8RDA+ high Vcore shutdown fix!

[Pita^Norf]

Originally posted by ctgilles
that's something I don't get. Desoldering the resistor would normaly be the same as cutting the trace or lifting the pin on the IC. Since you also open the cirquit by doing that.

I think the wisdom behind that is that it is much easier to undo the mod.

[bias_hjorth]

I wonder if you could do the same thing on the Nf7-S..

anyone??

[Sick0fthelies]

I skimmed through this thread and only found a picture that says "you can see where the resistor was removed" but i cant tell how it was removed. If that is too hard i will just cut the brdige but with Just a slight cut with a knife or a major cut with say wire cutters?

[Swiftech]

Hi guys,

I have been using this 8RDA+ Rev 1.1 for some time and it's being doing great. I'm running on an ANTEC 430W PSU but whenever I set the Vcore to 1.875 and above, it hangs and during the RAM checking and doesn't get into the BIOS correctly as well

This is pretty weird because it shouldn't act so, am I right ? My question is, has anyone with the same motherboard as I do have such problems ? I understand that this motherboard has an 'OVP' so here I am with it eh ? *lol*

Appreciate your feedback guys and does this mod help ?

[tomiapk]

I have exactly same problem wit the Vcore voltage rise. I just cant get it above 1.850. Computer simply doesnt restart.

Epox support sayd the problem must be in CPU...

What do you think and have you found a cure????

Regards!!!

-Tomi-


(My cpu is xp 2500+ barton)

[ryssen]

I have a Rev 2,1 and the chip is CS 5301 and the resistor is not the same, any one have this board?

[tomiapk]

Yep!!!

I had same cs5301

[Swiftech]

I guess theres no cure for this problem and I don't think it's the CPU. Why isn't the NF7 users having problems like us !

I need more Vcore !!!

[R.Rabbit]

i've got a rev 2.1 also, but with my mobo i cant run stabe for days(probably years) at 1.925vcore but once i go to 1.95 or above, it reboots flashes bios then nothing, complete dead mobo, but once i just reset the cmos everthings fine again, i emailed epox and they said exactly that this is the problem(but of course not how to fix it) but then i was pointed in this direction by another 8rda+ rev 2.1 owner on the ocforums, what im wondering is if i should/could try this on my mobo or is the layout just too different...

[Holst]

Originally posted by bias_hjorth
I wonder if you could do the same thing on the Nf7-S..

anyone??

I allready designed such a mod... have a search around.

As for the 8rda+ rev2 problems with the CS5301 its a problem with the boards design... one i was unable to solve when I had one... and I no longer have it.

[bias_hjorth]

Yes i´v seen holst. Nice job btw.
But its a long time ago now so I actually dont need it anymore ;-)

[R.Rabbit]

Originally posted by Holst
As for the 8rda+ rev2 problems with the CS5301 its a problem with the boards design... one i was unable to solve when I had one... and I no longer have it.

holst, so how did you solve it?

[Holst]

unable to solve it im afraid.

I think you may be able to get it to work if you set the vreg up manually (I.e. lift the pins and set the resistor devider and FID yourself) although this is someewhat complex.

Its a shame I got rid of my rev2 board as I quite fancy having another play with it now.

[R.Rabbit]

ahh that sucks maybe i'll do the pin voltmod if i bored

[The Byter]

at more then 1.9v i get boots... not shutdowns...even at stock speed.

will it help me ?

[trodas]

Hello, guys there

I was long time waiting to use this board and the mentioned CPU, so I borrowed it to my friend, who is mad overclocker, to test it/clock it and basicaly burn it in He managed 200 x 11 working, witch gives 3200+ rating...

...after almost half a year I was happy with my fanless watercooling, so I get my mobo & cpu expecting to get 3200+ speeds. For memory I used very expensive and kinda good Mushkin Hi-Performance 512MB PC3200 dimms, running at 2-2-2-11 and 216Mhz w/o a glitch (memtest 3.1a)
The problem - crash when running Prime95/folding as stress CPU testing Now I should mention that the crash is weird. Nothing like freeze, or such, just the mobo shut down completely, leaving me with black screen, weird sound (i use no fan, where the hell the sound getting?!) and the HDD let is light on
I should note I using perfectly good Enermax 431W CPU (I borrowed the one to my friend too, so the package are almost complete )
My ways how to fix the problem are these:
First I added a heatsink on the Mosfets:
http://ax2.wz.cz/show.php?p=wc&id=85&c=7&d=1&v=v2
No help.
Then I made MUCH better (with cooper bottom! http://ax2.wz.cz/show.php?p=wc&id=60&c=7&d=1&v=v2 ) NB cooling that the standard one, I even laped the NB!!! ( http://ax2.wz.cz/show.php?p=wc&id=52&c=7&d=1&v=v2 ) - see my monster heatsink!
http://ax2.wz.cz/show.php?p=wc&id=62&c=7&d=1&v=v2
Unsucesfull again.
Even as light change over 2100Mhz as 202 x 10.5 is unstable, in 3 days time (folding all the time) I got the weird crash...
Yesterday I saw this tread and the mod, so at today morning I did it, together with raising the voltage of NB from 1.58V to 1.81V (using 8RDA+ rev.1.1 - so no changes of chipset voltage in bios, just mod )
See there:
http://ax2.wz.cz/show.php?p=wc&id=73&c=7&d=1&v=v2

and there:
http://ax2.wz.cz/show.php?p=wc&id=74&c=7&d=1&v=v2

Now I has a very high expectations that the damn cpu will go to the 3200+ finally. I used 210x10.5 and 1.9Vcore (1.81Vchipset and 2.9Vdimms), starting at 31 degrees (ArcticSilver5 very thin layer applied and the CPU block tightened pretty hard:
http://ax2.wz.cz/show.php?p=wc&id=65&c=7&d=1&v=v2 ...with even tweakmonster CPUI shim) it made slowly to somehing like 50 degrees and then Prime95 report calculation error. Pissed I set 1.95Vcore (showing 1.98V in SpeedFan) and the calculations seems working for about 20minutes, and then - when I was almost certain I made it, the bloodyf*cking things again crashed as weirdly as it do all the time - everything blank, weird sound that fading quickly and that it is
I almost want to write Spartacus a big thanks for the mod, but seems that my problem are different
As one can probably see from the pictures/setup and equipment, Im far from being HW noob, I modding everything and Im not affraid of any mod (next to do - Vdimm voltage mod to get over 2.9V in attempt to get about 250Mhz FSB ), however this one pissing me hard
Any chances this is caused by the CPU?
I mean - is possible that older unlocked Barton with Vcore 1.65 and 166x11 (1826Mhz) default clock can't suddently manage 2200Mhz and 2100Mhz are the absolute limit?
I can't believe that!

I should note I believed with the shutdown, because the crash looking like shutdown (nothing except power switch holding for 4 sec reacting then), but since I desoldered the resistor, then It can't be...?
In bios I even have disabled the thermal shutdown, BTW.

With help of one nice friend in US i obtained Barton mobile 2500+, default Vcore 1.45V and most ppls reporting at 1.8V about 2500-2600Mhz possible, so I waiting for it like for a miracle and hoping that it somewhat fixes my problems, however the post from US to middle of Europe seems take forever and Im getting more and more pissed at the current CPU
I even tried to burn the f*cker with 2.2V Vcore, but the crap survived it...

I should note that this is before I did the Spartacus highVcore shutoff fix and when I keep the clock bellow 2100, the machine never fail
2.2V is a lot, IMHO
But it still not allow the damn CPU to get over 2100Mhz...
...I did not care about it too much - it's overclocking, your mileage may vary - when Zbyna not reported it working at 3200+... I see this a next-to-impossible now, witch of course leading me to slight suspicion, if he tested the 3200+ properly, or what the hell... I complained to him and he told me, that 3200+ it manage only with VERY VERY nice thin layer of AS3 he used this time... He even say that when he repasted it, it not made 3200+ again...!
But I believe that Im better and can do nicer layer. Besides, I used AS5, witch is supposed to be better...?

Hell, I don't know. If anyone there have any clues, please write down what is in your mind, even it can sounds pretty crazy, Im open to try it now, because I have spare older AXP 2200 and AXP 2600, so I can but this sucker down completely
Not to mention I avaiting the AXP mobile each day and waiting and pray for it like for a miracle, so when I exchange the CPUs, I hope I will be able to go at least to 2500Mhz (210x12 sounds cool to me ) ...
...witch then show precisely where the problem is...

My last mad idea is, that cooling matter for this CPU most, so try to burn my todays assembly of AS5 with 2.2V and 2100MHz Pime95 torture and then let it go to about 1.875 or 1.850 to see, if the weird crash happens again at 2200Mhz (200x11) or not

Better ideas?

[trodas]

...after reading well the whole tread I see that I should also note that my bios is the latest, 02/12/2004

Hell, if this is the CPU - should not the "end" be BSOF?
(BlueScreenOfDeath)

...this black screen and reaction only to PowerSwitch - after 4sec it starting again - seems to me, that something triggers the shut-off protection
I would like to know what and mainly how to disable it

In extreme case, I have one 2200AXP unlocked and one new, superlocked 2600 Both 256k cache cores, tough...

[trodas]

I did some other checkings there. First I tried relaxed timings on memory. 3-4-4-8 did not boot, 2.5-4-4-8 did
Then I tried how far I can go with FSB.

245Mhz - freeze at half of post screen
240Mhz - large errors with IDE during boot (ZIP)
235Mhz - ocassional errors with IDE (ZIP)
230Mhz - seems booting fine

That is intresting. Then I tighten the memory timings back to the fastest 2-2-2-11 and tried where I can pass memtest with what FSB - 216Mhz max. Not so bad for HI-Performance Mushkins with BH6 chips and 2.9Vdimm voltage only

So, what now? Now I tried running 215 x 9.5 - this is even bellow tested and rock-stable 210Mhz for my CPU, so it have to work, yea? Yea, but nope. After 4 hours of CoD playing it started crashing miserably
Since CPU temp are bellow 60 degrees then I believe that its more likely any strange limitation that did not allow my CPU work well with over 200Mhz FSB, no matter if the FSB increase are only VERY light, like 202Mhz instead of 200Mhz
Anyone with similar problems out there...?
Or at least ever heard that certain CPU can be "alergic" to over 200Mhz FSB?

[STEvil]

I had a rev 1.1 board, took me 2.2vdd to be able to bench at 233fsb.. posted up to 245fsb.

never was stable over 200mhz fsb really.

[trodas]

That sounds too bad...

But I know few guys, who pushed the board (stable?) over 200Mhz and it worked for them

Not to mention that there is lack of alternatives and I already like the mobo a lot - instead of Abit NF7 it did not need modded bios to make multipliers 10 and 10.5 working, their voltage and temp readings are fine (unlike the Abit) and so far, I modded it too much, as you can see from pictures, so I did not want to give up on it

2.2V on chipset sounds extreme to me, quite frankly I thinked about removing the chipset voltage mod, because it did not helping me reach more Mhz stable

I also can bench at very high frequencies, however the mobo, after all the changes, still tend to lockup weirdly, as described. My unfortunate attempts to drive my Barton over 3000+ ratings still failing. Finally I got BSOF when I trying absolutely stable and tested 3000+ (200x10.5) with 1.875V
Yes, finally a crash what I can undersntad Then I tried 1.9V and 3200+ (200x11) and the CPU show errors at Prime till the machine finally locked up (no mouse move, picture on screen, cool - another lock from overclocked CPU I can understand! ). So, I cranked the voltage to 1.95V - still errors. 2V - after 15min w/o a glitch, the strange shutdown come here again

I definitively don't like this. If anyone got any clue, what I should try to fix this, I try it

At this point I starting suspecting that any other protection kick-in Maybe cuting even more wires/desoldering more resistors?

Hmmm, for start I should add passive cooler, just like I have on the mosfets, on the protection chip next to them and then I thinked about cooling this little chip:
http://homepage.ntlworld.com/bubo/Li...ttle_chip.html

...as anyone can see, it getting pretty hot and maybe... Just maybe did not like my voltage experiemnts... The temperature is still about 63 degrees at MAX, even with the 2V and 200x11, so it should not kick in...

Any other suggestions, please, skiled dudes...?

[trodas]

I did another test today. With a perfectly stable 200x10.5 (3000+ rating) clock, I just cranked up the Vcore voltage from 1.9V (stable) to 2.0Vcore - unknown
Within 15min, the board weirdly crash - black screen, drive activity on, weird sound

Now I leaved it about 10minutes, and powered it on back again. In post I saw my CPU temp 37 degrees.
Way too low for 2.0Vcore, right?

So, now it's obvious - the mobo shut down, thats why the temp is so low (using watercooling and watrcooling was running all the time ), so weird crash = shutdown.
And yep, the problem is too high Vcore.
Hmmm.

Damn, I removed the resistor and still...?
When I get my mobile Barton 2500+ finaly, I definitively try remove the pin completely and add cooling as well

Any other ideas, friends, please...?

[fabbro76]

Hi guys,
i bought an Epox 8RDA3+ Rev 3.1 and I have a problem.....the system hangs when I set the vcore over 2V
Exactly the systems doesn't boot over 2V (example 2,025 or higher)

How can I do?

I don't know if the trace cutting for the 8RDA+ works perfectly on my motherboard....

Any suggestions?


Thank You very much and....sorry for my English :smileysex

[trodas]

Aaaaaagh, I made it!
The suggestion about "just removing/desoldering the resistor" is bad. At least on v1.1 of the Epox 8RDA+ mobo it did NOT working!!! When I desoldered the chip pin and lifted it to the air, the shutdown is gone!
Oh, yes, I added heatsink on the chip too as well as I added one Tweakmonster BGRA ramsink on the "small chip" very close to the cpu:
http://homepage.ntlworld.com/bubo/Li...ttle_chip.html

So far, this working well and I now got Prime95 as well as folding stable at 2.0Vcore (reporting 2.02V) and my oldie Barton managed the 3200+ speed with bellow 65 degrees temps now - I used specialy made cooper shim, because it seems helping me very much with temps

So far, great solution, thanks Spartacus!!!!


STEvil - well, I did not tried yet push the FSB more that 200Mhz (being happy that the 200x11 finally working rock-stable ) but I soon will try this
I know personally ppls, who manage more that 200Mhz from this great mobo, 215+ mostly w/o big problems.
Dunno then, we simply see

...even at 200Mhz "only" it seems to be so f*cking fast