Help Computer Freezing!

Toggle sidebar Toggle sidebar
K

KappaOP

Commendable
Mar 3, 2016
9
0
1,510
I have a few computers running the same motherboard: ASRock H110M-HDV. Everything is stock and nothing is overclocking. Yet these computers running the same motherboard/hardware are freezing/locking up the system daily. The error I'm receiving from event viewer is below. I'm always receiving a child device 1 or 5 error, nothing else. Computer freezes until it's manually rebooted. Thoughts?

Image of Error
System Specs
CPU Temp
 
Updated OP with link to system specs. Not sure why speccy is seeing the memory as "unknown". It's one stick of 8GB Crucial DDR4 2133Mhz.
 
The power supply is roughly the same across all of these systems having issues. About 8 computers in total. PSU used is either a 430 or 450w Corsair.
 
The Corsair CX 430 and the CX series in general (which I'm assuming is what you have) are known to fail early in life with leaking capacitors, which can unstabilize the voltages and cause freezes. How many out of the 8 computers are the freezes is this happening to. All 8? 3?
 
Yes, all 8. Which are all under a couple months old. I'm certainly not leaving that out of the realm of possibility. But I've never seen speccy pull "unknown" for ram before. Plus, within Speccy it shows the memory not operating at 2133Mhz, yet it's set for that within the bios.
 
The real frequency of DDR4-2133Mhz, at least the memory controller frequency for the CPU and the memory module, actually, is half. DDR stands for double data rate. Each clock, 2 chunks of data can be transferred between the memory and CPU (each chunk being 64 bits or 8 bytes). Original SDRAM before DDR memory only transferred a single chunk, or 64 bits. So 1000Mhz DDR RAM is = 500 Mhz SDRAM. They decided to make the naming system for DDR RAM frequency double most likely to avoid confusion with people who would think that DDR 400Mhz is worse than SD 500Mhz, when in fact the DDR 400Mhz would have more bandwidth.

The primary frequency of RAM is much lower than you'd think. For DDR4 memory, 2133Mhz, the data rate on the path between where the memory is stored (the memory array) and the I/O Bus is 2133 / 16 or 133Mhz. In DDR3, you'd divide by 8, DDR2, by 4, DDR, by 2. That's the difference in each DDR version (and a lower operating voltage). The amount of bits that can be transferred per clock between the memory array and the I/O Bus doubles each time, allowing for better bandwidth. The frequency you see on DDR mostly refers to the memory controller frequency between the CPU and I/O Bus.

Anyhoo, I'd be surprised if all 8 computers were experiencing freezing due to the PSU. Perhaps it is the electrical wiring in the facility they are all in? Do you notice a trend to when it freezes?
 
Thank you for your very detailed response! All 8 computers are at separate locations, otherwise I'd consider electrical. There is no trend on when they freeze, only that it seems inevitable within at least 48 hours that they require a reboot.
 
48 hours is a bit longer than I was initially expecting. Hmm. Okay, well, there are some things I can rule out. RAM. There's just no possible way that all 8 computers just happen to have faulty RAM, I cannot see this realistically happening. CPU, again, no possible way. Motherboard, I doubt it. Hard drive, again, no possible way. The only thing we are left with here is the power supply. This is a most incredible story you have! And just some solid proof of the CX series not being very stable.

Here's what you should do, you should buy a power supply and put it in one of the computers. See how that one holds up. If that computer does not freeze like the others, then I can assure you you'll have to replace the power supplies in the other ones. The CX series is only rated to work at very low temperatures, and it sounds like you have it running 24/7. They are notorious for failing quickly, and this is an utmost crazy event you may have here with 8 in total failing!

So yeah, replace the PSU in one of them with a better quality one. I'd be happy to recommend one within a budget of yours.
 
Thanks again for the info. What I'm wondering is that maybe I don't have faulty RAM, but incompatible ram with such a new motherboard. BIOS update may fix it however. These systems do run 24/7 however and are set to reboot once weekly. What would you suggest in regards to a better PSU for these conditions? Wouldn't need more than 500w, and would prefer modular. Price isn't a concern as long as it's the solution.
 
I've updated the OP with an image of the processor's operating temp. Although I understand that's not indicative of the temp of all hardware. Yes, video management software for business oriented surveillance is installed. Been using the same VMS for the past two years across different hardware without issues prior to this.
 
Well, if you want you can do the test. Just buy one single PSU. If freezes don't occur with that computer, it's safe to say it's probably the CX causing the problem. I don't think your PSU is faulty or has leaking capacitors, I think it's having some trouble working for that long duration. Why exactly, I cannot say for sure. Or, you can listen to Tiny Voices and see if there is a consistent software side to this. But the fact that the software worked fine for 2 years, and you got the CX 2 months ago, seems suspicious.

PCPartPicker part list / Price breakdown by merchant

Power Supply: SeaSonic S12II 520W 80+ Bronze Certified ATX Power Supply ($55.99 @ SuperBiiz)
Total: $55.99
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2016-03-03 18:51 EST-0500
 
I should have been more clear. The PSU being used in these troubled systems are within 2 months old. But the use of this brand/model has been used across 100+ systems without issues just in the past year alone. Maybe that tidbit might shed more light on this.
 


Oh, so you're saying that there aren't 8 computers in total, but there are 100+ computers in total, and out of them 8 freeze? Is that what you mean?

Are there any other systems with the 2-month old CX that don't freeze? Are some of the CXs still a year old?
 


Thanks for staying on this Turkey. To answer your questions, yes on all accounts.

 
So you're saying that only 8 out of the 100+ freeze, correct? That makes much more sense since only 8 out of the computers freeze. Since that is the case, and you run the same software on other machines that don't freeze, I do suspect the power supply will need replaced.
 
I very seriously doubt it's the psus. Have you troubleshooted the issue at all? Is the BIOS up to date? Drivers?

You can rule out hardware failure by dual booting ubuntu or some other type of linux OS and seeing if you still get a freeze up.
 
I can't see your images at work (all such sharing is blocked here).
I also doubt it is the PSU, not for all eight. Two possibilities come to mind. One is a RAM compatibility issue. Was it on the QVL for the motherboard? If your RAM is running with a 1T command rate, and those boards allow you to change it, set it to 2T and see if the lockups continue.
The other would be a Windows Update that has munged your systems, possibly in conjunction with other software or drivers you have on all of them (especially printer drivers).
Another possibility, related to the PSUs but not due to a failure, is if you're now getting some kind of power disturbance that is causing the crashes. Getting one good surge protector (an be a UPS) to put on one of them would be diagnostic. Use something better than a Wal-Mart special, perhaps even an Isobar.
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom