the root cause of NO Post.

I only ask this as when I set my VDIMM @ 2.8v, Hardware Doc has alsrm bells ringing telling me my VTT is out of range (too high)

As im increasing the VDIMM with the 1K VR, I did wonder if that would exaberate the VTT issue I see & therefore render me a NO Post scenario!?

MnM discovered that adding another Res. in Parallel lessend the VDIMM available, but at the same time, would not have pushed the VTT quite as high as was prolly happening with a NO Post with just 1 x 330Ohm Res.

Whatchya think lads?

Beers all around again!

It is interesting that MnM after effectively halving his resistance across r50 that the the board decides to boot up, this brings me to think if you started your resistance at say 100 ohms rather than the 300 would you have a working board now albeit with lower increase to Vdimm than desired?

Now I don't know really enough about VTT, but should this usually be set to about 1/2 of Vdimm as with most boards and is used to stabalise the Vdimm ?

If your board is telling you that VTT is out of range then it could possibly be the case that the VTT levels are'nt enough to stabalise the Vdimm you are trying to run and thus a no boot ?

[Kin Hell]

Originally posted by Pita^Norf
It is interesting that MnM after effectively halving his resistance across r50 that the the board decides to boot up, this brings me to think if you started your resistance at say 100 ohms rather than the 300 would you have a working board now albeit with lower increase to Vdimm than desired?

Now I don't know really enough about VTT, but should this usually be set to about 1/2 of Vdimm as with most boards and is used to stabalise the Vdimm ?

If your board is telling you that VTT is out of range then it could possibly be the case that the VTT levels are'nt enough to stabalise the Vdimm you are trying to run and thus a no boot ?

I dont have the kit to VTT mod & tbh, after thinking i'd fugged meh board 3 times & when u consider it appears to be ok for some but not for others, I'm kinda wonderin if it's really worth pursuing. ........Especially with the fact we are Increasing Resistance to Increase Voltage. Strictly speaking, thats total fubar against Electronic basics.
Perhaps i`m better off just buying OCZ- GOLD and having done with it but then the wallet is fugged real bad atm.

[crotale]

Why not try to post at 0 ohms..? Then it _should_ post...
Then upping it while monitoring the bios...

Question: What will happen if the resistance is completely removed? ("infinite" resistance)

[Pita^Norf]

Originally posted by crotale
Why not try to post at 0 ohms..? Then it _should_ post...
Then upping it while monitoring the bios...

The r50 is more or less at 0 ohms...last time I measure it was between 0 and 1 ohm.

Originally posted by crotale

Question: What will happen if the resistance is completely removed? ("infinite" resistance)

Infinate resistance is the opposite of zero, so to be at zero & infinity at the same time is not possible.

If what you ask is what happens if we replace R50 with infinite resitance, then thats an unknown? My guess is that there must be a limit that the Winbond chip will take and will not boot if this is exceeded, I think Nycunderground may have dumped 5k ohms across r50 pads and got a no boot.

[eva2000]

Originally posted by MnM
Thanks to you guys I have finished this mod.
I replaced the 0Ohm resistor with a 331Ohm one (IC7-G).
It works perfect. At 2.55 selected in bios it will actually give me 2.75. Didnt try any higher yet.

Funny thing is this.
I have done one motherboard first and it worked perfect.
Then I did my friend's one. He would get a blank screen and no post. We replaced the resistor (the 331Ohm one) with a new one, thinking we might have given it too much heat. Same result. No display at all. What I have done then, is I shorted the 331Ohm resistor with some twizers therefore giving it a 0Ohm value.
I got post. I removed the twizers not post.
In the end what I have done is I put another resistor (331Ohm) ON TOP of the 331Ohm already on the motherboard. I solder it in parallel. We turn the PC on - POST!!!!
Now with this mod, he gets only half the voltage from the normal mod eg. 2.55 selected in bios actually gives him 2.67 real volts.
He was happy with it as he already thought he had to spen $$ for a new motherboard. Plus in the end he still gets more than the 2.8 limit.

So 2 identical boards, two identical mods but only one of them worked. I think the components used on the boards and their tolerances play a big part in modding the vdimm. I am now sure that if I had a 400-450Ohm resistor around and would have actually solder it to my friends board will have worked giving him a bit better than just 0.1 increase over the value selected in bios.

Maybe this info can help the guys who cant get a post after the board has been modded. Try solding one to the motherboard and one on top of the resistor. At least you will get more than the defualt max voltage in the end.

nice MnM, would love to do this mod on my IC7 but never held a soldering iron in my life and have no clue :o

[NyCUndaGrounD]

I was wondering if it was possible to adjust the pot's knob while the computer is running?

[Vatt]

Yes, that is what I did.

I started MotherBoard Monitor and adjusted the pot untill I read 3.1v on the DIMM's

[Pita^Norf]

Originally posted by NyCUndaGrounD
I was wondering if it was possible to adjust the pot's knob while the computer is running?

It may be ok, but certainly not something I would advise.

I'd rather shut down properly than risk killing a board.

[crotale]

ok, I just went to a guy who maybe could do the mod...

To tell the truth: I was SCARED SH1TLESS!!! He was more shaky than my grandma'!
"Ooops, one pad almost got detached"
GHAAAA!!!

Well, I'm gonna try it now... I got a poor 1.6A and some cheap 256 stick laying around...

Pita^Norf: The resistor is a 0 ohm... Increasing the resistance will increase vdimm. Removing the the resistor equals infinite resistance... What happens then, that is my question...

What is the upper limit of the winbond chip on the vdimm?

[Vatt]

Originally posted by HWF
It would be great if some of the pro's could post some pics of the Vtt mod.

Yes, I think that is what I need also.

Let me explain:

My XMS3500 needs 3.0v for DDR466 2-2-2-5

I set in BIOS Vdimm at 2.8 and turned the pot untill I got 3.0v
Everything is stable and working perfectly.

Now, let's say I put in BIOS only 2.7v and turn the pot more so I ALSO get 3.0v. It SHOULD be stable... because 3.0v is enough for the XMS3500 and proven to work.

BUT.. it is not stable. Both times I have 3.0v , but the difference is in VTT.
With 2.8 in BIOS --> VTT is 1.4 and all stable
With 2.7 in BIOS --> VTT is 1.35 and UNSTABLE

I hope it's all clear to you all, what I mean ...

Now, I THINK that by increasing Vtt I will be able to higher overclock. I have come to FSB 282 and 5:4 divider, but even 3.2v on the XMS3500 doesn't get me any higher. It just freezes, and even in memtest (DOS based) it just freezes.

With SETFSB I can gain about 3 MHZ in windows but anything higher than 285 (5:4) is a big no no...


Please help me doing this Vtt mod. I have read on page 1 about the mod , but it's not very clear to me....

[[SLC]Tachyon]

OK- got a 15W soldering iron, rosin solder, but I could only find a 10k pot.

My uncle does soldering on a regular basis- so he's coming over tomorrow to do the solderwork.

What should I set the 10k Radioshack Cermet Resistor to? Can I start it @ 0 (same as if the 0 Ohm resistor was there) and slowly crank it upwards, watching the voltage rise to where I need it in the BIOS?

TIA

[hipro5]

OK.........I'll post photos(as soon as I get my friends digicam once again) with Vmem(not the "style" we are talking about here) , VTT , Vagp (too difficult Modd) and Vchip(Vdd)..........

VTT , Vchip(Vdd) and Vagp needs some traces to be cut.............

[crotale]

Woohoo, it works!
But I was really scary, my IC7 didn't like be installed barebone.. (?) I've been having that trouble before... Does it have something to do with the grounding?
Anyway, it's in a case now, working really nice!

Originally posted by [SLC]Tachyon
What should I set the 10k Radioshack Cermet Resistor to? Can I start it @ 0 (same as if the 0 Ohm resistor was there) and slowly crank it upwards, watching the voltage rise to where I need it in the BIOS?

That's what I did, works fine.
Double check to make sure you got a low resistance to start with, I don't really know what happens with alot of resistance, and I've asked here, but no one have replied to that yet...

BTW, I have a 22-turn vr, and even then it doesn't take much to adjust it, be careful!

[[SLC]Tachyon]

OK, we removed the R50 resistor *perfectly*- solder pads were in perfect condition.

*BUT*- when trying to attatch one of the wires to the top solder pad- it took a few attemps and in the process it seems we have melted the solder pad off.

Now is there any way in hell we can correct this? Can we just drop some solder onto where that pad was and will it be usable? how bout using some conductive paint and tehn just glueing the wires to the solder pads????

Any help appreciated!!!

[Vatt]

Hi,
excuse my English, I have just been to our town carnaval, and had a few to many

Even if you f*ck up the pads, it seems you can still use the legs on the chip.

It is in this very thread , it seems that an xtreme member has found the way to make it work.
(ofcourse you will only need it when you habe FUBAR'ed the original pads.)
I will find the exact post and edit it into this message,

Boy, it's difficult to type when you've had a few to many beers...

[NyCUndaGrounD]

Originally posted by [SLC]Tachyon
OK, we removed the R50 resistor *perfectly*- solder pads were in perfect condition.

*BUT*- when trying to attatch one of the wires to the top solder pad- it took a few attemps and in the process it seems we have melted the solder pad off.

Now is there any way in hell we can correct this? Can we just drop some solder onto where that pad was and will it be usable? how bout using some conductive paint and tehn just glueing the wires to the solder pads????

Any help appreciated!!!

I have fuggered up the pads too myself, and u can try adding some solder to the pads to revive it. That helped me save my board.

[[SLC]Tachyon]

Originally posted by NyCUndaGrounD
I have fuggered up the pads too myself, and u can try adding some solder to the pads to revive it. That helped me save my board.

The problem is- that for the top pad- there really is no pad there at all anymore. All I can see is the PCB- no metal is left there whatsoever. :/

[NyCUndaGrounD]

O in that case u might be screwed............... I mean even if the pads are gone there is still some kind of metal contact there in that space.

In my pic, there should be some kind of contact, if not how would the board get information from those contact pads.