TR Forums

Hi, long time lurker here. I recently build a machine loosely based on the most recent bargain box sample and unfortunately it won't post. You can see the parts below. Initially when it booted I got a continuous series of short beeps for a few seconds and then it shut off, then came on again with the same result. Now I get the on/off behavior but with no beeps. The fans just spin for a few seconds (and the optical drive light comes on if it's connected).

I've tried reseating the ATX connector and the 12V 8-pin connector, reseating the RAM, moving the RAM around to different slots, using a single stick of RAM, and disconnecting all the peripherals from the mobo and PSU. No difference in results.

I'm guessing that either the motherboard or PSU is busted, but I'm not sure how to tell which. Unfortunately I have neither another system to test (this is to replace a very old and now non-working system) nor a multimeter. Any suggestions as to how I should proceed? Thanks for reading, sorry if this is too vague.

Processor Intel i3 4150
Motherboard Gigabyte GA-H97M-D3H
Memory G.Skill Ripjaws 8GB (2x4GB) DDR3-1600
Graphics Integrated
Storage Crucial MX100
Enclosure Cooler Master N200
PSU Corsair CX430M

Did you plug in the correct 8-pin connection to the 12V on the motherboard? They also have an 8-pin connection for videocards, which can look very similar. It generally won't click in without a bit of work, but you never know. The correct 8-pin connection should be able to separate into two 4-pin connectors. Beyond that, I would reset the CMOS jumper and try to get it to beep at you again. When/if it does, write down the number of beeps, and whether they are long or short. (For instance, 4 short beeps, or 1 long 2 short)

The beep code should at least give you an idea of what's going on. If you're unable to even get beeps, replace the motherboard.

I know it's low-probability, but pull the mobo to make sure there's nothing between the mobo and the case that could cause short circuits, be it a stand-off in the wrong place or a screw that magically made it behind the mobo in the build process.

If that doesn't fix things, it's time for minimal boot (CPU/1 stick RAM) with the mobo sitting on its box, not in the case. Tedious process, but one that usually finds the culprit.

Thanks for the suggestions. I am pretty sure I used the correct 12V since the PCI-E cable on this PSU is modular and I didn't plug it in, but I will recheck. I shorted the CMOS jumper with a screwdriver for a few seconds and that did not seem to do anything.

I will try the mobo-on-box technique when I have a chance. I did take it out at one point to check and see if there was an extra standoff back there somewhere. I will say there don't seem to be any standoff mounts on the far right in this case which makes me nervous pushing the ATX connector into the board.

Another thing that I noticed was that the Mobo heat sinks were kind of a tiny bit loose, but it looks like they are held on with some kind of spring so maybe that's normal?

Assuming it doesn't work on the box, I guess I just return the mobo and buy a new one? I'm not sure if I should just return the PSU since it seems like 50/50 which is the problem.

Yeah, definitely check what The Egg said about the CPU power connector. Getting the PCIe and CPU power connectors mixed up isn't that hard to do if you are working inside a cramped case, and will result in grounding out the +12V rail. Depending on whether your PSU's short-circuit protection kicks in or not this can be "mostly harmless"... or it can cause wiring and/or connectors to melt. You didn't smell anything burning, did you?

Beyond that, you'll want to follow Ned's advice to isolate the culprit. Having spares of the various components to swap in would be helpful if the minimal boot test fails; if you don't have anything on-hand, maybe you can borrow from a friend or from work?

Oops, posted just as you were.

selfnoise wrote:

Thanks for the suggestions. I am pretty sure I used the correct 12V since the PCI-E cable on this PSU is modular and I didn't plug it in, but I will recheck. I shorted the CMOS jumper with a screwdriver for a few seconds and that did not seem to do anything.

OK, that's probably not the problem then. But double-check that you're connected to the correct modular jack on the PSU.

selfnoise wrote:

I will try the mobo-on-box technique when I have a chance. I did take it out at one point to check and see if there was an extra standoff back there somewhere. I will say there don't seem to be any standoff mounts on the far right in this case which makes me nervous pushing the ATX connector into the board.

I hate it when that happens. With the widespread use of ROHS (lead free) solder, the solder connections on the board are more brittle too. Increases the odds of cracking something when the board flexes. My advice in situations like that is to slip something (e.g. pad of Post-It notes) under the mobo to support it while pushing the connector on.

selfnoise wrote:

Another thing that I noticed was that the Mobo heat sinks were kind of a tiny bit loose, but it looks like they are held on with some kind of spring so maybe that's normal?

Yes, that is normal.

selfnoise wrote:

Assuming it doesn't work on the box, I guess I just return the mobo and buy a new one? I'm not sure if I should just return the PSU since it seems like 50/50 which is the problem.

That's where having (or borrowing) a spare PSU comes in.

just brew it! wrote:

selfnoise wrote:

I will try the mobo-on-box technique when I have a chance. I did take it out at one point to check and see if there was an extra standoff back there somewhere. I will say there don't seem to be any standoff mounts on the far right in this case which makes me nervous pushing the ATX connector into the board.

I hate it when that happens. With the widespread use of ROHS (lead free) solder, the solder connections on the board are more brittle too. Increases the odds of cracking something when the board flexes. My advice in situations like that is to slip something (e.g. pad of Post-It notes) under the mobo to support it while pushing the connector on.

Modular PSU to the rescue here, as you can attach the mobo side of things before you put it in the case and then dress it back to the PSU.

Thanks for all the suggestions. Here's my update:

I removed the MB and breadboxed it with the PSU. Same result. Removed the RAM to see if I would at least get some angry beeps. Same result.

I removed the CPU and I think there may be a bent pin on the socket... hard to tell since they are so tiny. So I think I will just return the motherboard and get a new one. I'm guessing it's either that or the power supply so I'll start with the MB.

I'm paranoid now, though, since I followed the instructions on seating the CPU exactly so I'm not sure how it happened. It's possible that it was bent out of the box and I missed it? Other than that... I did feel some resistance putting the lever back down but from videos (this sites, among others) that seems to be normal. The dimples on the CPU side were lined up with the socked and I jiggled it slightly to make sure it was set before putting the lever down.

Just dropping by to update: I bought a new motherboard and that did the trick. Went for MSI H97m-G43 instead of Gigabyte mainly out of superstition. Posting this from the new build... so far so good. Thanks again for your help.

I have almost same system, and same problem. I got 1 long and two short beeps. According to the Beep code it said faulty Video card. What I did I returned the MB and get another newer model GB which gave me same results. I did all the above troubleshooting processes which did not seem to work. Is it MB problem? no sure, so what I'm going to do is to change the MB with AsRock also for my interest I bought a Video card to check the MB before I returned it. Any suggestions. Thanks