kingpincooling.com

Go Back   kingpincooling.com > Overclocking & Modding > TiN's Dungeon

TiN's Dungeon The mad engineer's home at KPC.

Reply
 
Thread Tools Search this Thread Display Modes
  #111  
Old 11-08-2014, 11:41 AM
Masterchief79 Masterchief79 is offline
Ambient
 
Join Date: Jul 2014
Posts: 15
Default

Some news and new pics, if someone is interested.

With an eVc, I got the zombie completely working. Tested my 8800GTS with 1,46V @ 720MHz GPU in 03 Nature, without any issues. The 2900XT PCB gets warm, but it's ok. These cards worked with 100-120C VRM temp for years with an AC Extreme, which almost didn't cool the VRM at all


Here you can see the zombies final condition, soldered some extra cables and did some other mods. Just tried to bench it with Single Stage, unfortunately I got artifacts in the BIOS/Bootscreen and regular crashes when trying to load the driver. I'm guessing the memory is defective, or just doesn't like low temperatures.
It's also possible that I did something wrong when adding more cables, if you could just have a look at my work and tell me if you spot something

Anways, it was just a test, and I'm pretty amazed by the results :] Now I just have to solder this bad boy onto a working card And also, I'm confident I can turn my defective 5870 Lightning into a zombie aswell.


Pics:

Reply With Quote
  #112  
Old 11-17-2014, 05:53 AM
Masterchief79 Masterchief79 is offline
Ambient
 
Join Date: Jul 2014
Posts: 15
Default

I have a question: I soldered the zombie as shown above onto my 8800 Ultra.
Now, it works well, and the card runs 3D Mark 03 Nature and everything without errors at 1,3V.

However, when I bump then voltage up to 1,4V, the PC just shuts down instantly at load without bluescreen or anything. PSU is 1000W BQ. Do I have to adjust OCP/OVP on the zombo unit or is the error in my soldering? Maybe more and thicker wires? Voltage drop is about 10mV from zombie unit to ultra (at full load).

Last edited by Masterchief79; 11-17-2014 at 06:00 AM.
Reply With Quote
  #113  
Old 11-17-2014, 08:21 PM
TiN's Avatar
TiN TiN is offline
Volt-Nut :)
 
Join Date: Jan 2010
Location: Taipei
Posts: 1,449
Send a message via ICQ to TiN Send a message via Skype™ to TiN
Default

Likely OCP on your PSU is tripping. Try using other +12V rails..
Your connection kinda long, but should not be causing that issue.
__________________
xDevs.com
Reply With Quote
  #114  
Old 11-18-2014, 05:36 AM
Masterchief79 Masterchief79 is offline
Ambient
 
Join Date: Jul 2014
Posts: 15
Default

Yep changing PCI-E plugs helped. However, the card runs fine with 1,3V, but still crashes with artifacts on 1,4V, default clocks. Any ideas on that? O.o

Edit: Nevermind, now it also does that on stock voltage. Guess the card is defective after all :/

I should start zombifying working PCBs...

Last edited by Masterchief79; 11-18-2014 at 05:45 AM.
Reply With Quote
  #115  
Old 11-18-2014, 11:55 PM
TiN's Avatar
TiN TiN is offline
Volt-Nut :)
 
Join Date: Jan 2010
Location: Taipei
Posts: 1,449
Send a message via ICQ to TiN Send a message via Skype™ to TiN
Default

Probably BGA balls have bad contact on memory or GPU. It's common issue on old hot cards like 8800
__________________
xDevs.com
Reply With Quote
  #116  
Old 11-19-2014, 10:21 AM
Masterchief79 Masterchief79 is offline
Ambient
 
Join Date: Jul 2014
Posts: 15
Default

I know, I have had it on like 7 out of 10 cards(GTS, GTX and Ultras). No real idea on how to fix that though, i know you can always put them in the oven, but that's no real solution... Benching oven cards is a no-go.
Maybe with a hot air torch and then trying to apply a little pressure onto the mem chips... Or do you have a more reliable solution, except reballing
Reply With Quote
  #117  
Old 11-19-2014, 08:16 PM
TiN's Avatar
TiN TiN is offline
Volt-Nut :)
 
Join Date: Jan 2010
Location: Taipei
Posts: 1,449
Send a message via ICQ to TiN Send a message via Skype™ to TiN
Default

Get access to someone with BGA station and stencils for G80, take GPU off the board, reball all balls with stencil, clean PCB, check plating on pads, mount GPU back using ROHS temperature profile. Then your card will be fine in 90% cases. It's rarely memory, usually GPU (as it's heavy, have lot of thermal cycling (hot-cold-hot-cold) and faced down).

There is no other solution really, oven stuff will just make card work month or week, but outcome will be same.
__________________
xDevs.com
Reply With Quote
  #118  
Old 11-20-2014, 02:09 PM
Masterchief79 Masterchief79 is offline
Ambient
 
Join Date: Jul 2014
Posts: 15
Default

Yea that's true. Is it really not the memory? Yesterday I tried to apply pressure to the Ram chips after heating them up with a heat gun, so the solder would get a little soft and I could maybe repair some broken balls. I could boot the card afterwards, but still would get random driver crashes. Still better than the last state though, couldn't even boot windows anymore.

Finding someone with a BGA station would be really difficult, expensive and definetely not worth it These are old cards after all. I'm looking for a DIY-solution And I will definetely keep trying and report when I have found something that works :P
Reply With Quote
Reply

Tags
2900xt, engineering madness, overclocking solder guys, zombie


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 10:48 PM.

Contact Us - www.kingpincooling.com - Archive - Top
Powered by vBulletin® Version 3.8.3 Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.