• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

8800 series failures - red lines

Associate
Joined
4 Dec 2008
Posts
1,812
I have reflowed quite a few graphics cards aside from these two.

I have two cards here, an 8800GT and an 8800GTX. Both suffer red lines down the screen and red dots on POST, but still boot and work (albeit on low resolution graphics).

Neither are responding to repeated reflows, one was directed at the VRAM, one was directed at the core, and the last one was directed at both.

Does anybody have this problem, or did have this problem? Did you ever find any way of fixing it?

I've read of other people doing the oven trick on the same problem and succeeding, but these two just aren't picking up. I'm going to upload a video now.

Cheers
 
I would say they're knackered.

The oven trick to reflow the solder seems to work once or twice (when it does work).

If you've done repeated reflows with no result then I think they've had it.
 
The problem must lie somewhere else on the card, I'm sure of it. Either that or the core itself has had it and not the joints.

I've reflowed 4 other cards with artifacts and they worked first time and still today, and 'tis not the oven trick, I'm using a reflow gun and flux.

It's just this odd red line problem. I've tried reflashing BIOSes with modded clocks/volts to no avail. Though a new BIOS has made the fan and sensors work properly.

There's a large amount of cards failing this way.
 
Bump, anyone got any ideas? Even risky ones, the card only cost me £7 so if it gets killed it's nothing lost really. I buy a lot of these things and repair them.

I found it odd how it's driving PCI-E X16 2.0 128bit @ 8x 32bit, tracing the PCI-E bandwidth connections takes me back to a set of decoupling caps on the back of the core. Might try a whole-board reflow, I'm 99.9% certain the problem doesn't rest with the core or vram.
 
MMmmmm yes i always associated that with the physical memory being faulty.

Interested in what you are doing tho :) nice applied knowledge and skills there.
 
Thought so, just usually it's the BGA package that's gone, not the chip itself :p

Think I may have found the problem, took a look at the edges of the VRAM, on the 4th chip from the left at the top side there are two odd looking balls of solder. It's hard to tell because they're so small, but either there's not enough solder to connect to the board, or they're not centered properly. Either that or a pad's come off the chip itself, if that's the case, the thing's dead.

Might reflow again putting a small amount of pressure on the chip in question while I do it. Problem being, there've been 3 attempts by me already, one on the core, one on the vram, and one on the entire board... if this doesn't work, I'll scrap it.
 
Back
Top Bottom