Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!cs.umn.edu!ariel.unm.edu!chama.unm.edu!honeywel From: honeywel@chama.unm.edu (Honeywell Field Service) Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: IDE hard disk problems Message-ID: <1991May29.033504.8156@ariel.unm.edu> Date: 29 May 91 03:35:04 GMT Article-I.D.: ariel.1991May29.033504.8156 References: <3084@odin.cs.hw.ac.uk> <1991May28.115149.14644@esseye.UUCP> Organization: University of New Mexico, Albuquerque Lines: 61 Just passing on some information on disk problems. We had been experiencing some strange error messages on some of the AT class systems which would usually involve the disk or floppy. After some investigating it turned out that they all had one thing in common a mouse card with the IRQ jumpered to 2. It seems IRQ2 is somewhat special in the AT systems--they've cascaded two interrupt controllers together using IRQ2 and having something active on that interrupt caused the problems. I believe the error messages were soemthing the the effect that the disk wasn't ready and system hangs were not infrequent. I've been reading the articles on IDE and other disk problems with some interest. According to our depot repair people it does sound as if many of the problems reported are due to "Sticktion". (Hope the spelling is correct.) The symptoms are that the drive does not come ready, after investigating you will discover that the motor is not spinning. Repeated power up and down of the system may bring the drive back to life. The symptoms are also of an intermittent nature, i.e. the problem may go away for awhile. You may discover, as I have, that reaching in and gently moving the motor a fraction of an inch will cure the problem temporarily. What is apparently occuring is that the heads are actually sticking or adhering to the platter surfaces with sufficient force to keep the motor from spinning. The depot people reported that the problem seemed more severe with plated media. Some of the Seagate drives have experienced more problems with "Sticktion" than others, in particular the "100" series and the ST251 were mentioned. None of the manufactures seem immune to the problem and I was told that there seems to be a rash of problems of late involving IDE drives. (That was great news since I just got an IDE dri!) The only solution if you're experiencing the problem frequently is to replace the media--this means shipping your disk off to a repair center. They reported that sometimes just cleaning the heads seemed to cure the problem, but this is not something you can do at home. When I asked what was causing the problem, the answers were a little vague. The possibility of contaminant from the motor coating either the media or the heads was mentioned. I've been aware of "sticktion" problems in the past. When we first serviced what was then new technology CDC 600MB winchester disks we found out about "sticktion". The heads on those drives would stick if they were powered down for an extended period of time. The difference was that the motors had considerably more torque than our little 5.25 or 3.5 inch pc drives. If they stuck, the motor would just rip the head right off the gimbals and fling it around inside the HDA. From what I understood, "sticktion" has been around all along on the pc disk drives, just as it was on the mini's and mainframes. Out of 1300 PC's on service here, I'm definitely not seeing the problem that frequently. Most of these systems have 5.25 drives, we're just starting to install the 3.5's. The drives I have seen display this symptom had generally been in service for a couple of years and some of them were 5-6 years old, so I encourage you not to p -- ---------------------------------------------------------------------- HONEYWELL Third Party Computer Service -- 1(800) 525-7439 Mike Thomas, Senior Technician, Albuquerque, New Mexico honeywel@wayback.unm.edu (505) 888-5820