Motherboard and broadwell-e dead again

[NeoPTLD]

I have never seen a CPU fail that wasn't due to user error or operating them outside of spec. Back in the day, CPUs without IHS was common and it was common for CPUs to burn out or crack from installation error.

With modern PCs with a back plate assisted CPU mounting, a common cause of problem is crushing or shorting things out with a back plate incompatible with the motherboard or missing insulation sheet. The thick plastic sheet on back plate protects against shorting out but you can still crush stuff underneath if it sits over any surface mount component. Some shorts damage hardware, some cause malfunction or CMOS corruption requiring a jumper rest after clearing the fault.

Other thing is pegs or screws in incorrect places.

When I accidentally shorted out the back light control circuit on a laptop,It killed the back light dimming control a day after I worked on it. It cleared after I removed the short but it could have fried the motherboard depending on what stuff on the circuit was shorted out.

https://forums.anandtech.com/thread...nside-of-a-400-consumer-grade-laptop.2488539/

How did you independently verify bad CPU and bad motherboard?

 

[2blzd]

Because it's the same thing that happened last time. You can see my other thread in the motherboard section from September. I put a lot of of time and research into this the first time it happened. And it has the same exact symptoms and it died the same exact way. Red pled light in the same exact spot below the cpu socket as the last one. I got a replacement motherboard last time around, popped the cpu in and got a 00 q code which means no cpu, ie dead. So that verified whatever the culprit was, took out the motherboard and cpu with it.

New motherboard and new cpu fixed the issue. Or so I thought.

So I've ruled out everything else, what else can cause a a dead cpu and motherboard ? I'm running at stock clocks and volts, running my memory on its xmp profile which worked know first boot on all the boards. Ram has never been an issue. Brand new video card, it's worked flawlessly since I bought it.

Like I've said numerous times in this post, it worked and benchmarks fine for weeks. Then one day it dies idle afk. I'm not being defensive but I don't know what user error it could be that I'm missing.

If you can point one out please let me know so I don't replicate it for a 3rd time.

 

[2blzd]

I have rmad the psu and have a refurb/repaired one coming. (Sucks that my new system that I just paid for will be composed of refurbished parts. . Ugh. )

Intel is overnighting me a new cpu.
Will likely order a new motherboard and attempt to return the current one. I refuse to deal with asus rma and get back a subpar board.

My first test will be be the new psu in the old system to confirm its board/cpu. If that doesn't work do I attempt the new cpu in the motherboard I think is dead? Can I risk damaging new cpu?

Or do I test old cpu in new board first?

 

[BonzaiDuck]

I have rmad the psu and have a refurb/repaired one coming. (Sucks that my new system that I just paid for will be composed of refurbished parts. . Ugh. )

Intel is overnighting me a new cpu.
Will likely order a new motherboard and attempt to return the current one. I refuse to deal with asus rma and get back a subpar board.

My first test will be be the new psu in the old system to confirm its board/cpu. If that doesn't work do I attempt the new cpu in the motherboard I think is dead? Can I risk damaging new cpu?

Or do I test old cpu in new board first?

You probably already bought your PSU, and I'm comfortable to assume you know how to pick good ones.

I think I'm in a rut with an attitude explained as "don't even do the research-review-updates -- just get a Seasonic." I know that EVGA and some others are re-branded Seasonics. And there are other top-end PSU makers besides that one.

So I decided to invest in this one, instead of just getting the ~$100 "Gold" 650W unit:

http://www.tomshardware.com/reviews/seasonic-prime-titanium-650w-psu,4690.html

I think I spent $50 more. It's the last priority in the case-airflow strategy, and when I really stress the system, it doesn't even get warm. I hadn't even read that review until just now.

If I get more euphoric about this 6700K-build-in-progress, I'll want to budget a digital Vac-u-Jack, Oculus Rift and some X-rated VR program that includes the "Filthy Sanchez" coital position with Katie Morgan.

 

[Timmah!]

You probably already bought your PSU, and I'm comfortable to assume you know how to pick good ones.

I think I'm in a rut with an attitude explained as "don't even do the research-review-updates -- just get a Seasonic." I know that EVGA and some others are re-branded Seasonics. And there are other top-end PSU makers besides that one.

G2 is made by Superflower apparently, not Seasonic. Who are still highly acclaimed PSU manufacturer by themselves, mind.

 

[escrow4]

Sell the entire thing and then build a new box with parts preferably bought local and chosen by you to make sure they are brand new and not refurbs. If you don't go nuclear you'll probably still have issues.

 

[Arachnotronic]

G2 is made by Superflower apparently, not Seasonic. Who are still highly acclaimed PSU manufacturer by themselves, mind.

Superflower is good stuff.

 

[ehume]

So this hits at random times, most recently when your computer was idling. OK, you're getting a new PSU. Now get a UPS with power conditioning. Maybe you should get a UPS with true sine wave output.

 

[PliotronX]

PSU as mentioned.

Using an anti-static wrist strap and, ideally, mat? I've seen plenty of motherboards and GPUs on Ebay sitting on carpet and tables.

You may want to have your power quality checked. I know someone who kept killing Macbook power supplies. It turns out that when she plugged in a new one in her barn, which was separately wired, they didn't die. I assume the problem with their power comes from their solar panel installation.

Good points, I know its easy to get careless with esd but its a protective measure. Also, every time you see the lights dim in the house when an appliance kicks on is potentially wearing the power supply. Line interactive and sine wave UPS are pretty affordable protection these days as well.

 

[2blzd]

Sell the entire thing and then build a new box with parts preferably bought local and chosen by you to make sure they are brand new and not refurbs. If you don't go nuclear you'll probably still have issues.

I built it myself and bought and hand picked all the parts individually myself. .? going local doesn't have the parts or prices I want to pay



Anyway, is there anything else I can look at that might be causing this? I didn't overnighted the cpu cooler. . I double checked. . Everything in my case fits snug but it fits and there is clearance between everything.

Could bad cables from the psu cause it? I have some nice extension cables I bought for the 24pin motherboard connector.

I just feel like I'm a witch hunt and I'm not confident that even a new psu will not prevent this from happening a 3rd time. . It's not s fun feeling

 

[Ranulf]

Apart from what you're trying already, get a UPS battery backup. $100-200 well spent. A pure sine wave model helps, APC or Cyberpower are good brands. Ever since I had a mobo slowly go out on me after a brown out, I rarely plug in my desktop/servers straight into the wall or a power strip.

 

[ehume]

I just feel like I'm a witch hunt and I'm not confident that even a new psu will not prevent this from happening a 3rd time. . It's not s fun feeling

We agree. This is coming from outside your PSU. This is why several of us have recommended UPS systems. Like Ranulf, "I rarely plug in my desktop/servers straight into the wall or a power strip."

 

[superstition]

G2 is made by Superflower apparently, not Seasonic. Who are still highly acclaimed PSU manufacturer by themselves, mind.

Seasonic has had a ton of complaints about coil whine. Both Seasonic and Superflower released models that failed the hold-up time test at Techpowerup. Seasonic released a 1050 watt PSU that was so loud that Anandtech's reviewer said it "can be heard from rooms away".

They're considered high-quality manufacturers but they're not above making design errors. And, any product can ship with defects.

It's also not like this has to be an either-or situation. It may be that the PSU is marginal and your room has less-than-clean power as well.

 

[2blzd]

Well there is regularly 3-4 computers LANing pretty hard in the same room my computer failed in. And they are lot older and crappier PCs then mine. None of them have issues.

I can't see how only my computer is effected by this bad power and no one else's is. The machine I had plugged in for 7 months before I built this one was fine as well.

All this points to my power being fine?

 

[Madpacket]

It's your power supply. Also don't use the extension cable with the RMA'd unit. Superflower make better supplies than SeaSonic or CWT. These things happen. Take your motherboard out of the case as well and stress test it for 48 hours.

Also make sure your +12, +5, and +3.3V lines are within a 50Mv +- tolerance reading the values in your BIOS. With a Superflower Leadex design anything outside of 50Mv means you likely have a bad power supply.

 

[NeoPTLD]

Intermittent issues are difficult to detect in most fields. When you're reading other people's complaints and you haven't seen it directly, it's even harder. CPU rarely fails as I said. If the CPU has broken, it was likely killed by something external to the CPU

CPUs today operate between 0.5 to 2v. The various fairly large chips on the motherboard with coarse pins are part of the secondary power supply which is the substation that steps down 12v to whatever voltage called for by the CPU. A failure here can expose CPU to 12v and could cause the main PSU to go into lock-out. When PSU goes into lock out, it's prevented from coming back on until you power cycle by unplugging or the rear switch. Did this happen? It would be screws/pegs in wrong places and digging into traces or a glitch in design that causes the regulator to act up.

Look around the web to see if other people are having similar problems with this motherboard.

Any signs or smell of overheating around any of the power components near CPU?
Do you have a separate video card? If the main power supply is erratic enough to repeatedly kill the motherboard, it's unlikely that video card regulator or drives get out unharmed every time.

Standby UPS is pointless except for protection against power outage. So called power conditioner UPS has a transformer that can drop or boost the line voltage 5 or 10%. Since most 80Plus power supply today are designed to work between 90 to 265v, trying to maintain voltage at precisely 120v is pointless.

 

[2blzd]

So you're now telling me my hard drives and video card are dead? Uhhh. . . What?

My vga led light would light up on the motherboard if that was the case. .


Anyway, should I try the new cpu in the motherboard that I think is dead first?

 

[Burpo]

I would, and without heatsink, so I could keep my finger on it when powering it up. If it doesn't get warm, it's the board. If it does start getting warm and no red led, it may post. If it posts I'd put the HSF back on and try a boot.

 

[NeoPTLD]

https://forums.anandtech.com/threads/dead-motherboard.2485033/

I don't really know what you've got going on, but something seems amiss.
So, you powered up a board with just the 24-pin connector, and a 4-pin CPU connected from another PSU?

So you're now telling me my hard drives and video card are dead? Uhhh. . . What?

My vga led light would light up on the motherboard if that was the case. .


Anyway, should I try the new cpu in the motherboard that I think is dead first?

A blown regulator can take out the CPU. So I wouldn't. There's only so much we can diagnose over what you say.

 

[Torn Mind]

Because it's the same thing that happened last time. You can see my other thread in the motherboard section from September. I put a lot of of time and research into this the first time it happened. And it has the same exact symptoms and it died the same exact way. Red pled light in the same exact spot below the cpu socket as the last one. I got a replacement motherboard last time around, popped the cpu in and got a 00 q code which means no cpu, ie dead. So that verified whatever the culprit was, took out the motherboard and cpu with it.

New motherboard and new cpu fixed the issue. Or so I thought.

So I've ruled out everything else, what else can cause a a dead cpu and motherboard ? I'm running at stock clocks and volts, running my memory on its xmp profile which worked know first boot on all the boards. Ram has never been an issue. Brand new video card, it's worked flawlessly since I bought it.

Like I've said numerous times in this post, it worked and benchmarks fine for weeks. Then one day it dies idle afk. I'm not being defensive but I don't know what user error it could be that I'm missing.

If you can point one out please let me know so I don't replicate it for a 3rd time.

He said user error or running out of spec. Both things do not have to be true. Just one. Running things out of spec is not user error when it comes to PSUs. .

 

[2blzd]

https://forums.anandtech.com/threads/dead-motherboard.2485033/

I don't really know what you've got going on, but something seems amiss.
So, you powered up a board with just the 24-pin connector, and a 4-pin CPU connected from another PSU?


A blown regulator can take out the CPU. So I wouldn't. There's only so much we can diagnose over what you say.

That wasn't done on purpose. Just ignore that. Was breadboarding the first board that died in sept and was trying different psu's to see if the board or psu was dead. Had left the cpu 4pin plugged in from the psu that wasn't powered or plugged into the wall. It was just a coincidence that I had both plugged in at once, only one psu was actually used

 

[2blzd]

The psu fits snug in my case and the exhaust fan blows down. . My case has legs but it sits in some deep carpet. There are some screen filters between so it's not exposed.

I've never smelt anything burning and I've never had temp issues. Could this be an issue at all? The psu fan being pointing down and covered ?

All my builds have been like this and I've never had issue. Just double checking and trying to eliminate all variables that might be causing this.

So don't use extensions cables? They're nice Silverstone tek cables. What could extension cables do/cause?

Last thing I'm curious about...my main boot drive is an 840 evo. The sata power port on the drive has a slight crack on it the right angle perpendicular part..no actual pins are affected just the plastic. Drive runs fine, benchmarks fine, no errors or performance issues. Could this cause a short? Again just trying to eliminate all the possibilities.


New 6800k arrived yesterday, rma'd psu arrived today and looks brand new(yay). New motherboard and some Noctua NH1 thermal paste come tomorrow. I think thr first thing I'm going to do is take the old cpu out and try to power on. If it doesn't power the fans up or give me a q code then the board is dead for sure. I'm just swing that really for piece of mind/curiosity.

Regardless of the status of any of the old parts after my tests, I'm just going to use all the new ones because I have them and want to start with a clean slate.

 

[ehume]

The psu fits snug in my case and the exhaust fan blows down. . My case has legs but it sits in some deep carpet. There are some screen filters between so it's not exposed.
. . .

I don't know if this is relevant or not, but PSU fans blow in to cool their contents. The PSU exhaust goes out the back.

Older PSU's have fans in front. They blow in. Newer PSU's have their fans on the top or bottom, depending on how you orient your PSU. The fan blows in -- up if you have the PSU with fan side down; down if you have your PSU fanside up. You can test this; you don't have to take my word for it.

Resting a computer on a carpet is OK, but in such a case you need to have the PSU resting fanside up. Air will go into your PSU and out the back. Again, test this.

I don't know if this is relevant to your case, but the fan in a PSU should be an intake fan, not an exhaust fan. If it actually blows air out, the fan was installed incorrectly. Lord knows what else is wrong with it if that is true.

A thought: could your PSU be overheating and turning itself off? Could your PSU be overheating and killing your stuff?

 

[2blzd]

Wouldn't having psu fanside up, intake all the hot air coming from my video card (and rest of my case for that matter) since it's right above it? Is that OK?


As far as overheating, I've closely monitored Temps in everything I do multiple times a day and I never saw anything abnormal. Then again I never really monitored psu Temps. Both times my computer died it was idle with no load and I was afk. I don't see how temps can come into play when a system is idle and under no load..but then again I don't know everything obviously so who knows. .

 

[ClockHound]

So don't use extensions cables? They're nice Silverstone tek cables. What could extension cables do/cause?

Are those cables compatible with the G2? PSU cable pinout is not a standard. Everyone has their own. But, if the 24 pin was misconfigured would have killed stuff on first turn on. Usually.

Still, the first rule of troubleshooting is to eliminate variables. So...try it without the extension since more wire length = less current = more trouble. Plus the G2 have terminating caps to reduce ripple, but plugged into an extension cable means they aren't terminating caps anymore. And we haven't determined if the extension cable is flawless, have we?

Last thing I'm curious about...my main boot drive is an 840 evo. The sata power port on the drive has a slight crack on it the right angle perpendicular part..no actual pins are affected just the plastic. Drive runs fine, benchmarks fine, no errors or performance issues. Could this cause a short? Again just trying to eliminate all the possibilities.

At this point I wouldn't trust anything. Personally I wouldn't trust an 840 evo for anything other than mad science bit rot experiments.

New 6800k arrived yesterday, rma'd psu arrived today and looks brand new(yay). New motherboard and some Noctua NH1 thermal paste come tomorrow. I think thr first thing I'm going to do is take the old cpu out and try to power on. If it doesn't power the fans up or give me a q code then the board is dead for sure. I'm just swing that really for piece of mind/curiosity.

Regardless of the status of any of the old parts after my tests, I'm just going to use all the new ones because I have them and want to start with a clean slate.

Hope the clean slate really is squeaky clean. No parts recycling, right?

And on the new system, wouldn't hurt to run a monitor program with logging. Just in case the killer is a power virus installed by the Win10 data mining faeries. ;-)