New build. Used (working) parts. When connecting PSU, the CPU and system fans spin and stop continuously. The green power led stays lit. There is a red led for AMT that is on and off as power seems to cycle. No post or any beeps.
I've tried removing the battery to reset the cmos. Same thing. Tried a second PSU. Tried with and without RAM. Etc. Even tried connecting a network cable to LAN to see if I could get this thing to boot from LAN or reset state. I get a nic led to light but nothing.
When I press the power switch for 10-15 seconds the "cycle" stops and nothing happens excepts green light. A button press just restarts fans and stop again. I'm thinking the board is bad or the bios is stuck in some sort of state.
CPU = Intel Core 2 Duo E6420 @ 2.13GHz
RAM = G.SKILL F2-8000CL5D-4GBPQ PC2-8000 4GB 2X2GB DDR2-1000 CL5-5-5-15
I've been struggling to get this board (bought on ebay) up and running. I've build systems for years but this thing is a challenge (for me anyways).
First thing I would try is to install the latest BIOS available for this board via the recovery method. See the documentation for your specific board for information on this method.
Thx for taking the time to help. I downloaded the latest bios .BIO file and copied to an empty usb flash drive. But the board wont even "read" it. The second I apply power it just starts the fans and then stops, up and down, again and again. No post. I see the usb stick lighting up when the fans spin but then it stops.
The document is not very clear on how one would clear the cmos. I think it is just a matter of removing power and the battery for at least 5 minutes. But do i need to use the jumper? If so on which pins? Also, does the sequence in which we plug/unplug the ATX power connectors from the PSU matter? How about the cpu fan? Do I need to plug and unplug that as well? What about the CPU? Would it help to remove and reseat? I did this but can try again.
Or I could just set fire to the board and dance in circles while jumping and see what that does...
Here's some answers for your questions:
- You should do two things, (1) disconnect the PSU from the wall (the motherboard is getting standby power otherwise) and (2) remove the battery for 30 minutes.
- It doesn't really make that much of a difference, but I remove the ATX connector before the ATX12V (4- or 8-pin) connector. As I mentioned above, if the PSU is connected to the wall, it is providing standby power to the motherboard (it is this power that lights the green LED). If the system is connected to the wall but sitting in a "powered-off" (S4 or S5) state, no power is being delivered via any of the standard 12V, 5V or 3.3V rails. This means that there is no power being provided on the ATX12V connector at all; it is only the standby power rail on the ATX connector that is receiving power. That's why I remove it first.
- Presuming that your system is powered off, you can remove the processor fan cable from the motherboard connector at any time. While you should never power up the system without the processor heatsink attached, you will be OK for a while if you do so with the heatsink attached but the fan not plugged in. The heatsink provides a mass that will extract some processor heat and ambient airflow will allow some of this heat to be dissipated. Eventually, however, the processor will generate enough heat that it cannot be dissipated by the heatsink and silicon temperatures will increase. In this case, the processor has the capability to throttle its own performance so that it will take much longer to reach a critical level. If the processor does reach a critical level, it has a thermal-trip circuit that will protect it by shutting down the processor (almost like a fuse, but one that resets when the processor eventually cools off).
- When I remove the heatsink from my processor, I only unplug the fan if I need to move the heatsink well out of the way (like when I am removing the motherboard from the chassis).
- In your case, I don't think re-seating the processor will help.
OK, I don't want to scare you but, if clearing CMOS and doing a recovery of the BIOS doesn't work, you may have a permanent problem. One of the possible causes for the symptoms you are seeing is a corruption of the Management Engine's (ME) portion of the flash. The Management Engine is one of a number of microprocessors in the chipset that is responsible for a whole bunch of capabilities including power-sequencing. [Aside: In your (4-Series) system's case, the ME is also responsible for system thermal management and is making all decisions with regards to fan speed.] Anyway, if the ME's area of the flash is corrupted, the system is never going to run to the point where a flash recovery operation can take place. From your standpoint, this is as bad as a hardware failure; only an Intel technician can reload the flash from scratch. It may be that this is why the motherboard was for sale on E-Bay
Let's hope that this isn't your problem...
P.S. Regarding your "press the power switch for 10-15 seconds" comment, if you press and hold the power button for more than 4 seconds, a circuit in the chipset will hard power-off the system. This is not programmable or maskable; it will always occur. Continuing to hold the button beyond this point accomplishes nothing; releasing the button has to occur before the circuit resets and another power-on can be attempted...
The saga continues. After maybe 50 tries I was able to start the board and update the bios to the latest version. I have no clue what I did different if anything but it booted.
I went in the menus, updated the date/time, etc. Shut down, then finalized the board in its case. Now it is not starting again.
When starting up the system the fans spin once and then stop. The power led is solid green and now the difference is the AMT led is solid red.
I also validated the voltages from both PSU i tested with using a voltmeter and it is all good. Question: is there any difference if the testing I do takes place outside a case? I do my testing on a table making sure there is no short. I mean, could the board not start because some contact is not made with a case or something?
No, there is no contact with the case required. You should always do tests like this outside of the chassis. This avoids issues like shorts and ground loops caused by the chassis or any extra things built into it. I would remove the board from the chassis and try it again on the bench.
The seller finally agreed that I had a non-working board and refunded me. I will not continue working on this board. That you very much for all your help Scott. Appreciated.
I saw there are diagnostics or debug cards we can plug on the board and see what is wrong. Maybe that could help me out? I ordered one anyways just for fun and "learning".
The diagnostic boards are of limited value. They only provide two capabilities, display of BIOS POST codes and display of power rails. If your board already has a BIOS POST code display, the diagnostic board does little for you. You must have a PCI (not PCIe) slot to insert the diagnostic board into, so your choices are typically limited to older boards. Most modern boards either have only PCIe connector(s) or they have utilize a bridge IC to add the PCI connector and this makes it impossible to use the POST code display capability of the diagnostic card (because board cannot access LPC bus).