buildzoid Posted April 24, 2015 Posted April 24, 2015 (edited) I'm E-powering a GTX 590 and I'm using only 1 E-power so the VCC of the cores is bridged(I have more wires on one plain that the other so I want them to share current) cutting down the resistance from VCC to GND. This a problem because when I measure GND to GND I get 0.3/0.2ohm and when I measure VCC to GND I get 0.5/0.6ohm. This means 1 of 2 things I have a small short or each of the GF 110 cores is has a resistance of about 1.1 or 1 ohms so combined I'd have 0.5ohms through them to GND. So could anyone give me how much the VCC to GND of a single GF 110 is? Because if it's 1 ohm I can stop messing around with looking for a short that doesn't exist. The multimeter I'm using just read 1.8ohms VCC to GND on an HD 5850 so I'm just going to assume that my shared voltage plane is cutting the resistance of the GTX 590 in half and I'll try run the card in 16 hours. Stop me before I kill it please... EDIT: there is no short on the card but due to some other issues that I know how to solve it won't start Edited April 25, 2015 by buildzoid Quote
der8auer Posted April 27, 2015 Posted April 27, 2015 Unless you have the same multimeter it will be impossible to compare because of the different internal resistance. I'd say the resistance is fine. Which multimeter do you use exactly? Quote
buildzoid Posted April 27, 2015 Author Posted April 27, 2015 (edited) der8auer said: Unless you have the same multimeter it will be impossible to compare because of the different internal resistance. I'd say the resistance is fine. Which multimeter do you use exactly? It's a Duwi DMT 7977. As for why it won't start. When I was soldering the wires I has some solder drip onto my PCI-e pins and pin 17 and 16 on the side with the cores are gone so I'll need to fix them. I can't fit the card into my main system so I'm waiting waiting to get my new benching rig up and running before I try run the card again. I pulled the PWM controllers off the card and so I'm not sure if there is an issue with the PWR good signal since the public data sheet for the CHL8266 is 2 pages and doesn't give any details on how the chip works. Also do you think 30 12AWG wires for +V and 30 for GND is enough or do I need to add even more? Edited April 27, 2015 by buildzoid Quote
buildzoid Posted April 27, 2015 Author Posted April 27, 2015 Strat said: 1 Epower enough for 2 Fermi GPU? I'm not using LN2 and assuming clocks of 1Ghz at 1.4V I should only need 800A(and I don't even plan to go that far). The MOSFETs on the E-power do 60A at 125C and there's 14 of them so it should be fine. Plus the OCP is 900A. Quote
TiN Posted May 26, 2015 Posted May 26, 2015 GF100/110 is about 0.400-1.000 ohm (delta measurement with zeroed lead resistance). Also for 1G 1.4V you will be around ~300-350A tops if your temps are over +50°C. Quote
buildzoid Posted May 29, 2015 Author Posted May 29, 2015 TiN said: GF100/110 is about 0.400-1.000 ohm (delta measurement with zeroed lead resistance).Also for 1G 1.4V you will be around ~300-350A tops if your temps are over +50°C. 300A per core right? because a GTX 590 is 300A to the cores at stock. Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.