unlocking problem!

[peonyu]

Hi, Im having big problems with unlocking, heres the deal: I got the chip in, checked the internet on how to do it, then got super glue, a razor, some tape and bought some aluminum paint at walmart.

Now ok I started by filling in the pits, all was good it appeared! Then I chizzeled the paint down abit [without tape over the contacts tho..], then i applied the aluminum paint and cut between the bridges to ensure no overlapping. After I attempted to boot it up...and nothing! I get no video no beeps the system gets power and the fan runs but I just get no screen! So I take the cpu out and examine it some more, and I noticed that between the first two contacts i razored to hard cause I could see a little twinkilng of copper..could this be grounding my connection or something? I was thinking of maybe supergluing over the copper to ground it but I want to ask some of the pros on what to do cause this is my first time and I hope the chip isnt dead!! please give me some ideas!

 

[ejsmith2]

I dunno about the aluminium paint.

But about the only thing you can do is grab a multimeter, and use a really fine point probe, and see if anything is grounded or shorted.

"I personally think filesystems should be rewritten from scratch every 5 years..." --- Hans Reiser

 

[peonyu]

bump

 

[Lonemagi]

Like ejsmith2 said, test it. Thats how you know if you made contact, or grounded the pins.

--
"To be prepared for war is one of the most effectual means of preserving peace."
"There is nothing so likely to produce peace as to be well prepared to meet the enemy."
-George Washington

 

[svol]

Yes thebchance is very high that you grounded one of the bridges which can give very weird results. You can check the resistance between the ground (large dots and AMD logo are grounded) and the right pins of the bridges (the ones closest to the end of the CPU) with a multi-meter. Their resistance should be above 100 Ohm and not 0 Ohm.

My dual-PSU PC is so powerfull that the neighbourhood dims when I turn it on

 

[peonyu]

ok I tried it and im getting equal resistence from both the AMD logo and the pins, around or above 100 OHMs. Im not to familiar with AMD chips so assuming the logo shouldnt give off that kind of resistence right? And if it is the bridges that are grounded can i just put a dot of super glue over the enire L1 bridge area to ground out any resistence and be ok or would that just mess up another pin? Heres a pic I dont know if you can get anything from it:

<IMG SRC="http://www.imahosting.com/sigs/cpul1.jpg">
the scratches are inbetween the last 3 pairs to the right, 1 scratch between each pair

thanks in advance

edit, Toms doesnt support html..

 

[svol]

The scratches should be directly connected with the ground so if the unlocking paint touches it the resistance should be around 0 Ohm. Looks like the bridges aren't touching it... that could mean that you've cut to deep and reached the internal data channels.

Maybe you can try to remove the paint and see if the CPU will boot then. Or try the CPU in another PC to see if it isn't the motherboard not accepting the unlcoked CPU.

My dual-PSU PC is so powerfull that the neighbourhood dims when I turn it on

 

[peonyu]

Actually upon closer inspection I am getting 100 OHMs on the bottom row of conductive dots, including the arrow, but am getting 0 OHMs on top 5 dots. Please excuse my newbness here, but if the top five dots are getting 0 that means that theres a ground right, and those dots arent supposed to be grounded right? So should the bottom row get 0 OHMs as well, or should the bottom be at 100 and the top at 0, also im not trying to unlock anymore, I just want my cpu working normal again! Thanks in advance, your really helping me here!

edit; the razor cuts with the copper showing also gets about 100 OHMs

<P ID="edit"><FONT SIZE=-1><EM>Edited by peonyu on 03/13/03 04:24 PM.</EM></FONT></P>

 

[skligmund]

to get a URL to work,
type inside of brackets [] URL
type html address
type inide [] /URL

should look like this [ URL]http://www.com[/URL ]
(without the spaces)

Water cooling is for the weak. Get liquid nitrogen.

 

[phsstpok]

This article might help you understand if you can follow it. (I can't. LOL!)

<A HREF="http://www.beachlink.com/candjac/TbredDecode0.htm#Table" target="_new">http://www.beachlink.com/candjac/TbredDecode0.htm#Table</A>

and especially this part, "UNDERSTANDING THE CIRCUITS USED TO SET AND RESET MULTIPLIER SIGNALS"

<A HREF="http://www.beachlink.com/candjac/ResLogic.htm" target="_new">http://www.beachlink.com/candjac/ResLogic.htm</A>

I think these statements explain your problem.

<i>And a direct connection from Vcore to the grounded Signal Bus would only create a short circuit. So direct grounds, such as accidentally connecting L1 bridges to surface grounds or sub-surface ground plane when closing the L1s, not only set LOs but also prevent resets to HI by mobo/bios if and as needed, with unexpected Multipliers as a consequence.</i>

I think you cut into the sub-surface and your aluminum paint grounded one of the L1 points to it.



<b>99% is great, unless you are talking about system stability</b>

 

[svol]

If unlocked correctly the top row should be the same as the bottom row... looks like something went wrong. The last time I measured a fully unlocked CPU the top row had endless resistance and the lower row something like 150 Ohm (don't remember exactly). After unlock the top row should have the same resistance as the bottom row as you connect them with eachother.

The best thing you can do is completely remove the CPU from the conductive paint you used. This can possibly be done with rubbing alcohol depending on the paint type.

My dual-PSU PC is so powerfull that the neighbourhood dims when I turn it on