MSI GTX 1660 Super - all rails shorted?

[Zakyn]

I received a broken MSI GTX 1660S, not detected in the PC. There is no short circuit on 12v and 3.3v. 5V is present. So I moved to 1.8V, and it's missing, so no other voltages will be there. Against ground I am measuring 0.0ohms on 1.8, PEX, VMEM and VCORE. Is it dead core, or Turing GPUs has that low resistances? And its possible that it burned in a way that all rails are shorted? As sanity check I removed GS9216 buck converter and replaced it with new one but still no 1.8V. Can anyone please confirm that I can put this GPU away?

[McWhitePink]

are you using ohm mode? if you use ohm mode and measureing 0ohm, then it probably the core is dead. or maybe there are something under the core, like liquid metal.

[Zakyn]

Yes I am measuring ohms and it is 0.0 ohms. Core is clean, so its gone, thanks for help.

[McWhitePink]

Yes I am measuring ohms and it is 0.0 ohms. Core is clean, so its gone, thanks for help.

Yeah, likely to be dead, at least VMEM should be measuring like 30ohm.

[McWhitePink]

But it's really strange anything can burn all the rails at once. They don't even tied to same 12V input.

[ductranvinh]

The IC to supply 1.8v power (GS9216-GS9219), the open control pin (pin 2) must have a voltage of 3.3v. But U509 pin 4 will supply signal voltage to pin 2 of the GS9216. U509 to open, you must have pin 1 voltage which is the 5v power signal line to work well. Normally, if the card has been operating for a long time, the pin signal line will Number 2 of u509 is lost because Q519 (double mosfet 2N7002DW) is shorted or damaged, causing the voltage to go straight to ground. Q519 can be removed, if you want you can replace it.
I usually don't care about internal resistance because in many cases there is an internal short circuit where the 1.8v line goes almost to zero but that doesn't mean the gpu is dead. You can remove the 1.8v current management ic and apply 1.8 -1.9v voltage directly to the 1.8v line coil. If the current is higher than 1.5A, the gpu is definitely dead.

MS-V375 r2.2 boardview - https://www.badcaps.net/forum/troubl...r2-2-boardview

[ktmmotocross]

remove 1.8v coil and inject. u will see what is burning

[Zakyn]

remove 1.8v coil and inject. u will see what is burning

Coil removed injecting 1.8V, power supply max current 4 A, nothing is geting hot so i guess the core is able to to dissipate 7W of heat?

[ktmmotocross]

Coil removed injecting 1.8V, power supply max current 4 A, nothing is geting hot so i guess the core is able to to dissipate 7W of heat?

4A is low. u need at least 10A which can be see with thermal cam, but that gpu is toasted

[Traptor]

Coil removed injecting 1.8V, power supply max current 4 A, nothing is geting hot so i guess the core is able to to dissipate 7W of heat?

Try removing the bios chip. For me, the bios chip caused the short circuit on line 1.8

[ktmmotocross]

Try removing the bios chip. For me, the bios chip caused the short circuit on line 1.8

it will be burning hot with that 7W heat