New Build Prob

Brightonclock · 28 May 2012 at 09:53

Hi all,

Completed my new build friday night but have had an issue, hopefully someone can help as am running out of ideas.

What happens is, sometimes, on pressing power on, the system does not seem to want to boot at all. Other times, the PWR LED flicks on for just a second, and I see all the fans start to spin up, then nothing. Yet at other times, a single, press, the whole thing jumps to life, with no problems. Out of 4 shut downs yesterday, 2 booted up fine twice, 2, did not.

Once it has booted, it seems to be 100% stable, left on for hours, played some diablo 3, cod etc on it, no problems. Temps are good. No BSODS/Crashes/Reboots or anything. When I restart, its always fine, straight back in. But when I shut down, its a gamble as to whether it will 'fire up' again seemingly. It always does, eventually. Sometimes after 3 presses, sometimes I have to leave it 10 mins. Whether not I disconnect the plug from the PSU first or not seems to have no bearing on whether it will start first time again. I have tried removing the CMOS battery also after the bios flash, same problem. It's almost as if it WANTS to start at times, or is trying to, but something is stopping it/cutting it out. But then also, on some presses, its like its stone cold dead.

I have tried running it off the integrated cpu graphics, with the graphics card out, same issue, so I don't think thats causing it.

I have rebuilt it about 4 times now looking for any obvious problems, I can't see any. All screws/connections look OK. The PSU is new and if anything overspecced for the machine. The auxiliary pwr connector is a 4+4 into the mobo's 8-pin socket, I take it that is OK?

Could the RAM be doing this? Testing each stick is a pain so I have not done that yet, Ive tried basically everything else I can think of. as the HSF has to come off to get at them. It is low voltage, would this be an issue? Once it has booted, its totally stable, so I don't know.

I'm almost out of ideas myself, can anyone offer any advice at this point? (my grasp of electronics is limited)

CPU: 3570K
Mobo: ASUS Maximus V Gene Z77 updated bios to 0813
RAM: 16GB (4x) Samsung Green 30nm 1600Mhz low profile
HSF: Noctua NH-D14
Case: Silverstone Temjin TJ08-E
PSU: 750W Silverstone Strider 80 plus silver
GFX: Sapphire 6870
All bios options are factory, no oc etc yet.

Thanks,

Dan

Kitarist · 28 May 2012 at 10:01

Maybe the PSU is faulty. Do you have any other PSU that you could test ?

olivier renault · 28 May 2012 at 11:22

You checked your front panel connectors and buttons?

Brightonclock · 28 May 2012 at 11:28

I might be able to wire up the old 500W I have from my old comp to it, should have enough cables/adaptors to try that I guess it is worth considering it at this point. I'd have thought if there was a prob with it though I'd get crashes/instability once its running?

The connectors to the board look are OK from the front panel yeah. I have not examined the actual wiring from the button itself, suppose it is worth a look when I get in !

Idleman · 28 May 2012 at 11:36

What voltage is your ram running at?

olivier renault · 28 May 2012 at 11:38

Brightonclock said:
I might be able to wire up the old 500W I have from my old comp to it, should have enough cables/adaptors to try that I guess it is worth considering it at this point. I'd have thought if there was a prob with it though I'd get crashes/instability once its running?

The connectors to the board look are OK from the front panel yeah. I have not examined the actual wiring from the button itself, suppose it is worth a look when I get in !

If you have a dodgy button, you can try the screwdriver trick, basically quick-short the power button wires to similate a button press.

Brightonclock · 28 May 2012 at 11:51

http://www.overclockers.co.uk/showproduct.php?prodid=MY-017-SA

Thats the mem, I'm not sure what voltage it has defaulted to offhand (at work at the mo so cant check. From what I recall it was quite low. Perhaps I should set it to 1.5V or something, maybe its undervolting and causing a prob? TBH I probably should have checked the Mobo QVL list for RAM before I got any, as it seems to be a newish one out.

Can you explain quick-shorting? Sorry for my tardedness.

Idleman · 28 May 2012 at 12:03

Try using the mobo power-on switch and see if the problem persists. If it is ok, then you know it is the case button. If not the problem lies eslewhere. Psu or ram.

olivier renault · 28 May 2012 at 12:04

1600MHz (7-8-8-24 1N) @ 1.40v

I'd try that. Maybe 8-8-8-24 2N is you have stability problems.

Can you explain quick-shorting? Sorry for my tardedness.

It's a way to simulate a power button press by touching two pins with a screwdriver on the front panel header.

In fact, don't bother, As Idleman says, your motherboard has a start button. So you can try playing with that and see if you can repro the bug. If you can, it's not the case.

Also check 'TDC Current Limit', or something similar, but you haven't overclocked and it's stock.

Brightonclock · 28 May 2012 at 12:44

Sorry I forgot to mention that yeah I have tried it with the Mobo's start button, still the same, so I guess we can rule out the case switch as you say.

So we are down to the PSU being, essentially, naffed in some way. Or a RAM problem. Either, at least one stick of the four, is knackered, or the voltage/timings could be amiss. OR the board just does not like it.

I don't think it is using timing's that low offhand, even though it supposedly can. I will check this when I get in before I do anything.

olivier renault · 28 May 2012 at 13:17

Sensible is 8-8-8-24 2T 1600MHz, 1.5V, but I think you can run at 1.4V on those quite comfortably. Would definitely look at the RAM before doing anything drastic.

Brightonclock · 28 May 2012 at 13:59

I'll try setting that when I get in and give it a try. It's a frustrating prob, works so well apart from this!

Brightonclock · 28 May 2012 at 19:45

OK, so I got home, hit the power button 1st time, all good.

I check my RAM - 11-11-11-28-2T default, 1.35V

I set it to 1.5V, 8-8-8-24-2T, restarted, booted fine into windows.

Shut down. Power pressed. nothing

(( Seems like RAM voltage may not be the problem. Several presses later, it attempted to spin up the fans about 3 times, but zip. Unplugged, 5 mins, replugged, nothing initially. Few presses later, all good again. Aaa!

Is the next step to start taking out the RAM? Is there nothing I can do to test it in-situ reliably, the HSF has to come off with each switch and to get at the RAM which will make for a time consuming test

Anything else I am missing before I try another PSU/removing memory?

olivier renault · 28 May 2012 at 21:39

My guess would be on the PSU. I'm nowhere near Brighton unfortunately, otherwise I would have lent you a HX620 to try.

Unless you're around Birmingham, and your alias is some cunning ruse...

BTW cold boot issues can be motherboards.

finding faults can be a pain. Last thing to try, do you have anything in the bios sounding like 'HDD Delay'? Something that gives the system a few seconds to warm up.

olivier renault · 28 May 2012 at 21:50

http://www.eteknix.com/reviews/memory/samsung-green-ddr3-1600mhz-8gb-30nm-memory-kit-review/

So it works with that board. Try two at a time I suppose.

I think the original timings are the recommended values, so I'd stick with that.

Brightonclock · 29 May 2012 at 09:17

Yeah I am in Brighton, but thanks none the least!

Cool, well that article is a good start at least, it SHOULD work...

I'm going to try two things tonight.

I'm going to take everything out the case, and put it on some card or something, then try startups/shutdowns, this will ensure no chance of a short right?

I take it shorts are caused when metal contacts the mobo? If I can do like 6-8 startups/shutdowns in a row with no errors, I'll consider it sorted. If I still get the prob with it out the case, I do have an old Antec 500W PSU from my old comp. It was in use up to the other week, and has always been solid in my old amd. I think, it has all the connections on it I'll need, hopefully if not the load of new cables I got will have any adaptors needed, so I will try it on this and hope for the best.

I think after that, the only thing left to do would be a process of elimination with the RAM sticks, to hopefully find the prob

After that, I might just bag it all up and chuck it in a skip !

olivier renault · 29 May 2012 at 11:07

Brightonclock said:
Cool, well that article is a good start at least, it SHOULD work...

I'm going to try two things tonight.

I'm going to take everything out the case, and put it on some card or something, then try startups/shutdowns, this will ensure no chance of a short right?

put it on the motherboard cardboard box, but NOT on the electrostatic bag.

I take it shorts are caused when metal contacts the mobo? If I can do like 6-8 startups/shutdowns in a row with no errors, I'll consider it sorted.

yeah that can happen. Usually through stand offs that touch a part of the underside of the motherboard.

Brightonclock · 25 Jun 2012 at 13:50

Bizarrely, this sort of resolved itself. I'm not 100% sure on how, but I think it may have been caused by SATA cabling on metal. When installing my new ssd the other day, I did read that shorts can happen where SATA cabling touches metal. I have repositioned the cables, and it seems all good for now. I would have thought they were shielded against that sort of thing though, anyway, I am happy it did not end up in the skip after all!

Many thanks for your help all