Member Avatar for BILL S
BILL S 0 Junior Poster

I tried swapping a formerly working VA-10 Abit to an Emachine T5026 case, when that mobo proved dead except for the light near the PSU connect. Now, when the Abit was in that eMachine case, I used a Compaq S3000Z switch to test, aligning the power terminals, and the case itself. Thing I dislike is Emachine's spongy power button, this one seems cursed. After trying all that, only the board light comes on, same as the 5026 board. (Intel 775)
The VA-10 is a lower spec model AGP8 Socket A board, I swapped a former eMachine CPU into the Abit (2400+)
When I moved that CPU back to a good eMachine FIC board, it didn't power up, so far, it could be I swapped the fan plug around with the Abit heatsink in my HP/Compaq with the AM37 FIC board. IT did power up after not working initially. Needless to say, I have a few computers, it can get confusing, but the Abit is not firing up any more, and to use the T5026 case, I need IEEE1394 and USB headers for the front features, USB, Firewire and mem card reader.

I have three AMD Athlon Socket A chips, starting at 1.3Ghz, that one's still good, then the 2400+ is about 2Ghz IIRC. I think it's fried from this last incident. I believe there is an Abit VA-11 and VA-20, the 20 model has SATA added. AGP8 versus PCI Express for the simple reason, I had bought a nVidia 6200 for AGP8. Hate to hamper it in the AM37's 4x slot. The AM37 has a high rate of failure so in a way, it's surprising this one works still! Someone actually tossed the mobo and case out and I rescued it. Since it is a Compaq S3000Z, can anyone locate a front panel and side panel for this machine?? I didn't see one with it, when I rescued it.

  • 1 Contributor
  • 0 Replies
  • 133 Views

Dani AI

Generated

For the mixed-parts situation described by , the most effective next move is a strict, minimal bench test and a short diagnostic checklist that eliminates variables in a fixed order.

  • Bench test first: motherboard, CPU with heatsink/fan, one known-good memory stick, PSU, and a system speaker. Power the board outside the case on a nonconductive surface.
  • Verify the PSU rails and standby power with a multimeter or PSU tester before blaming the board. If unsure, swap in a known-good ATX PSU.
  • Clear CMOS (jumper or remove battery for a few minutes) and try again; some weird states are resolved this way.
  • Reseat the CPU and inspect for damaged pins or thermal paste bridging contacts. Confirm the CPU fan is on the CPU_FAN header — some boards will halt on fan-fail.
  • Test with one compatible memory stick in the primary slot only. Listen for POST beeps (attach a small case speaker); no beeps often points to CPU/power issues.
  • Do a visual inspection of caps and VRMs for bulging, leakage, or burn marks. Replace failed caps only if comfortable with soldering, or factor repair cost into replacement decisions.
  • If the above fails, isolate by substitution: known-good CPU in the suspect board, or the suspect CPU in a known-good board, to identify which component is bad.

For front-panel USB/FireWire/card-reader needs: many OEM cases use proprietary daughterboards. Practical fixes are a universal 3.5" front-panel bay that connects to a standard internal USB header, or a cheap PCI FireWire add-in bracket if the motherboard lacks a 1394 header. To test an unknown power switch assembly, identify the two power-on pins with a continuity meter, or momentarily short the power-on pins (carefully) to rule out a bad front switch.

Work methodically, document each swap, and avoid powering up with uncertain wiring.

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.