HDD burned/smoked out, all other HDDs died too. Why?

Toggle sidebar Toggle sidebar
B

Boomer83

Honorable
Jun 13, 2012
41
0
10,540
Hey everyone. I would like to start by thanking everyone in this community for helping me on so many occasions with great solutions to tons of problems. Unfortunately, I couldn't quite find a similar enough issue to figure out what I should do about my problem, so I'll just create a new question.

Okay, so I'll start from the beginning. Two weeks ago, I built a new gaming rig (full system specs at bottom). I imported a few HDDs from my older build since they already had a ton of media on them and they weren't terribly old. Fast forward to last night, the computer was working fine, I was playing BF3, and all of the sudden, the computer shut off and something made a popping sound. I smelled a tiny bit of smoke, so I unplugged everything and let it sit for a few minutes. Now, it was raining out, and I live in a woody area, so I figured it may have been a surge, or a brown out (even though I have a surge protector) so I tried plugging it back in and starting it up. The PSU and CPU fans started up and some red lights flashed on the mobo, but nothing happened. So my first thought was that my PSU blew out. I guess this is where I made my mistake. I swapped my PSU with my older PSU (Thermaltake 850W) and booted up. Everything started running but after about 3 seconds, whisps of smoke started to come from underneath one of my HDDs. At the time, this particular drive was plugged into the PSU, but not the mobo because it had been giving me boot issues with remnants of an old Windows install. My oldest drive was also plugged into power and not data. The other three HDDs were plugged in to power and data, and both SSDs were as well. So I unplugged the HDD that generated the smoke and turned it back on, and it went into the BIOS perfectly fine, but it no longer detected any of the other HDDs, only the SSDs.

TL;DR: Computer turned off, something popped, smoke came from one HDD, all other HDDs also stopped working, SSDs work fine.

Fixes I've tried:

1) Since the SSDs worked perfectly fine, I tried swapping data cables, power cables and SATA ports with one of the HDDs and it didn't work.

2) I tried plugging the RE3 drive into an external dock, and it didn't seem like it even began to spin, nor was it detected.

3) I plugged the RE3 drive back into my cold computer, and I couldn't feel it spin up and it also wasn't detected.

More details:

- BIOS is set to RAID
- 5V rail is reading ~5.080
- I'm pretty sure that this is the surge protector I have: http://www.newegg.com/Product/Product.aspx?Item=N82E16812107193

Questions:

1) Why did this happen?

2) Should I be worried about plugging new drives into my PSU?

3) Is it cost effective to recover data from large drives that failed under these conditions?

Build:

Intel 3770K Ivy Bridge @ 3.5/3.9GHz
16GB Corsair Vengeance DDR3 1600
EVGA 670 GTX FTW
Corsair 1050HX PSU
2 Intel 520 Cherryville 120GB SSDs (RAID 0, brand new, still work perfectly)
2 WD Caviar Black 500GB (RAID 1, about 4 years old, didn't fully test but BIOS won't detect them)
1 WD RE3 1TB (about 2.5 years old, this was the drive I used when doing the testing)
1 WD Caviar Black 640 (about 1.5 years old, this was the drive that generated the smoke, no data cable at the time of failure)
1 Seagate 160GB (about 7 years old, untested, but I've been trying to kill this drive forever, no data cable at the time of failure)

Thanks in advance!
 
Sort by date Sort by votes
Do you have access to another computer to test the drives that didn't smoke? If you can confirm that they're all dead, and they were all plugged into nothing but the PSU, you might even be able to get Corsair to pay for them. All your other parts look very new—was the PSU new (under warranty) too?
Incidentally, that 1050W unit was a huge waste of money. Two 670s will run perfectly well on a good 650W box.
Data recovery is remarkably expensive. Take this as a lesson: when you're spending a huge amount of money on a build like this, the least you can do is back up your data!
 
Upvote 0 Downvote
1) Sounds like the PSU was wicked bad
2) YES! If multiple drives fried when plugged into this PSU, I wouldn't plug it to another drive. Unless I had old, 20 GB hard drives I didn't need and wanted to prove that the problem is the PSU.
3) Cost effective depends on the value of your data and your time. It will usually run 1.5 to 3.0 thousand dollars per drive. I did it, once. Some members here will recommend replacing the electronics board, or merely fried components on it. Some people have gotten this to work; I have not and I am prejudiced against it.

It sounds like some of your motherboard fried, too. If drives that seem to be good aren't detected, you may have lost, at a minimum, your drive controller.

Note that this is all guesses based on the description and my previous experience; there is a good chance that I have missed the target.
 
Upvote 0 Downvote
Kajabla: Yes, I have enough parts from my old PC to test it. This involves me putting my mobo on a desk, plugging in everything but a monitor, and turning it on, and I couldn't feel the drive spin. I also tested it in the external dock, and nothing worked. But later tonight I can put it into my mother's computer to see what hers sees. But I feel like I'll be beating a dead horse. And yes, the PSU is two weeks old, so it should be under some kind of warranty.

WyomingKnot: I honestly don't know what to think about the PSU. I mean, everything else runs fine. My two SSDs work, and technically, when the smoke came out is when it was plugged into my older PSU. I feel like the new one had some failsafe and turned off when it detected an HDD fault, and the old one didn't, so it just killed everything. I'm not really sure if that's how PSUs work but it seems to make sense. And no, $1000 is not worth it for me haha.
 
Upvote 0 Downvote
Is there a way to test a PSU? And I still technically don't know exactly when they failed. The new PSU popped (or it might have been from the HDD) and the old one set the HDD on fire (basically). So this is really confusing.
 
Upvote 0 Downvote

Depends. If, for example, the HDD uses the 12 volt power, and the SSD ignores it and uses the 3.3 (?), then a zap on the 12-volt line would fry the one and not the other. But this is just speculation.
 
Upvote 0 Downvote


I was thinking something along those lines, but the two SSDs were even on the same cable as one of the HDDs.

And ebbote59: Everything is SATA, and as like I said above, the SSDs were even powered by the same cable from the PSU as one of the HDDs.
 
Upvote 0 Downvote
WyomingKnott - good speculation.
3 1/2 in HDDs use both the +12 V and the = 5V, while most SSDs (and 2 1/2" laptop HDD) only have the +5V connected. (ie you can run Most SSD off of the usb port which only has the 5 V.). For SSD in a Desktop the +12 V is available at the plug, just NO connection from plug to SSD circuitry.

Best guess is that the orginal HDD +12 V went amuck, this took out the HDDs (But not the SSDs), probably shorting a component in the HDD that smoked. 2nd PSU just added injury to insult by appling +12 V to shorted component. For 3 1/2" HDD I think the +12 is for the motor, +5 is for control and data.

Possible, but less likely, a failure in the first HDD was reflected Back onto the =12 V line feeding the other HDD, thus taking them out along with the orginal PSU.

Then the question, what about the MB and other components.
Many compontens on the MB also use the +12V rail such as CPU and memory, and GPU. But most use the +12 V to supply onboard voltage regulators, ie +12V -> VRs down to around 1 Volt for CPU, Ram the +12 V -> VR to 1.5V. That may have protected these types of components. With a Multirail PSU the effect may have only been on one rail, emphasis on may.

The bottom line here is that some MB components, if not failed, may have been weakened and are now ready to fail in the near future.
 
Upvote 0 Downvote
So I guess I'll be giving Corsair a call. And does this mean almost every other component in my system could be damaged? I do have the Asus Sabertooth, (forgot to mention that in specs) which is supposed to be able to survive these kinds of things, so maybe that saved everything else.

EDIT: And this PSU only has one 12V rail @87.5A
 
Upvote 0 Downvote
Since The system works when on SSDs, that is boots to windows. I'd run prime95 and reverify that memory and CPU are fine. Then run furmark to verify that GPU is AOK.
Need to plug a "working 3 1/2 in HDD in to check out controller, But since SSD and DVD drive - OOPs you did check that the DVD drive is OK as it is also on the same +12V as the HDDs.
Passes, then No choice but to continue using and cross fingers no other MB component was damaged. If MB failure down stream, you will have to cross that bridge then.

That is the main disadvantages of the single rail +12 V PSUs. You can draw way inexcess of 20 Amps on any give +12V line. ONly limitation is exceeding the 87 Amps, or the "SHORT" burns open. With multiral PSUs you are limited to 20 Amps (generally) on a single rail.
 
Upvote 0 Downvote
Yeah, well I did get a message saying the OC on my ram (just getting it to its rated speed: 1600) was bad, but after a reboot it stuck perfectly fine. I also ran BF3 later for an hour or so and had no problems. It's no Futuremark, or Heaven, but it's certainly stressful. It did freeze at one point, and the SSD RAID reported a failure, but I think I may have loosened the SATA cable when I was messing around with the case (hopefully). And in terms of the DVD drive... well uh, I don't have one. When I bought my new mobo I figured I would just transfer the old drives over. But of course, I forgot they are IDE, and this board has no IDE. So a SATA DVD drive should actually be arriving today. But I don't think I'll be plugging it in any time soon.

And should I even risk plugging in a working HDD with my current PSU? Im sure the controller is fine, I tried my SSDs in multiple ports and they worked.
 
Upvote 0 Downvote
I think Wyoming and Retired pegged it. Your problem was a shot through your +12v rail caused by something that is hard to identify with 100% certainty. It was probably a component failure or short on the smoking HDD.

As mentioned, the +12v feeds a HDDs motor and passes through the HDD's conditioning/control circuitry. This may explain why your other HDDs won't spin up, they took the shot too. It also explains why the SSDs are fine as they don't use the +12v.

It's a very very good sign that all your ports on your SATA controller work, it sounds like the board might be fine. Your CPU and GPU also sound like they're fine.

And really, your PSU probably isn't to blame, it was probably component failure on the smoking HDD. BUT for peace of mind if I was in your shoes I would call Corsair, get a real person on the phone, explain the situation and tell them I don't trust the unit and would like a replacement. You certainly paid enough for peace of mind with your PSU.

Unfortunately, it sounds like your HDDs are toast - they won't spin on a known-working system. The irony is the data is probably fine, it's likely just the circuitry that controls the motor that is cooked. Using that assumption you may be able to find someone to attempt to repair the HDDs control boards but as mentioned the cost may be prohibitive.
 
Upvote 0 Downvote
I seen Brownouts and Surges destroy systems or cause problems.

But your older drives could of been in the process of dieing and finally just did after a slight surge. Does sound like your other devices are fine.

Best recommendation.

Test to see if your PATA port is good with and older PATA optical drive.

Drop the old drives and stop using them. They might destroy your new system. Replace them with a new SATA drive.

Most important for older electroncs and your new system.

Invest in a quality UPS for your computer. Surge protectors are useless protection and do not protect anything from brown outs and power outages.



 
Upvote 0 Downvote
Yeah I'll definitely see what Corsair recommends. Any advice on how to get elevated quickly within their support?

And in terms of the HDDs, it's super frustrating because there's most likely no platter damage, and a repair may be as simple as replacing the PCB. BUT, I know two of the drives are under warranty, and the other two may also be covered, so there's a risk reward in doing a DIY job. And data recovery is way too expensive.
 
Upvote 0 Downvote
Since you built it recently, depending on where you bought the PSU you can usually return for a replacement within a specified period of time. That would be much easier than dealing with Corsair. For instance, if it's Newegg you've got 30 days for them to receive it from the invoice date for them to send you a replacement.
 
Upvote 0 Downvote
Yeah, talking to them is worth it since you've got dead HDDs, see what they say/offer especially with regard to those.

But if you end up choosing the warranty route they might send you a refurbished unit. That's why if it comes down to a replacement, you should return it to the store for a brand new replacement.
 
Upvote 0 Downvote
Just a short note on these RE3 drives - they are supposed to be used under control from a RAID capable controller with specifics in the logic, so not every external controller and not a RAID device may recognize /"boot up"/ them.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom