Printable View
Well here are the pics I promised.
Infinity....RIP
http://homepage.ntlworld.com/dresuk03/Burned_PCI.jpg
9800....Long live (touch wood)
http://homepage.ntlworld.com/dresuk03/Burned_9800.jpg
Relay...
http://homepage.ntlworld.com/dresuk0..._relay6%5d.jpg
and finally the successful mod....
http://homepage.ntlworld.com/dresuk0...-droop-mod.jpg
mong
hy
i discover a very strange thing about nforce 2 whith more temperature on chipset it runs stable at 250fsb with less temp not stable,i test with other board(epox 8rda3+) and as the same problem,wath conclusion?chipset nforce 2 like temperatures?
please if possible can someone make de test?
i test it with a window open,after with the wiondow open but chipset covered to improve temperature.
Well, that would make sense on why my board doesnt run above 220 very easily.. huge copper heatsink, lapped nb, etc...
Would also answer why the one guy who burned the nb on his NF7 got better results after that.. I forget who that was, though.
Maybe i'll take my nb cooler OFF and let the sucker burn a bit ;)
The mod 's looking nice mong but the board :rolleyes: ;)
I got my LP B from rma today, and hey it's a taiwan one :D
I don't want to kill this board so I 'll wait for an infinity to come here (one of the two) and maybe mod this...
I afraid, you know? :(
I still have not got my antec tc 550 back from rma... all that rmaing sucks :mad:
P.S. I wanted to test the vdd mod with the board out so that I am able to measure the vdd mosfet (before sticking coolers on it)...
did anyone realize the voltage staying the same with 52k, 65k and 80k??? Weirdo :confused: always 1,95V...
R U sure you've got the right measuring point?Quote:
Originally posted by SAE
The mod 's looking nice mong but the board :rolleyes: ;)
I got my LP B from rma today, and hey it's a taiwan one :D
I don't want to kill this board so I 'll wait for an infinity to come here (one of the two) and maybe mod this...
I afraid, you know? :(
I still have not got my antec tc 550 back from rma... all that rmaing sucks :mad:
P.S. I wanted to test the vdd mod with the board out so that I am able to measure the vdd mosfet (before sticking coolers on it)...
did anyone realize the voltage staying the same with 52k, 65k and 80k??? Weirdo :confused: always 1,95V...
Always the same voltage seems kinda weird, doesn't it?:rolleyes:
I've used the same point as the NF7-S. (that soldering patch, where a mosfet could've been, upper leg.)
I know how you feel SAE,
Come on guys, we need some more volunteers to test the mods!! Anyone fancy PM Holst to ask him to take a look at Vlad's mods.
Vlad I dont doubt your work, but is always best to get a second opinion.
Come on peeps, get the soldering iron out and test the mod!!
mong
Hmm. I am not exactly that dumb I cannot recognize the measuring point... :rolleyes:Quote:
Originally posted by DivineEvil
R U sure you've got the right measuring point?
Always the same voltage seems kinda weird, doesn't it?:rolleyes:
I've used the same point as the NF7-S. (that soldering patch, where a mosfet could've been, upper leg.)
I now have been using the dfi design (3 infinity, now LP B) since late november. I know where to measure. It's by far the hottest mosfet on the board... it's the leg pointing south of the board.
The same point as NF7-S? That sounds a little weird though since both boards use a very different layout :confused:
What board are you referring to?
@mong
Thanks :D
Today got my LP B, and it's a taiwan one :thumbsup: (RMA lasting one day... they got it and exchanged it same day, yesterday).
Yipiiiee... (knock on wood it works... ;) )
Still have not had time to test it thouroughly ;)
I know u're not stupid or anything. :toast:
i've double checked my DFI and you're right.
I've used the upper soldering pad directly under the cap that's most left, under the AGPslot. (marked EC44)
(edit: oops it's most-RIGHT)
not 100% sure about the "upper part" but that's the location i used.
It's "kinda" the same location as the NF7-S ;)
sorry 'bout that.:rolleyes:
Never mind ;)Quote:
Originally posted by DivineEvil
I know u're not stupid or anything. :toast:
i've double checked my DFI and you're right.
I've used the upper soldering pad directly under the cap that's most left, under the AGPslot. (marked EC44)
(edit: oops it's most-RIGHT)
not 100% sure about the "upper part" but that's the location i used.
It's "kinda" the same location as the NF7-S ;)
sorry 'bout that.:rolleyes:
Hm well my LP B also died, for no reason. I really don't know how this happened. First it got unstable, then reboots after about 5 mins and then .. nothing anymore. No bootup!
so desoldering mods and RMA :rolleyes:
Now I'm waiting for my new DFI..
EDIT:
I had a Beta BIOS on it, willt this be a problem when RMAing ?
On the telephone some employee said: "You'll get a new one, but we'll check if you did some BIOS update or something."
But where's the sense when checking afterwards ??
e
Did you guys remove the resistors and replace them with ones that were 3.3k or did you leave them in place? Vlad Draculea can you post pics?
Sorry to hear your brd died Sai, i think DFI are getting strict on brds flashed on beta bios. I hope they dont mess u abt with RMA due to this.
Good to see you did the mod Vlad, I have my new Infinity here and it is alot worse than my old one :(
With my mobile barton i can run 2.6 ghz with 1.8volts prime stable but as soon as I start going over 1.85 and push up the mhz Im getting the reboots and instabilty whilst running prime.
Also 3dmark01 is alot worse now, I cant run a test at 237x11 (2607mhz) without being kicked to desktop.
I have a few questions abt the mod, with the Vdrop mod, how do we know if it is doing what it has been put in place for. We have to ue a 50K pot and make sure not to put zero resistance, but how can we tell when we need to stop turning the pot, is their something we have to look out for??
iceman2g, yeah I removed the resistors, if u go and re-read the instructions Vlad has posted he says you remove the three isen resistors.
mongoled
Anyone rmaed a board with beta bios on it ?Quote:
Originally posted by mongoled
Sorry to hear your brd died Sai, i think DFI are getting strict on brds flashed on beta bios. I hope they dont mess u abt with RMA due to this.
r
e
Well i got the mods in place on my new board, I couldnt hold of any longer waiting for other peeps to try this out!!
Things are looking good so far...touch wood.
Vlad, I need some more advice from you, Ive noticed that even though i have gained alot of stability i still find my voltages are jumping all over the place.
How do i need to use the vdrop mod to control this. Do i need to be increasing the resistance on the pot or decreasing it? The vdrop pot is set to 30k at the moment.
Also I am overvolting in windows :) even though my vcore and vsense pot are set at 50k, i havnt even touched those yet, too scared
;)
So to sum it up this is what i have, three 3.9kohm resistors replacing the three smd resistors on the mthbrd, vcore and vsense mod inplace and vdrop mod set at 30kohm.
mong
-EDIT-
doh, i better mention what i have got out of this so far
:)
http://homepage.ntlworld.com/dresuk0...z-super_pi.jpg
prime95 stable at 235x11.5 = 2708 @ 1.85volts (bios). Gonna let the mobo and cpu settle in at these settings for a few days, oh and most importantly (at least for me anyways) i have all fans set at 5volts = nice quiet very fast PC
:) :)
n
When u say 'decrease the drop pot' do u mean decrease the resistance, the ohms? Say from 30k to 20k (the value here isnt relevant)?
mong
g
Thanxs for the advice Vlad,
Im a bit perplexed at the momnet abt a problem im having. I would say that the setting sin my sig are stable. I have had prime running for a couple of hrs ,which i stopped and have just finished rendering a scene for a friend. The render has been for the last 16hrs, not one crash, so i think it safe to say that the PC is stable, but....
I keep losing the onboard lan, their are a few things I have in mind, so i will run it by u guys. First of all, the operating system i have installed is the same one i had with my last Infinity. I did a straight swap, what chance of this causing the LAn instability? Ive tried upping/lowering voltages it dosent help. Ive got the southbridged sinked with the nthbrd sink and all other hotspots sinked so i dont think heat is my enemy.
The ram has been tried and tested above 235mhz easy.
One thing which i have noticed is when the lan goes it will freeze the PC for a few seconds. Has anybody got any ideas on what is causing this.
Ive just installed a diff lan driver to see if that helps. Ive also tried threee bios so far with same problem 131, 121, and da alpha b02.
mong
Do use a pci nic ;)
It works... the onboard one sucks at high fsbs...
You found 3,9Kohm to be best? I don't know if I should risk losing my lp b?!?! ... it's very good cause I never could do memtest_windows errorfree longer than up to 13% in dual channel at any voltage/vdd/settings... now I can do it without errors over 2hours (quit test then) ... and I am very pleased...
It's undervolting badly though, but I was able to complete a superpi 32m at 258x11 fastest settings with 2-17 b02... but that was using 1,7V vdd and with 1,9V it errors on me quite quickly :confused: 3d action needs more vdd though :( That's weirdo. ;)
Erm... wanna try this too...
Sorted three smd resistors out (from dead nf7-s' ;) ) having a resistance of about 4,18kohm...
Would this be safe?
I hope so. I wanna replace the onboard ones with these without having to mess with cutting my sinks ;) SMD resistors are a lil harder to solder but I hope it works (often soldered smd resistors with all kinds of mods).
And would it be possible to start off with a 10k pot for vdrop???
Help please, cause I want to do this today :D
Hey SAE I here your call,
:)
First of all i didnt experiment, I just used 3.9kohm resistors. I think it would be ok to use the resistors you have, just read how Vlad explained the calculations he made.
Another thing I noticed is that just by using 3.9kohm resistors the vcore voltage on my mthbrd increased. I had all the other pots in place but they were all still set at 50kohms when i noticed the overvolting.
As for the LAN I think ive figured it out. I wouldnt say 235mhz on the fsb is high ;) not enuff for me to be losing my LAN. The only thing i did which seems to have made a difference is to take off the nthbrd heatsink from the southbridge and replace it but instead this time with thermal paste rather that thermal pad.
I havnt tested it enuff, but it has been stable all day today. Talking abt stable, my computer rebooted abt 20 minutes ago for no apparent reason
:(
I wasnt doing anything stressful just browsing the net. Its strange I had the PC rendering for 16hrs non stop with settings in sig, some gaming sessions AND some prime95 sessions and never crashed. Oh well ill never now what caused the crash
Id say a 10k pot would be good, I have my 50k pot turned 25 full round clockwise (30 turns takes it to 0 ohms) and my mobo is still alive...touch wood....
Strange thing is I havnt seen a drop in my vcore by doing this, something which Vlad said would happen. Maybe i need to turn my vdrop pot lower still??
Anyway hope this helps, come on I think u must have done the mod by now even without anyones advice, I know i couldnt wait myself
;)
Gluck
mong
Did the mod as you wrote ... before any answer given ;)
Only had three 4,66-4,68kohm smd resistors and used them...
But I was not give the opportunity to test for stability :(
Was in bios and already that was a torture to wait for... the prom lowered the temps so slowly that it lasted over 10minutes. In bios temps rose this quick I had to immediately shut down the rig :(
I think - cause it often had hiccoughs lately - my prom is leaking :mad:
Not enough refridgerant to cool the sucker down...
Gonna ask for help in the refridgeration forums... thanks for ya help anyway and good luck with ya rig :thumbsup:
I will follow the thread though ;)