Yeah but I bet it wasn't volts:DQuote:
Originally posted by Ragnarok
Well JC i hope u dont treat ur wife/gf like u treat ur mobo!.. :D
Printable View
Yeah but I bet it wasn't volts:DQuote:
Originally posted by Ragnarok
Well JC i hope u dont treat ur wife/gf like u treat ur mobo!.. :D
15000uF 6.3v are good for this job?
:D
I did it on 8RDA. i took 2 caps from A7V333 and another 4 from A7N8X. it booted fine and voltage rails were very stable for the first time i saw something like this on this POS board. However, when i set the fsb to 210 (it could do 217fsb at 1.8VDD before) It froze and now can't get past "0D" code.... friggin POS, i got one more mobo, so i'll re-do the mod later.
I'll probably do the mod this evening, so I'll let you guys know how it works out.Quote:
Originally posted by BrainStorm
Nohto, let us 8RDA owners know if it works or not!
sorry if this is a little off topic....
but after reading this thread i learned alot more about capacitors and thus my fear of them was reduced. So i took out my broken KT7A that had blew a capacitor like a year ago and i fixed it.
i just removed the popped cap, then soldered on some big fat brown cap from radioshack. My board is now running perfectly fine, i just need a HD for it and its back in action.
It is just incredible some of the things u learn to do thnx to the internet. As long as educated ppl like you guys keep sharing knowledge freely on forums, peasants like me who dont go to college still have hope to learn a few things.
cheers :toast:
Welcome to Xtreme Mister420
Well done!:D
:toast:
Hi to all rev2 owners!
Do u think that soldering those caps is the only way to use vcores over 2.1V?
I cannot use vcores higher than 2.1V at ANY FSB speed!!! :(
my rev2 will not boot at higher vcore than 2.08V at any cpu or fsb setting! I can set 133x10 @ 2.125V -> no boot.
also 210x12 @ 2.1V -> boot and completely prime95 stable for hours.
So I increased 3.3V, 5.0V and 12V lines on my Antec PSU with adjustable VRs.
Now the multimeter reads 3.53V, 5.25V and 12.02V directly from the PSU. BUT! No improvement here! :(
Still stuck @ 2.1V.
Do I really have to solder those caps on?
Hello everybody. This is my first post here, just wanted to say i have been looking forward to a modd like this to appear. Have myself a Abit NF7 and due to an little accident with me trying to Vdd mod it, it died. But anyway, i could run 217 fsb with a Vdd voltage of 1.7. And if this mod actually can make it possible for me to run higher fsb's, i'm gonna do it =) Now to the problem, which of the components listed on this page is suitable for this modd ?
http://www.elfa.se/elfa/produkter/en/2012731.htm
/siefer
Welcome to xtremesystems siefer! :toast:
Are you intending to replace the existing caps or solder them on the back? If the latter then the third one down (4700u 25v) is suitable. Note that those caps are designed for mounting directly on circuit boards, they have very short leads (you could always solder some wire).
I'm still waiting to see if this works for anyone else, like Hognert said above it might be a board specific thing. Three negatives so far...:(
That sucks...been able to re-do yet and did it help any?Quote:
Originally posted by QuadDamage
I did it on 8RDA. i took 2 caps from A7V333 and another 4 from A7N8X. it booted fine and voltage rails were very stable for the first time i saw something like this on this POS board. However, when i set the fsb to 210 (it could do 217fsb at 1.8VDD before) It froze and now can't get past "0D" code.... friggin POS, i got one more mobo, so i'll re-do the mod later.
Nohto, what about your results? Anything? Or can we take the lack of a post as bad news?Quote:
Originally posted by Nohto
I'll probably do the mod this evening, so I'll let you guys know how it works out.
What would you use to replace existing ones? 2nd one down?
Thanks:D
No, the top one. Existing caps are 3300u + 4700u (add capacitance in parallel) = 8000u, so 10000u should be OK (I think, somebody correct me if that's wrong :)).
30mm wide is way too big for replacements though, the existing caps are only about 10mm.
I haven't had a chance to do it yet. I am running 2 new 20A circuits for the computer station from the sub panel. I had to add a chamber inside my 5 cu ft frezer, because it was leaking a little, and I am exhausting my 8000 btu/hr air conditioner so it won't heat the room up.Quote:
Nohto, what about your results? Anything? Or can we take the lack of a post as bad news?
I am also adding duct system for the air cond. to go to the server and to the tower that has the Prom. in it. I had to run #3 to my subpanel that I installed in the garage because of the extra load the 60A circuit wouldn't be enough so I upped it to a 100A. Needless to say I've been a little busy.
Thanks:DQuote:
Originally posted by MrLavender
No, the top one. Existing caps are 3300u + 4700u (add capacitance in parallel) = 8000u, so 10000u should be OK (I think, somebody correct me if that's wrong :)).
30mm wide is way too big for replacements though, the existing caps are only about 10mm.
Damnation...sounds like you got quite a setup there. I'm just in my bedroom "computer room" upstairs. *lol*Quote:
Originally posted by Nohto
I haven't had a chance to do it yet. I am running 2 new 20A circuits for the computer station from the sub panel. I had to add a chamber inside my 5 cu ft frezer, because it was leaking a little, and I am exhausting my 8000 btu/hr air conditioner so it won't heat the room up.
I am also adding duct system for the air cond. to go to the server and to the tower that has the Prom. in it. I had to run #3 to my subpanel that I installed in the garage because of the extra load the 60A circuit wouldn't be enough so I upped it to a 100A. Needless to say I've been a little busy.
I have now tried the capacitor mod as well. I used caps rated at 4700 uF and 25 volts. They're approximately 16mm in diameter 33mm long. A little large for replacing on the board itself. They were all that was available locally. The electronics shop swindled from my pocket, approximately $2.80 US for each individual cap. Motherboard makers know what they are doing, when they use components that you can't source easily....:)
My method involved attaching wires to the caps and then soldering to the underside of the board. That way I was able to place the board in the case and have the caps showing at the top of the board. I tried to keep the wires as short as possible for my application but it did not help me.
My results were less than stellar. Infact I lost about 3mhz fsb speed. I run this board 24/7 at 233.86 fsb and seem to be able to hit into the low 240's but with a higher Vdd. This could be ram related or a chipset issue. The North Bridge is cooled by a Copper thin fin design Sockat A cooler with a 60mm Global Win fan with a "3DfxCool" sticker on there! Bit of free advertising for Bart at 1coolpc. :)
Initialy it did not seem less stable. However once you pushed your fsb to high and reset, the board did not seem to be able to come out of this state and increase the Vdd voltage again. Instead it made the annoying beeping and siren sounds.
I'm certain you all have seen that when you change settings in bios the board overides your original modded Vdd. And uses a low setting. Then increases it again when booting at your overclocked settings.
I didn't try soldering the actual caps without wires to the back of the board and that more than likely may have helped, but I'll leave it for now. Really I can't recommend anybody do the mod the way I tried however there may still be variables I didn't take into account such as capacitor quality etc.
Shot of cap next to a new and unused T-Bird 1.4 to give an idea of size.
[LORD EDIT: IN OTHER WORDS IT SUCKED BIG TIME] :stick:
MrLavender:
Thank you for your help. I was planning on just adding capacitors on the existing ones.
/siefer
Not all caps are the same people.
Bolting on just any cap with high fahrad capacity wont exactly help stability, and I guarantee wires will kill most gain in most applications.
Get low impedance high-grade caps... Panasonic, Rubycon... should do. This is not a smoke and mirrors mod. The only possible reason it could have a negative outcome is if the filtering capacity or quality has been reduced.
Solder in something that could be construed as an improvement, keep the leads short, and there should be a stability gain.
What can I say... Im actually starting to really disslike this board. Seems there are lots of different things preventing a high and stable FSB.
Adding the caps did two things on my board. I can use a 250MHz FSB at multi 5-11, and there is no longer any use in upping the Vdd over about 1,8V at any multi. Thats all fine and dandy but this board is still a pain in the ass...
At a higher multi the max FSB still goes down... I thought the caps fixed this but after further testing its clear that they didnt.
At 11,5 the max FSB I can even boot at is 236 and its very unstable
12 - max boot 225 (unstable).
12,5 - boot 216 (stable) - will not POST at 217.....
13 - boot about 213 (unstable)
14 - fails to post at 200MHz....
These are the exact "limits" i experienced before the caps mod.. only difference is that I now dont need a very high Vdd. So there is no doubt that after multi 11 some new problem kicks in and spoils everything... Tried all bios settings there is of course.. changing Vcore, Vmem, Vdd and so on.. nothing helps. Guess next time ill go epox ;).
edit: One more thing. My board actually perform better with the cpu being detected as a 166MHz model. Got my 1700+ detected as a 2700+ now and that is more stable then 1700+/2100+. But then again all these boards seem to suffer from different problems so...
Hognert: What's the top that cpu reach? All those multipliers and fsb's is about the same speed, from top:
2714
2700
2700
2769
2800
Have you ever considered that's about what that cpu ca handle? But then ofcourse, if you'd been up in the 3g'ish with that cpu I can understand it.
we need a 2.1 :(
what people need to realise is this mod is NOT going to help get any faster fsb if your memory wont physically go any faster in the first place. all this mod will do, assuming it does actually work, is remove the problem with the board itself refusing to go to higher fsb.
how do u tell? easy. if your max stable fsb on tightest 7/2/2/2 timings is the same as your max stable fsb on 7/3/3/3 timings then in theory your memory is NOT whats holding you back and this mod might well help. on the other hand if 7/3/3/3 gives you a lot more on fsb than fast timings, then your memory just wont go any higher and this mod is probably pretty useless to you. The theory behind this is if your mb limited and NOT memory limited, then timings will not make any difference to how high you can get since memory isnt whats limiting the fsb. Thats my theory on it anyway.
i can do 223fsb on 4/2/2/2 and 223fsb on 8/3/3/3. upping vdd doesnt help much, so im assuming this mod should help me gain more fsb because it doesnt apear to be my memory limiting it.
Ewok
my rev 1.2 board would do 222 cas2 , 250 cas2.5 slow timings.
this rev 2.0 board goes as far as I want on cas2 ... 230, 233, 236 with lockups from NB ..... I can relax the timings, it will NOT do 1 MHz more than with max timings.... yet the mod yielded me absolutely nothing, so that's another theory shot to pieces.
doesnt mean my theory is wrong, especially as you only used 2200uf caps and hognert used ones more than twice that. just wanted to point out that if it doesnt gain anything, people should make sure their memory isnt maxed out before they can say the mod didnt work.
obviously your memory will go higher so maybe it needs higher quality/capacity caps? or maybe it was the wire you used taking away any benefit they might have given?
certainly doesnt look like its helping many though, other than hognert only 1 other person seems to think its helped and the same person also experienced it making things worse so something strange is going on. i dont think ill bother with it anyway, from the success rates so far i dont think its worth the hassle trying it, especialy as knowing my luck ill just fry the board.
speaking of which i fried something on a fan header when i was trying to measure voltages and i touched 2 pins at once with the voltmeter, it sparked but nothing else happened. cant see anything wrong with it, but plugging a fan into it no longer works, im wondering if i fried the tiny cap thats next to the fan header? it LOOKS ok though, so maybe something else is fried. board works fine still, other than being stuck at 223fsb.
maybe you should try sticking the caps on the ones near the mosfets and see if that helps instead?
I dont really know how far the cpu can go since Ive only tested it on this somewhat unrealiable board :/, It ran 3dmark at 2780MHz though and since 11*250 boots up fine I really cant see why the ****ing board refuses to even post at 12,5*217+.Quote:
Originally posted by Marquzz
Hognert: What's the top that cpu reach? All those multipliers and fsb's is about the same speed, from top:
2714
2700
2700
2769
2800
Have you ever considered that's about what that cpu ca handle? But then ofcourse, if you'd been up in the 3g'ish with that cpu I can understand it.
Messed around with the board the whole bloody day today.. didnt learn much... Its like the board actually got a mind of its own..
Why do 250MHz easy as pie to multi 11 and then all the sudden turn into a pile of crap?... Sounds bios related to me...
Cant figure out what more to do with the board atm.. guess ill just wait for a proper bios realese, v11 sucks pretty bad.