Seller’s Note “The fan spins first couple second on gpu, then goes off. It boots into windows but the drivers don’t install. Shows error(Code 43). I changed thermal paste, and tried different drivers still no luck. Selling for someone for repair or parts.The card come from job lot so history unknown. Please see all the pictures before buying. It has scratches, marks, wear and tear as it is a used item.”
Summary
- Clean condition, board appears undamaged. Nice little card, would imagine it is a slightly lower power variant, single fan design and poer just from PCI-E.
- Resistances appear ok: Vcore 3.3 Ohms, Vmem 121 Ohms
- Ok, so on starting, I am noticing some behaviours:
- Fan spun for a few seconds only on first boot. GPU doesn’t seem very hot, so perhaps this is fine.
- Unusual delay to picture on boot – Very long pause I felt on power on before a picture was displayed.
- Windows does indeed load, think windows tries to load the driver and fails (momentary black screen).
- Code 43 as reported by seller.
- Think Mats is the next stop in case of a potential Vram issue.
- Running Mats:
- Was displaying text (possible error message relating to o/s), but turned to blank screen with back light after 5-10mins.
- Stayed in this state for 30 mins or more. I gave up and touched the shutdown pins, then the Mats test started, completed an passed!
- Very strange behaviour, haven’t seen that before.
Pre Checks / Multimeter Only
| Test / Checks | Result | Notes |
|---|---|---|
| 1. Visual Inspection Eye / Magnifying Glass | ||
| 2. Visual Inspection Microscope | ||
| 3. Check for shorts on 12v PCI, 12v PCI-E, 3.3v | ||
| 4a. Resistance VCore (~0.5) | 3.3Ω | |
| 4b. Resistance VMem (~60Ω) | 121Ω | |
| 4c. Resistance 3.3v Rail (~960Ω+) | ||
| 4d. Resistance 5v Rail (~5KΩ) | ||
| 4e. Resistance 1v PEX Rail (~123Ω+) | ||
| 4f. Resistance 12v BUS (14.7kΩ ) | ||
| 4g. Resistance 1.8v Rail | Ω |
Power On / Test Rig Checks
| Tests / Checks | Result | Notes |
|---|---|---|
| 1. Observations (e.g. PCB heat noticed, blank screen, power light on monitor, bios screen, o/s loaded, artefacts, ) | ||
| 2. Voltage VCore (~0.88v) | 0.79v | |
| 3. Voltage VMem (~1.4v) | 1.37v | |
| 4. Voltage PEX Rail (~1v) | – | |
| 5. Voltage 3.3 BUS | – | |
| 6. Voltage 12 Ext | N/A | |
| 6. Voltage 5v Rail | – | |
| 7. Voltage 1.8v Rail | 1.8v | – |
Update 02/01/2022 – Reflowed VRAM & Reflashed the BIOS, no change
Reflowed all 4 VRAM chips and reflashed the BIOS, but this has not changed the behaviour at all. It could be that there is still a VRAM issue, but that replacement is the only way forward or perhaps the GPU itself is the issue.
The VRAM is marked Micron 7VA47 D9TCB
This can be identified from here https://www.micron.com/support/tools-and-utilities/fbga?fbga=D9TCB#pnlFBGA, which reveals the actual part number is MT51J256M32HF-80:A
4 of these chips will likely cost around £40, which is quite an investment without clear evidence that it is a VRAM issue or a way to identify a particular chip.