Greetings All,

I'm a new addition to the forum and I hope that this is not a done elsewhere topic but ...

Recently I was working on an emachine 770 computer that looked in a bad way.

The power supply had let go and was totally dead. Having swapped in a working power supply of the right type (P4 with molex connector) the case lights and the cpu fan started working but nothing else. The disk drive was moved to an external case and worked well enough attached through USB to another, working computer.

The memory strip was tested by insertion into another machine and appears to be dead.

Unplugging everything from the motherboard still gave no sign of life and the audible POST warnings have remained obstinately silent. The Caps lock light on the keyboard flashes when the machine is switched on but stays off when the caps lock button is toggled up and down. BIOS never appears.

Its all consistent so far; the cpu + motherboard is not functioning, the memory is dead and the graphics card may be suspect.

However, I'm really seeking advice and guidance on how to determine if its the cpu or the motherboard or the cpu and motherboard that has failed. The usual method is to put the cpu into another machine and test the cpu and repeat the process with the motherboard.
The method can be risky as a blown cpu can destroy a working motherboard and vice versa.

Has anyone got any better methods?

Thanks in advance

RBowen47

Dani AI

Generated

Good question, . Swapping parts can work but it’s risky and expensive. ’s suggestion to try replacement parts is valid as a last step; first follow a short, low‑risk diagnostic sequence that usually tells you which component to replace.

  1. Check the PSU rails and standby power with a multimeter or a cheap PSU tester. The purple wire (+5VSB) should show ~5V with the PSU switched on. To check the main rails safely, use a PSU tester or, if comfortable, momentarily short the green wire to any black ground to spin the PSU and verify +12V, +5V and +3.3V.
  2. Reconnect or fit a simple motherboard speaker and watch/listen for beep codes. If the board provides POST codes (LEDs or a two‑digit display) use those. Consult the board manual for code meanings.
  3. Use a POST/diagnostic card if available. It’ll show where initialization stops and is much safer than swapping a suspect CPU into another machine.
  4. Do a careful visual inspection: bulged or leaking capacitors, burn marks, damaged traces, or bent pins in the socket are strong indicators of motherboard or mechanical damage.
  5. As a last step, and only after the above, try a known‑good CPU in the suspect board or the suspect CPU in a known‑good board. Follow ESD precautions and avoid repeated hot swaps — one controlled test is enough.

Interpreting results: good PSU rails + POST codes that stop at early CPU/init stages point toward the CPU or its voltage regulation. POST codes that show chipset, memory, or I/O failures, or visible board damage, point to the motherboard. If diagnostics are inconclusive or there’s uncertainty about safe testing, a local repair shop or a used parts swap (as suggested) is the pragmatic fallback.

Recommended Answers

All 2 Replies

buy a cheap mobo/cpu and check that way.

you can buy a working version of either for around $50

OR swap out the ram for a new pair to see if it's the ram.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.