Friday night head scratcher help :(

The corsair is a great unit and it is a very very odd fault.

I think the hx1000 is now obsolete so may get a hx1050 in return. If thats the case I may keep the 1050 as it runs an 87a single 12v rail with the 3.3 & 5v running a separate rail, the hx1000 runs dual 40a rails with the 3.3 & 5v sharing with the 12v.
 
Not yet the gs800 will get picked up tomorrow and it's getting wired straight in, its not even going in my case!
 
Well I give the hell up, sending the CPU back to the seller and motherboard back to gigabyte get this sorted once and for all.

New PSU same problem, fantastic :mad:
 
All thats connected is the 24 and 8 pin power connectors, lights on for a second then everything off, fantastic.

I have remembered a multi-meter today so I'm going to jump the PSU and check the readouts from the 12v, 5v & 3.3v on both PSU's.

Is there any way (with a multi-meter) to check the Motherboard?
 
Well this is REALLY screwed up now.

I just checked the voltage output from both HW1000W and GS800. HX readings first, GS readings second:

3.3v: 3.35v / 3.34v
5v: 5.08v / 5.10v (so much for that faulty 5v eh? maybe trying to trick me into buying a PSU from him!)
+12v: 12.10v / 12.11v
-12v: -11.80v / -11.86v

Now heres the strange part, the motherboard has two 8 pin EPS (Cpu power) outputs, so I thought what the hell I'll plug the second cable in. Now everything powers up APART from the motherboard, everything stays on, the PSU, the GPU is spinning up, the pump is working just nothing on the LED's from the motherboard and no display.

Any ideas? The fault is confusing me more and more by the second :(
 
Inputs sorry, HX1000W has two EPS outputs :D

Going to try using the 24pin from the GS800 and the EPS connectors from the HX1000W (as in the off there is actually something wrong with the 5v it will only have to output on 12v)
 
Last edited:
Next phase of testing I managed to get everything on but no post. Here is what I did.

HX1000W: 2 x 8 pin connected (jumped)
GS800: 24 pin connected

= Nothing, same fault as before.

HX1000W: 1 x 8 pin connected (from other rail thats 12+3.3v, other rail is 12+5v, jumped)
GS800: 24 pin & 1 x 8 pin connected

= Both PSU's stay on, even get some life from the LED readout, just no code, looks like a broken zero with no post, GPU does not spin up.
 
The code only appears when I jump the the PSU with a single 8 pin cable connected. I presume as power is then forced to the 8pin it gives the scrambled code as the motherboard its self is not actually running. When I run everything from the HX1000 (including the 2nd EPS connector) everything 'almost' comes on, the motherboard lights go completely out and the PSU keeps running, the fans keep running etc...

The code, I'll try get a pic later, I genuinely feel sick from this now! The Motherboard warranty is 100% through the retailer (OcUK thankfully), though I have messaged GB directly but I don't expect a reply for a good few days.
 
No code I'm now thinking this:

The motherboard has two 8 pin EPS connectors, If I plug one in individually from either rail or PSU its insta-fails. If I plug in two from either rail or PSU everything stays on just no life from motherboard.

GS800 - 1 x 8pin
HX1000 - 1 x 8pin
=
Power up, scrambled LED code on motherboard, GPU does not spin up. Pressing power button acts as before, GPU/MB lights flash then off again but both PSU's stay on, so the active 8pin from the jumped PSU must be forcing the activity.

HX1000 - 2 x 8pin connected
=
Power up, nothing on LED, GPU spins up fine

SO the only thing I can see I have left to test to be sure its the motherboard/CPU is to get an adaptor for the GS800 to jig up a 2nd EPS direct from that supply and see what happens?
 
Cool, I'm going to remove everything (again) and see what happens, making sure to give the socket and power connectors a good look at.
 
Right to start from scratch I have just the motherboard, CPU, PSU & 1 stick of memory (tried all 4 for ever thing I tried)

I tried as you said, with no CPU, the same happens with the exception it just loops, only have to hit the power once and that start up shut down keeps happening so its looking like the motherboard. Luckily I got the motherboard from a well known member of staff here and he is helping me figure this one out, if it requires RMA he can do it directly for me with OcUK so no major worry on that part.

Also a picture of the socket pins, I can't see anything wrong but then again I'm not used to looking for bent pins :)

Edit: Also a video of what happens (right click, save as, 19mb .mp4)

Edit: Requested a return under DSR for the CPU, more power then I need, getting a 3820 with some change, hopefully should have the motherboard sorted by that point.
 
Last edited:
Well thats everything packed up ready to be shipped out, hopefully should have the motherboard tested/replaced by the end of the week and also have a refund for the processor so I can go retail this time with a 3820.
 
Not yet, I will get that sent once I find suitable postage, though I measured al of the outputs from it and got the correct readings?
 
Nope got nothing :(

Afaik I have tried everything possible, my hunch is the 12/5v rail has damaged the motherboard. If the motherboard is damaged then I may well be out of luck with the UD7, as they don't make them anymore, was a limited edition production from what I can gather, so may be looking at a G1 Sniper or Big Bang II as a replacement.

As it stands I'm happy to pay £100 in shipping and testing fee's over 3 components to find out whats wrong.
 
Last edited:
Little update, had the motherboard tested today and its fine! Looks like the CPU may have been DOA, so I'm getting refunded on the chip and I'll be buying a retail 3820 from OcUK as soon as the refund clears.

Serves me right for buying from ebay, never again. I'd rather pay more from OcUK and know I'm getting top notch CS if I have a problem.
 
Space Monkey said:
Glad you are finally getting to the end of this ordeal :)

Are you still RMAing your PSU?
Click to expand...

I'm going to wait until I get the 3820 (hopefully be ordering over the weekend) and see if the HX1000 fires up on that, if it does I take it the guy that tested it either didn't load it correctly or was trying to blag me into buying a PSU from him. If it does fail then I still have the GS800 (which will be getting tested first) to fall back on while the RMA is sorted.

And I am ecstatic that I'm finally getting to the root of the issue.

Rusty0611 said:
Not getting the 3930K then? :D
Click to expand...

Nah, my new plan is 3820 then move over to IB-E next year, when that happens I will be going for the very very top end chip, so hopeful of something 8C16T or better.
 
Small update on things, cpu has had dsr accepted and refunded, will be ordering a 3820 over the next few days and hopefully end up with something a little more functional.

The only concern I have is the psu situation, hopefully prove either way if the hx1000 is shot.
 
I'm going to be trying it firstly with the suspected working GS800, if that boots I'll then swap PSU's over to the suspected faulty HX1000. If both are working fine I'll be putting one up for sale, buying a new PSU (Possibly OCZ 1200w, at a nice price atm).

Will be doing all building out side of the case for now, as I want to know its working before I go sorting out my waterloop :)

It is not Friday, but if this problem is still kicking me in the stones on Friday I'm going to admit my self to a mental hospital :D
 
You must log in or register to reply here.
Back
Top Bottom