how to increase CHIPSET voltage?

[Guest]

Dear experts

My A7V default chipet voltage the VIO only regsiter 3.4V when its supposed to be 3.56V

when I jumper it to 3.69V it only came out 3.59V

I am thinking of cutting one of the ATX 3.3V supply and stick-in one of the 5V supply into it.

Which 3.3V should I cut? what other components are connected to the path together with the chipset? i.e. AGP, CPU I/O etc, is there any other?

Of course I am not going to stick-in all 5V, will use a (diode) of some kind to reduce it to say 4V or 3.8V then stick-it-in.

any help appreciated! thanks !


Best regards
cx5

 

[Guest]

Since the VIO is adjustable, it suggests that it isn't using the 3.3V ATX pin at all, but through a regulator.

<A HREF="http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=626527#626527" target="_new">http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=626527#626527</A>

Hope the above helps!

 

[Guest]

I was hoping you would be the one who replied!

Thanks for the bad news, I was praying that it wan't taking it from a mobo regulator but direct, seems my wish did not come true 🙁

If so, your advice using resistor on R something is the alternative like toms tips on the A7V CPU up-voltage

Very well, I will attempt is after my last examinations tomorrow.

I will stick in a fix value resistor (say 56k as adviced) + a trimmer, I'm sure you know what I am trying to do!

As for toms tips I will do the same, perhaps with a higher resister value because I am not so brave, just hope the larger resistor doesn't make the voltage go higher instead or else hahaha!!!

Well, good luck to me for the exam and good luck to me not blowing up the mobo and cpu.

Best regards
cx5

 

[Guest]

It seems to me that a regulator would be easier to deal with -- by playing around with the 'feedback' pin on the regulator chip.

Yes, this is very similar to Tom's VCore mod...it's just being done to another regulator =)

Good luck to you, and be careful!
(and don't forget to report how well this mod works!

 

[Guest]

I have another question?

When I add the 56k resistor, + 20k trimmer how safe do you think is it to vary the voltage after the computer is turned on and running.

I am interested to see what the voltage level will change when the total resistance change!

Actually all I want to know is that if the resistance is more than 56k will the voltage increase? or decrease?



Best regards
cx5

 

[Guest]

I really don't know how safe that will be...

but... When resistance drops, voltage rises.
when resistance increases, voltage drops.

 

[Guest]

<A HREF="http://www.stuffedcow.wox.org" target="_new">A blurry scanned picture</A>.
Hope this might help

 

[Guest]

Dear Henry

CPU
ori max mod
650 900, 963 Mhz, still finding the max max
1.6 1.82 1.92 voltage

all my modified voltage is less than toms value. 🙁

As you can see my A7V does not produce high voltage in the first place 🙁 may be because I shorted it once when using a metal to start the PC without the switch 🙁 sparked and smoked abit. Thank god its still working

-CHIPSET 1-2, 2-3, 3-4 (jumper pin numbers)
--manual 3.36 3.56 3.69
detected 3.2, 3.4, 3.59

again as you can see my A7V have lower voltage than normal

3.72 3.92 4.02 voltage after modification
109, 110, 110, FSB

FSB 110 failed on both 3.92V and 4.02V
(Red flashing in BIOS)

I am using a dual 20k ohm trimmer in series to give a total of 40k instead of 56k because I know my A7V is already producing less V and I wanted more.

Seems like the chipset reached its max, will find more info about the actual max, lets hope it can take up to 4.3V, do you know the max?

Will post picture to you later.

Oh yes, I found out that the VIO jumper pin 3 is connected to the point you recommend, so I solder to the base of pin 3. I'm sure you know why, because everything is so close together.

Also I found 2 grounds point. In the picture you post, on the left, there is a empty transistor 3 legged. The pins on both sides are ground not the middle.

You can take a multimeter and measure it, also the VIO jumper 3 and verify.

Another quesiton which made me confused from A7V manual

when I increase the VIO, I up everything for CHIPSET, AGP, PCI, CPU I/O, DRAM etc

does it mean only on the input output stage or all actuall hardware?

my meaning of input output only is voltage increase not to the actual hardware but bus transfer.

or else it is the actual hardware, if so this means the my SDram is being fried with higher voltage, AGP card as well, sound card as well etc.

Can you clear me up?

Best regards
cx5

 

[Guest]

I'm glad to see it worked

I believe the max voltage is 5V (since the converter converts 5 V to 3.x V). I'm not certain if the KT133 will take that high though.

About the IO voltage... I think it affects the power of those components. So your RAM would probably be running at 4V as well, along with the other components listed.
I'm not certain if that is a big problem, but check on the heat from RAM, etc. just to be sure...

 

[phsstpok]

I'm sorry, I'm not technical enough to do what you are attempting. (I can just barely understand it). However, I was just wondering what your goal is. Are you do all this so that you can increase the FSB?

 

[Guest]

yes, pretty much just to overclock FSB a bit more than ~110MHz...

It's a modification very similar to the one used to raise VCore, but done to VIO instead.

 

[Guest]

Well, there you have it, Henry answered you the question

My dream hope is to go FSB133Mhz and CPU 1000Mhz

Henry, the CPU is already running fine on FSB108 * 9 = 972Mhz

It is just the dam VIA CHIPSET not wanting to go higher.

My maddest bit is that FSB 117 will post and boot but windows will surely crash/hang. Even when BIOS performance set normal instead of optimal and Dram at 117Mhz also, lets just say everything in BIOS reduced, but no use.

My new question is, say heat is never a problem, as what voltage will the CPU or CHIPSET + RAM FRY?

simply due to voltage I mean, is there such scenario???

Best regards
cx5

 

[Guest]

I forgot to mention, the trimmer total of 40k, after connected to the mobo only got 5k probably due to other resistor paralleled with it.

Anyway, I didn't dare to turn the trimmer down until I consult you.

one things for sure I am not going to give it 5V

Ah if only the chipset is removable like the CPU then I can change it to VIA133 capable. so I don't have to worry if I fry it due to voltage

And the DRAM is worrying me, I don't want to fry so many thing at the same time. Frying chipset is bad enough, that means the whole mobo is gone, £125 man, and if my RAM goes as well, then [-peep-] me man bla bla bla crying 🙁~~~~~

somemore the AGP, PCI, bla bla bla, so many things connected to it.



Best regards
cx5

 

[Guest]

hmm... posting at 117MHz sounds promising...

I still haven't tried this yet... I fried my Athlon 800 a while back...My homemade water block melted off the processor because water circulation stopped.
(Perhaps it's the CPU? underclocking the CPU could be an option to test the FSB only... )

no, I have absolutely no idea how much voltage will fry if heat is not a problem.... The maximum voltage rated by VIA is 3.45V. And I believe someone took an Epox KT133a board to 185MHz at 4.2V, so 4.2 should still function on the KT133 too.

Removing the KT133 shouldn't be that hard...just put a lighter underneath it. But soldering a new BGA chip on there is an entirely different matter!

 

[Guest]

Burning the chipset with a lighter, hahahaha that was very funny and chear my day.

If I can get my hands on one of those new chipset, then I can ask my department, acoustics and electronic to use one of those fancy hot air gun, quick type, very hot to remove the BGA chip and put in the new one and reheat the entire thing.

But those chip are not for sale are they? 🙁

If only my dream mobo would come out

which is, socket to CPU, north, south bridge, ATA as well

and for graphics card, socketed GPU and ability to change RAM

but these things will never come out so dream on me.

Bye!

Best regards
cx5