Desperately need help.

[Kikojui]

New PSU has arrived.


I used thee PSU tester to see if it was working, it is.

I plugged it all in and now the PSU is ticking every second, and wont stop unless you unplug it.


I'm guessing something is causing a fail-safe to work or something? any information would be great, thank you.

[pontiacg5]

All of the lights on the PSU tester are lighting up now?

Remove the riser cables to start, especially if you are using ribbon riser cables. Really, you shouldn't have any more than the 24 pin motherboard plug and the 4 or 8 pin EPS 12V plug, the little guy by the CPU. No GPUs, no hard drives, no USB drives/headers, etc...

If that does power on, start adding things till it no longer turns on. It's possible something is hooked up wrong, and it's causing the power supply to short out and shut off to save itself.

[Kikojui]

After unplugging everything I tested the PSU again, same readings as when I first got it out the box, so seems that it's fail-safe is saving it breaking.

http://i206.photobucket.com/albums/bb191/kikojui/140317-153045_zps2d740d4d.jpg


What the Rig location looks like (just in case you imagined it might be built into a case)

http://i206.photobucket.com/albums/bb191/kikojui/140317-153229_zpsf8af5f3c.jpg

The motherboard WAS standing on this tub (As I thought the airflow would be better, I also think this somehow might have been the problem?:

http://i206.photobucket.com/albums/bb191/kikojui/140317-153156_zpsba7fae9b.jpg

I've got it laying flat now, I guess the board won't get that hot, I'm new to this so flag me on the obvious.

http://i206.photobucket.com/albums/bb191/kikojui/140317-153121_zps5f33a988.jpg

It's laying on this plastic mesh sheet ( 2 layers of it):

http://i206.photobucket.com/albums/bb191/kikojui/140317-153130_zpscd0b3a94.jpg

The wood on top was for holding the weight of the heavy cards, they reached with riser cables and the motherboard being on that tub.

http://i206.photobucket.com/albums/bb191/kikojui/140317-153121_zps5f33a988.jpg

Lastly, my components that where removed, placed on a sheet to help prevent ESD.

http://i206.photobucket.com/albums/bb191/kikojui/140317-153339_zpsed50d55a.jpg




Back to the question:

I took everything off minus the CPU and the RAM, the PSU still makes a ticking noise (constantly every 3/4 of a second) and even after turning off and unplugging, it makes that noise for a further 10-15 more seconds. NOTE* During each tick, the PSU and the CPU fan both spin momentary is if being jolted with each tick.


I'm starting to think the board is dead, I was wondering it was the way I have the rig set up (hence all the pics)... When I first plugged in the PSU today it did do the boards checks for about a second (lights all working, fans spinning etc) then it started the clicking...

I've no idea, if there is anything you want to see about the rig set-up (photo wise) I'll meet any request if you think it will help find a solution.

[pontiacg5]

Have you removed and reset the ram yet? Are you sure you've got it in the right slots, according to the mobo manual? The sequencing doesn't look right to me, but it shouldn't stop it from booting. Improperly seated ram sticks will cause a boot loop though, since ram is one of the very first things initialized by the board.

That mobo also appears to have a post status LCD at the bottom, I'd bet it's not lighting up with anything when the fan kicks over for a second?

The only other thing I can think of is that motherboard wants to be grounded for some reason, though I've never found one that doesn't work ungrounded. Does that motherboard have a on-board power switch? If not, how are you turning it on?

[Kikojui]

Have you removed and reset the ram yet? Are you sure you've got it in the right slots, according to the mobo manual? The sequencing doesn't look right to me, but it shouldn't stop it from booting. Improperly seated ram sticks will cause a boot loop though, since ram is one of the very first things initialized by the board.

That mobo also appears to have a post status LCD at the bottom, I'd bet it's not lighting up with anything when the fan kicks over for a second?

The only other thing I can think of is that motherboard wants to be grounded for some reason, though I've never found one that doesn't work ungrounded. Does that motherboard have a on-board power switch? If not, how are you turning it on?

Removed the RAM, not re-set it.

Even with just a CPU in, it still preforms this ticking over / PSU fail-safe activation.

I'm not sure if you've followed this thread, but this rig has been up and running twice. First time it ran for about a hour and then the PSU died. Sent the broken PSU back for a replacement PSU of the same model, which arrived, lasted about a hour and also died.

Now I decided to change the brand, and this time it won't even boot.  I'm thinking when the second PSU died could it have damaged the MOBO with it?

As for the RAM being in the wrong slots, you could be right, but since it's not even starting and the rig has been up and running I'm failry sure that isn't the problem.

The LCD isn't lighting up like it would when the rigs working.

I think it's just sending little electrical jots to itself and the PSU over and over.

(Annoyingly this flat has no other computer to test anything, so I just got to go with what tools are available to me and this page)

Thanks for the replies, I'm going out of my mind here!

EDIT* There is an on-board power switch

[pontiacg5]

Yeah, it will do that. You need RAM in the board for it to post. Otherwise, it will loop when the RAM check fails,  so it restarts trying the whole process again. If the RAM was not seated properly to begin with, it is possible the retention clips pushed partially seated sticks all the way out, maybe?

The power supply does not turn on and off like that, not without the motherboard requesting it. I am almost certain the power supply is fine, based on the tester you have.

You do know what I mean when I say 12V EPS plug, right? You've got that plugged in, too?

I've had power supplies spectacularly blow up in the past, but I've yet to loose a mobo that was less than a decade old. Seeing how it was just working, I think we are missing something obvious...

The onboard power switch just means there's nothing crazy going on with the front panel header and ungrounded chassis, a good thing.

[Kikojui]

Yeah, it will do that. You need RAM in the board for it to post. Otherwise, it will loop when the RAM check fails,  so it restarts trying the whole process again. If the RAM was not seated properly to begin with, it is possible the retention clips pushed partially seated sticks all the way out, maybe?

The power supply does not turn on and off like that, not without the motherboard requesting it. I am almost certain the power supply is fine, based on the tester you have.

You do know what I mean when I say 12V EPS plug, right? You've got that plugged in, too?

I've had power supplies spectacularly blow up in the past, but I've yet to loose a mobo that was less than a decade old. Seeing how it was just working, I think we are missing something obvious...

The onboard power switch just means there's nothing crazy going on with the front panel header and ungrounded chassis, a good thing.

As for the RAM, I've tested it both with and without RAM with the same outcome.

The PSU is fine, I, I'm just not sure what's causing it to tick over.

>>> I've no idea on the 12V EPS plug.  I'm going to guess and say it's the CPU plug that go's into the board near CPU, 6 pin if I remember correct?

Possibly missing something obvious, but I look at it this way:

Everything worked, then a PSU died (maybe because of something to do with the set-up but nobody can say how or why that would happen) Then another PSU arrived, tested that one and while I was out the room that one went down also.

Ordered a 3rd, better PSU and nothing works, even with the bare minimal plugged in.

What else could it be?

[pontiacg5]

12V EPS is either 4 pin, or 8 pin CPU power. Those EVGA supplies only come with 8 pin EPS plugs, IIRC. A 6 pin PCI-e plug is exactly opposite of a 8 pin EPS, so if you somehow got a VGA plug in there the PSU would likely cycle like that in a dead short.

You had it on riser cables while it was working, right? The riser cables you use do not have extra power, there's no molex plug attached to them? Besides plugging a 6/8 pin pci-e into the EPS plug, they would be the next best candidate for a electrical shorting point that could damage the mobo...



 

[Kikojui]

12V EPS is either 4 pin, or 8 pin CPU power. Those EVGA supplies only come with 8 pin EPS plugs, IIRC. A 6 pin PCI-e plug is exactly opposite of a 8 pin EPS, so if you somehow got a VGA plug in there the PSU would likely cycle like that in a dead short.

You had it on riser cables while it was working, right? The riser cables you use do not have extra power, there's no molex plug attached to them? Besides plugging a 6/8 pin pci-e into the EPS plug, they would be the next best candidate for a electrical shorting point that could damage the mobo...



 

It was the 8pin CPU power, I used VGA for VGA even for a beginner they made them bright red, with labels on them hahaha!

Riser cables worked fine for the GPU, they have no molex. The GPU have 2 VGA cables each (they use a LOT of power).

The only feasible explanation is something (either how it was built, how I plugged in the new board, OR when the second PSU died) Broke the Motherboard...

[pontiacg5]

I can't remember if it was that power supply, or only the NEX-1500, that also has 6 pin VGA plugs on the black EPS 12V cpu cable. Just checking 

I'm afraid I'm out of ideas at this point, it may be time to RMA the motherboard.

[Kikojui]

Sending off the motherboard tomorrow.

This has been a massive headache spanning over nearly a month of issues. I really hope this mining makes up for all the grief!

[pontiacg5]

The only other thing to try is clearing the CMOS, but I thought it such a long shot I didn't bother to mention it. Some boards have a hard button for it, but other motherboards have a jumper that needs to be moved. Just pop that watch looking battery out, move the jumper, and let it sit for ~30 minutes. Then put the battery in and move the jumper back. All while the motherboard is unhooked, of course.

The good news is the fans and board would likely light up if either the CPU or RAM were dead. It's still OK, hopefully, but you need the board to turn on before fixing that stuff!

Don't give up just yet 

[Kikojui]

Got the new Motherboard, realized I didn't have the smaller part of the MB plugged into the PSU (was probably what the problem was).


It's all running now except one thing... the monitor is not receiving a signal...


So what is a good way to sort that out?


Thanks in advance.


*EDIT*

I removed both GPU, and it boots fine, I've tested it with only 1 of EACH of the GPU's Both times with different riser cables to no avail...

[pontiacg5]

Ah, yeah that may have had something to do with it! You mean the "+4" piece of the long 24 pin plug?

As for the video output, have you tried all three outputs yet? Try the onboard first, then the top card, then the last one, resetting in between each test. You may need to boot with onboard video, then change BIOS settings to use pci-e graphics. I'm not too familiar with AMD, only intel...

Those riser cables you have aren't very good, in fact they are just about the worst riser cables you could have got. I've posted my reasoning behind that in the past, if you want to go look through my old posts for technical stuff. Basically, x16-x16 is not needed, and just creates more unnecessary failure points. If you haven't already, I'd try without the risers. If it works without riser cables, you might want to look into a "1x pci-e presence detect jumper." Basically, tells the card to run at 1x mode. If the riser is on the edge of compatibility, as far as signal integrity, that might just be enough to make it work.

 

[Kikojui]

Ah, yeah that may have had something to do with it! You mean the "+4" piece of the long 24 pin plug?

As for the video output, have you tried all three outputs yet? Try the onboard first, then the top card, then the last one, resetting in between each test. You may need to boot with onboard video, then change BIOS settings to use pci-e graphics. I'm not too familiar with AMD, only intel...

Those riser cables you have aren't very good, in fact they are just about the worst riser cables you could have got. I've posted my reasoning behind that in the past, if you want to go look through my old posts for technical stuff. Basically, x16-x16 is not needed, and just creates more unnecessary failure points. If you haven't already, I'd try without the risers. If it works without riser cables, you might want to look into a "1x pci-e presence detect jumper." Basically, tells the card to run at 1x mode. If the riser is on the edge of compatibility, as far as signal integrity, that might just be enough to make it work.

  

Hey there, thank you for the fast reply (doing all this as we speak).

I've tried the different ports, I will state now that I'm NOT plugging the monitor into the GPU, plugged into the on-board (which worked previously in the last 2 attempts)


I'm looking at how change BIOS to those settings, It's not as simple a process as it probably  should be.


As for the Risers, they where just so the Rig didn't overheat, as the cards are huge and would have very poor airflow, I can imagine this will get very hot once I start to mine (which is the only purpose of the rig).

I will try putting the GPU into the board direct as a test however, it will then let me know if this is the riser cables fault.



EDIT*   Nope, tried them direct into the board, same results. Going to work onto this BIOS... Weird I never needed to do this before?
All that is new is the MOBO.

[pontiacg5]

To avoid wasting time "1x jumping" I suggest running the cards direct, if they don't work right on the board a riser cable won't help!

I wouldn't try mining like that though, for obvious reasons.

The BIOS settings are different from board to board, you might google the particular part # for a guide, maybe.

[Kikojui]

^ I put an EDIT above, I tried direct, no success.


This rig is only to mine with, Without a riser of some sort it will not have the airflow to mine. So if those cables are so bad I'd love some suggestions on other risers or some magical way to plug both HD7990 giants into a mobo and them not overheat hahaha!!



EDIT* I switched the onboard to PRIMARY, and it works, I'm guessing it was because the primary was PCI-E but no monitor was plugged into the PCI-E.

Now, a question, if it's not set PCI-E to primary, will that effect the Rigs ability to mine since I'm using both PCI-E to mine with?
 

[Kikojui]

I finally got everything working, and (after a loooooooooong / stressful day) started using cgminer.

I keep (constantly) getting <Above target> message usually followed by <Worker temporarily banned>

Every time I move forward I hit another wall. any Idea what this means / is?

Getting about 600khs X 2 per card, which is a little less than it should be also (but I just want to mine a coin so all this hasn't been for nothing)