Weird HDD Stats on Brand New Mini i7 2.3

pongo99 said:

Hi,

I bought a new Mini earlier in the week (i7 2.3 version, with stock 1TB disk and 4GB memory), found it was making weird noises from the hard disk so took it back today and got a replacement only to discover this one has the same issues.

But that's another story. The issue here is that what i found when checking the SMART status of the drive has taken me aback somewhat. The stats on the one that i returned showed that the HDD had been on for about 23 hours and had a Load Cycle Count around 640. The replacement i got says the drive had been on for a total of 46 hours and had been powered on 453 times, with a Load Cycle Count of 1340 something. I didn't note the Power Cycle count of the first unit unfortunately.

These stats seem awfully odd for an ostensibly brand new machine. 453 Power Cycles is incredibly high, and 46 hours seems like a lot of use fresh out of the box. That's something like 9 start-ups per hour.


Does anyone have an idea what this might indicate?

Should i take it back again?




Thanks.

Click to expand...

pongo99 said:

You might think it's pedantic of me to want to return it for such a minor issue, but i'm spending money to buy a product that i expect to last upwards of three years, and with such an awkward start to its life cycle i can't help but feel uncertain for the future and especially the reliability of such an item.

It's just baffling to me how a hard disk can be spun up 453 times before ever reaching the end consumer. With such doubts i believe it's worth taking a chance on another unit just to assuage my anxiety.

The cycle count goes up by one every time the computer is turned off and on again or wakes out of sleep mode. How can it be possible for this to have happened more than four hundred times in 46 hours of its lifespan? And that's even before reaching a retail point. Something just seems fishy here...

Click to expand...

Drives exit the assembly room in a small crate and pass along a conveyor belt. From here, drives can get routed to any of several areas and machines for long-term testing. One such test machine is called Excalibur. In the picture on the right, you can see an operator feeding a drive onto Excalibur’s conveyor belt. That object behind the glass just above head level is Excalibur’s robotic arm. The robot grabs the drive and whisks it back down the aisle you see behind it. The drive gets inserted into a bay, whereupon its media is filled with a servo pattern to fill the tracks with data. This will confirm that all of the media is able to store information.

Excalibur can hold about 5,000 drives. We saw other smaller test units, but WD needs Excalibur to handle the location’s volume. At any given time, there are usually several experimental models in Excalibur’s test ranks, both 2.5- and 3.5-inch. The machine cranks through several tens of thousands of drives annually.

While we’re seeing WD’s test procedures on pilot line drives here, know that the exact same tests are applied to the production lines in Asia. Even though WD manufactures over half a million drives per day (the company reported almost 50 million produced in Q4 of 2009), each one goes through an Excalibur system and testing process just like these relative few.

Click to expand...

DitteVilladsen said:

Hard disks are subjected to "long-term testing" in the factory. Slide 40 of this tour of Western Digital shows one of the testing machines:

http://www.tomshardware.com/picturestory/525-western-digital-tour.html



The idea that you happened to get two instances of a brand-new model of computer, whose assembly takes place in China and production of which only began a couple of months ago, that had somehow been subjected to hard disk replacements with used hard disks that were previously owned by two separate owners who had bizarre habits of turning their computers on and off ten times a day is ridiculous.

The disks just underwent normal factory QA. Your buyer's remorse is approaching seriously neurotic levels, I think.

Click to expand...

The downside of parking the heads is that HDs are usually not designed to do this every few seconds. Typical limits range from 300,000 to 600,000 (link) load cycle counts.

Click to expand...

Now, after four months since discovering the problem went by, my load cycle count only increased by about 500 (which is about the number of times i sent the mac to standby). Isn’t that a figure compared to the 36500 within the first month?

Click to expand...