New Computer Build Not Working

Steenamaroo said:
I don't think I've ever used/needed a multimeter with a computer. :eek:
I do always have a bunch of spare parts, though.
Click to expand...

Yes, as I said, swapping is a good substitute for a DMM. It's just quicker and gives fairly easy to see diagnostic (at least on power supplies). Less guesswork, I guess.
 
Thinking about it, there could yet be something in the PSU idea. If the temporarily replacement only had 4 pin CPU connector, it'd be worth finding out for certain if that's adequate.
 
Steenamaroo said:
Thinking about it, there could yet be something in the PSU idea. If the temporarily replacement only had 4 pin CPU connector, it'd be worth finding out for certain if that's adequate.
Click to expand...

According to this sticky on the Toms Hardware site, you can (see item 2):

PERFORM THESE STEPS before posting about POST/boot/no video problems! - Product - Systems

They say it should be the 4 pins closest to CPU. In my case it was the 4 furthest away, but it would only fit there.
 
Reading a few things here.
Apparently the board is fussy about memory order - saw someone mention that they had to move their memory to the slot farthest from the CPU.
Since you have two sticks I guess you'd either go for the two farthest from the CPU, or the two light grey ones. I can't see their numbering.


Also, about the CPU power socket - just to be super clear, you have the main power harness where it goes, but you also have a 4 pin connector plugged into the the 8 pin cpu power socket, the only way it'll fit?
By all accounts that should be fine. The extra 4 pins are there for higher power for overclocking or whatever, apparently.
 
One other thing to check are the screws and standoffs that you used to attach the MoBo to the chassis. While fighting my no-power-up-battle with an Asus Maximus, I found information about making sure the screws and posts were not allowing a short, and also warnings about taking care not to accidentally press the back of the exposed processor socket into the chassis.

The Z97 versions has small LED indicators, on the MoBo, indicating status of the components (ie. DRAM/ PWR/ Boot_Device/ CPU/- POST status etc.). They will stay lit up until the problem is solved. [see Page 1-36 to 1-38 of the Z-97 manual] I want to think the Maximus showed Orange error, and Green functional, on these small LEDs.

You may have checked all of this already. I am just thinking out loud regarding the battles that I fought, to maybe trigger something for your troubleshooting. That was my first Asus board to work with and I found it pretty finicky. The bent pin was the result of a very low-altitude, innocent (I thought) drop of the processor, onto the pins, during installation. But once I had all the kinks worked out (Pin straightened, BIOS updates for the i7, etc.), it is an unbelievable, beast of a machine.
Dale

PS. I just read Steen's post about the RAM order... He is absolutely correct! The Maximus did not play well with the 16GB (2@8GB) Corsair Vengeance RAM that was not on the exact model on their supported hardware list, even though it had same voltages and timings, AND it had to be installed in slots that did not match the manual's reccs. (Trying to remember side-by-side/Slots 2&4/1&3, or something to that effect.) Single stick would work in one slot, but not others; two sticks side-by-side... I cannot remember exactly how I finally discovered a working config. << That was another heartache too.
 
Last edited:
DaleVO said:
One other thing to check are the screws and standoffs that you used to attach the MoBo to the chassis. While fighting my no-power-up-battle with an Asus Maximus, I found information about making sure the screws and posts were not allowing a short, and also warnings about taking care not to accidentally press the back of the exposed processor socket into the chassis.

The Z97 versions has small LED indicators, on the MoBo, indicating status of the components (ie. DRAM/ PWR/ Boot_Device/ CPU/- POST status etc.). They will stay lit up until the problem is solved. [see Page 1-36 to 1-38 of the Z-97 manual] I want to think the Maximus showed Orange error, and Green functional, on these small LEDs.

You may have checked all of this already. I am just thinking out loud regarding the battles that I fought, to maybe trigger something for your troubleshooting. That was my first Asus board to work with and I found it pretty finicky. The bent pin was the result of a very low-altitude, innocent (I thought) drop of the processor, onto the pins, during installation. But once I had all the kinks worked out (Pin straightened, BIOS updates for the i7, etc.), it is an unbelievable, beast of a machine.
Dale
Click to expand...

Aye, apparently his doesn't have the LED readout. :( Shame. Internal speaker should do the same thing though.
At this stage, I'd probably be running the thing on the bench. No case - just literally board, cpu, memory, power and essential I/O.
 
The mobo may not have a proper segmented LED numerical readout, but almost certainly the internal PC speaker will beep out a code, or there will be one or two LED bulbs somewhere on the mobo that will indicate POST issues. It's just a basic feature of motherboards, and the instruction manual will have the codes listed with explanations.

I've installed 2 Z97 motherboards in the last couple years. One has a segmented LED display, the other had something more primitive.

What exact model of motherboard are you installing? My money is on CPU thermal overload, based on your symptoms so far. If you leave it overnight, does it run for several seconds upon first boot? Then only for a second on subsequent boots?

Assuming that you're using an ASUS Z97-A, here are the LED indicators for various POST issues (page 1-36):

http://dlcdnet.asus.com/pub/ASUS/mb...WEB.pdf?_ga=1.158465914.1375006287.1463155320
 
Thanks for the ideas, guys. It helps just to talk this stuff through.

There is a dram status indicator on the board, it is supposed to light or blink when there's a ram problem. It is not lit. Accordingly to the manual two sticks of ram go into slots A2 and B2 - the lighter colored ones. That's where I have them installed. I also tried it one stick at a time. Everything gives the same result, just a quick spin of the cpu fan.

And, as mentioned I pulled the ram and put it in a Z87 board I have in another computer. Same slots. It worked fine.

I sourced the memory on the Kingston web site. You put in yiur motherboard and it tells you what memory will work. I also found a site that allows you to enter all your components and it checks for compatibility. I entered the psu, mobo, memory, cpu, and even the case, and it said there were no compatibilty issues.

If it's not the CPU, my guess is it's something with the psu or a short as Dale suggested. I wanted to swap the cpu first, because it's the only thing I haven't checked at all, and I'm nearing the 30 day return period. If that doesn't fix it, my next step is a bench build, first with the existing psu, and if that fails, with the psu from my other computer (identical make/model). If all that fails, I'm then clueless - since everything will have been exchanged and/or checked with known working components.

The computer I built for my office is nearly identical. Except with an earlier version of the mobo (z87 vs 97) and a different processor (although still an Intel core i5). It has been deadly quiet and rock solid. This was supoosed to be a slightly updated version of that. :(
 
Tadpui said:
The mobo may not have a proper segmented LED numerical readout, but almost certainly the internal PC speaker will beep out a code, or there will be one or two LED bulbs somewhere on the mobo that will indicate POST issues. It's just a basic feature of motherboards, and the instruction manual will have the codes listed with explanations.

I've installed 2 Z97 motherboards in the last couple years. One has a segmented LED display, the other had something more primitive.

What exact model of motherboard are you installing? My money is on CPU thermal overload, based on your symptoms so far. If you leave it overnight, does it run for several seconds upon first boot? Then only for a second on subsequent boots?

Assuming that you're using an ASUS Z97-A, here are the LED indicators for various POST issues (page 1-36):

http://dlcdnet.asus.com/pub/ASUS/mb...WEB.pdf?_ga=1.158465914.1375006287.1463155320
Click to expand...
It's the Z97-A, as you assumed. I was only aware of onboard LEDs for dram and standby power. The standby power was lit, the memory one was not. This is how it's supposed to be, I believe. However, I just double checked the manual and there are also LEDs for CPU, VGA, and Boot Device. My understanding is these light up if there is a problem. I've already pulled the CPU and can no longer check it, but I'm pretty sure none of these lights were on or I would have noticed them.

For info., the machine doesn't run any differently when left overnight versus subsequent starts. Just a couple of spins on the fans.

No internal speaker on the mobo, but I'm assuming the LEDs perform the same tests.
 
Steenamaroo said:
Are the CPUs compatible? Swap em out! :)



Oh... Why's that?
I mean, I was going to suggest trying a different CPU or trying your CPU in a different machine, but just assumed you wouldn't have the gear to do so.
Click to expand...
I believe the CPUs are compatible, but I'm a little averse to screwing with my rock solid build. I'm not even crazy about pulling the PSU, but I'll go that far. :)
 
Last edited:
Yeah, I can understand that. Fair enough man.

It's just so odd that it booted and worked briefly. I'm still come back to either a: Something changed that you don't know about or b: The CPU got cooked somehow.
Neither seem particularly likely but bench running it without the case would be my next step for sure.
 
Couple of things have come to mind. Some electronic devices have a thermal shutdown system that only re sets if they are taken below 0C. You could buy a can of electronic tech's freezer spray or just leave the CPU in the freezer overnight, well protected from moisture. (btw, this DOES work, sometimes, for forked USB drives!) .

Mounting screws and their bushes: Yes, possible PCB "whisker" shorting but I have known the earth pad to be insulated by flux or other residue. This did not cause a no-go but contributed to a strange earth loop hum problem.

Ah! Old brain just kicked in. 3 weeks ago I had an intermittent SATA to Molex cable that shut a PC down.

Dave.
 
ecc83 said:
Couple of things have come to mind. Some electronic devices have a thermal shutdown system that only re sets if they are taken below 0C. You could buy a can of electronic tech's freezer spray or just leave the CPU in the freezer overnight, well protected from moisture. (btw, this DOES work, sometimes, for forked USB drives!) .

Mounting screws and their bushes: Yes, possible PCB "whisker" shorting but I have known the earth pad to be insulated by flux or other residue. This did not cause a no-go but contributed to a strange earth loop hum problem.

Ah! Old brain just kicked in. 3 weeks ago I had an intermittent SATA to Molex cable that shut a PC down.

Dave.
Click to expand...

Ok, thanks. I believe I tried booting with all the expansion cards removed and all SATA devices disconnected. The only change I noticed was that in addition to the CPU fan, the case fans spun a rev or two. My guess that was from less power draw without a video card. But should eliminate any SATA cable issues.
 
Did you pull all the power/reset/led connectors like I suggested earlier?
One of those in the wrong place could totally do this.
 
Steenamaroo said:
Did you pull all the power/reset/led connectors like I suggested earlier?
One of those in the wrong place could totally do this.
Click to expand...
Yup! MoBo has a power button, makes it simple to start even with all the case wiring disconnected. I pulled everything. Usb. Audio. Along with power, reset, and led's. No love.
 
if you replaced the motherboard, are you sure you got it seated in the case correctly? Not knowing what is causing this, I'd take everything out of the case, and connect the motherboard and power supply up that way. Lay it all out on a table and connect it up outside of the case. While you have the motherboard out, check for scorched spots on bottom. If you don't have the motherboard speaker connected, it might be giving you an error code. Connect it to hear what, if there is one. Your manual should tell you what each code is.

I don't see how you could overheat a CPU if it was just on for a couple seconds. And CPUs are equipped with thermal protection so they will shut down before they burn up, unless there is some flash heat, like lightening strike.

Go back to the start and watch everything you do, to make sure you get it right. Trouble checking in this way can be maddening, because you might be missing the obvious.
 
"I don't think I've ever used/needed a multimeter with a computer.
I do always have a bunch of spare parts, though. "

As an olde valve tekkee a meter is my third hand, goto device. I put a basic (Ei-Systems) computer together for a grandson a year ago to get him W7. For reasons not moot here, he did not get it and it just stood in my hall. About 2 months ago I decided to fire it up and see if it would handle W10. The B thing was as the Dodo! My first check was to download a voltage pinout of the power plug and check voltages. All present and correct so WTF is wrong with it I do not know! Very low specc machine so not fussed.

BTW. hitting "quote" only got me half the text which is why I have copied and pasted in the relevant passage. 'King Computers!

Dave.
 
ecc83 said:
BTW. hitting "quote" only got me half the text which is why I have copied and pasted in the relevant passage. 'King Computers!

Dave.
Click to expand...

Are you using Internet Explorer? I had the same problem when using IE and complained many times behind the scenes to the admins. There is not much they can do. In the end, I reluctantly switched to Firefox and it works as it should.
 
You must log in or register to reply here.
Back
Top