New HDD's (x3) disk write errors!! Please help I've tried MANY things. I suspect its down to consistent high bittrate usage

Question

So I have now bought 3 different hard drives. 2x 1tb and 1x 2tb. Over the course of the last 12 months or so.

The first one lasted no time at all, and I got a refund for it. The second one did last several months, I had occasional "I/O error" in Qbittorrent but I decided to ignore them since they only seemed to happen in bittorrent and whilst heavy usage was happening, the data was also fine once the downloads were completed.

Eventually that drive gave in and I lived without the extra storage for some time (i'd almost even forgotten about the previous drive failures), then I bought this brand new 2TB hard drive. As soon as it arrived I checked it with HDSentinel and it says 0hours usage, 1000+ days remaining.

I just got it, so I figure if Im gonna break it its got to be within the first week. So I have re-downloaded Steam (which I uninstalled a long time ago) and decided to install my entire library to the new drive. I also opened qBittorrent and set this to download lots of large, random torrents.

I'm getting the damned Disk Write Errors again!! The errors appear in Steam and qBittorrent. It seems to happen when the downloads speeds are very high AND I'm downloading multiple files at once (COULD THIS BE THE PROBLEM? OR IS IT OK FOR ME TO HAVE AS MANY SIMULTANEOUS DOWNLOADS AS I WANT?)

I have fibre internet and the data throughput is often at around 10MB for long periods of time (the disk usage is often 4-5 times that due to steam installing the files or unpacking for example, whilst downloading), as I say downloading multiple files (games/movies/etc/etc/etc) at the same time to different folders on the same drive. Its the secondary drive. The primary SSD has NEVER given me an error. I've tried:

3 new hard drives 3 different sata cables All sata sockets on motherboard HDSentinel reports: (It still says 100% health for this recent new drive even tho i have the errors often) ChkDsk - Says no errors found Speedfan - All temps are fine

The errors I speak of have happened whilst I am doing nothing with the computer apart from these downloads.

SORRY FOR THE LONG WINDED POST. I wanted to make sure to explain exactly what the problem is. I can't believe that it's just dodgy new hardware for 3 in a row, so there must be something in my set up, or my usage practices, that is causing these errors.

I noticed already the new drive makes the occasional clunk when the fails are occurring.

System specs:

Asus Z97-P; 32GB RAM (Crucial DDR3); i7 4790k (4.0Ghz); Asus 1060GTX 6GB; Desktop case (cheapest available); 1 case fan (cheapest); The most recent harddrive is (Seagate ST2000DM006) (as stated it happened on two Toshiba 1TBs also basically the same symptoms);

Is there a way to test it by moving data onto it quickly? I tried just copying massive folders I have on my SSD but it gave time of over 24hrs so I quit that.

Answer 1

I'm tempted to blame either:

• Your SATA controller(s) on the motherboard are toast.
• Your power supply is providing noisy / bad quality power.
• The ambient temperature in your case is too high.
• You're not installing the HDDs properly, and they're vibrating badly, which causes them to fail early.

If it were the SATA controller, then, one would think, the SSD would be having problems too (unless that's M.2 NVMe, which doesn't use SATA).

If it were the power supply, then, one would think, the CPU or GPU or RAM would've been toasted by now (except that the HDDs might be getting power from a different rail than the motherboard/GPU, and maybe that only rail is defective).

If it were the ambient temperature, you'd probably experience at least some other temperature-related negative symptom. But it could still be temperature (localized due to poor circulation).

Of course, if the HDDs are installed poorly or vibrating excessively, that's not a problem that would affect any other system components. So I'd start there.

The "clunking" you hear might be the disk attempting to park and reset the read/write heads, which is something it ordinarily only does when it powers on/off, but this could also occur if the system thinks that power cycling the HDD will help with the I/O error, or if the PSU is unreliable.

Unfortunately, this is a situation that is difficult to pin down to one specific problem, and could even be a combination of problems. I think the only way you can solve this for certain is to start replacing components, improving the temperature in your case (if it's much hotter than ambient temperature especially), and make sure nothing crazy is happening, like:

• Liquid damage to the HDD's controller board (like, leaking from a water cooling setup)
• Physical contact between the components on the controller board and the case, while the system is powered on, causing a short

---

BTW, for a bit of meta about why this is so hard for me to answer (and why, perhaps, I shouldn't have attempted to answer), see here.

Answer 2

I Originally wrote this question. It turns out it was all caused by an adapter in my PC PSU to change the old Molex into SATA power plugs. If I use the HDD in on of those slots it does these problems. If I use the built-in SATA plug it all works fine again.

My advice. DON't USE MOLEX-SATA adapters for power