Hard Drive Torture

Yep, another weekend and another project... In this case I had just wiped 15 hard drives so that they could be transferred to surplus. Being the curious sort, I wondered if there was a quicker way to do this than running format over and over again. The quickest way would be with a magnet. This begs then begs the question: "Will it work?". After considering this for a good day or so I figured I might as well test some drives "scientifically" to see what other trauma a HDD could survive. To find the answer I went out and bought four cheap old 500MB hard drives. As usual, Computer Success came through again with my used parts needs.






I decided it might be a good idea to run a surface scan on all of the drives, considering their age. Three of them tested correctly, but the fourth (one of two Maxtors I bought) had four bad blocks. This did not concern be since nothing I was going to do would only harm one or two blocks and would easily stand out. For the actual tests I decided to use Norton Utilities 2001 in DOS mode running continuous scans. I also had to create a new IDE ribbon cable for these tests. Normal IDE ribbon cables are only 18" long and this could have made it difficult for me to traumatize the things. So I bought some cable and made a 2.8' cable. Along with a power extension cable, I could now freely move the drives about while they were running.

The first test I ran was the magnet test. For this I used a set of magnets from car stereo speakers. They are stacked on top of each other in a cylinder about 1.75" tall by 3.25" wide. They are fairly powerful, but nowhere near what a rare earth magnet is. The first test inolved merely passing the magnet over the HDD case a few times. After a surface scan found no damage, physical or otherwise I removed the cover and passed the magnet right over the platters. Again there was no damage, even to the stored files. I found this fairly amazing, since the magnetic media is extremely thin and I thought it would be very susceptible to outside interference. Since this wasn't the case I decided to up the ante with a stack of magnets I had previously removed from other HDDs. These are far, far more powerful. I think they are probably of the rare earth variety, but do not know for certain. Anyway, I ran these over the platters without the benefit of a cover. They were powerful enough that I nearly hit the platters a few times but narrowly avoided contact. This time there was evidence of extreme damage. By that I mean even the BIOS could not detect the drive. I figured this meant something somewhere was definitely damaged... Since there was no way to scan the drive, I just gave up here.




The next test involved a pen. Since I have never seen physical HDD damage, I decided to make some so I could see the pattern show up during the scan. This was a very simple test. I just turned it on, started a looping surface scan and pulled the lid off the drive. I then touched the surface of the platter and held the pen there. Repeat. After a while there was a faint, almost white ring on the platter, but no sign of ink or scratches. Even after I pushed fairly hard there was no visible scratches. Of course, that doesn't matter since Norton Utilities devided there were quite a few problems with that white line. In fact, it found bad blocks all over the place. Now, as you may know, bad blocks slow a scan down. Way down. In light of this I stopped Norton after about half the drive was scanned. just for kicks I shorted out a few of the contacts on the controller card and discovered I could restart the computer that way. Good to know, huh? I also discovered that I could push the armature around and it would not cause any damage and would seemingly easily find its way back to where it was supposed to be with only a small pause in the scanning.


The third test involved fire. Since fire looks cool and some things burn, this seemed like a good idea. What I did was this. I pulled the cover off, then tried just holding that nice Bic flame between the platters. The flame blew out. I tried again and again the flame sputtered and died. The spinning of the mirror-like platters was creating a decent air current. I decided to try a different approach and held the lighter over the top platter. This time it stayed lit and I held it there for about ten seconds. I waited for the scanner to complete one cycle and saw no damage whatsoever. Again I had thought that the thin layers on the surface of the plattermight be suceptible to the heat. It wasn't. At this point I was upset that there fire has had no effect. So, my hand started wandering and lo and behold, the fire came to rest on the Read/Write heads. There was a little pop and the armature parked. Another one bites the dust. When I was adjusting the contrast of the picture to the right I discovered what looks like a plume of hot gas above the lighter. I think that is pretty neat since I couldn't see it with my eyes and only saw it in the picture after brightening it.



No, I did NOT intend my CA flag to be in the
reflection with the fire...

Eight armatures, handfulls of screws, and a
bunch of platters stacked together. Perhaps
even a partridge in a pear tree...

After I was done with all these drives, I decided to take 'em all apart. You can see the results to the left. I also had a few other drives lying around that I dismantled.

Perhaps in the future I will blow up one of my hard drives just to see how it handles that ;). Of course, I don't have anything with which to do that... Maybe a few firecrackers... Or even better, tape one melting in a blast furnace! Or, maybe i'll just go to bed now.

Or, perhaps, I could WRITE A SECOND PART!!!

Back to the Main Page!