paul@devon.UUCP (Paul Sutcliffe Jr.) (09/06/87)
In article <8899@ut-sally.UUCP> nather@ut-sally.UUCP (Ed Nather) writes: > In article <1319@faline.bellcore.com>, gtchen@faline.bellcore.com (George T. Chen) writes: > > Help. I have an XT clone with a 20M hard disk problem. When I first turn > > of the computer after leaving it off for a while (like in the morning or > > when I get home from work) DOS is unable to read from the harddisk. When > > I try booting from a floppy, I'm still unable to access my C: drive. I keep > > getting a "Invalid drive specification" [...] > > > > What's going on? Could there be a problem where the harddisk actually > > needs the machine to be warm? > > The same set of symptoms appeared on my XT. After a lot of flopping around > which I won't describe, I removed the connectors to the hard disk and cleaned > all the contacts, which looked a bit scuzzy to me. I used a rubber eraser > till they appeared shiny again. When I re-connected them everything worked > and has been working ever since. I don't *know* if it's the same thing, but > it might be worth a try ... I've also seen the problem encountered by George, but my solution was somewhat different from Ed's. My system, too, started having trouble booting from the hard drive until it was warmed up. After a while, it wouldn't boot even after a several hour warm-up period. At this point, I decided to check things out. It would seem that because of repeated hot/cold spells (hot when running, cold when powered off), the socketed IC's worked themselves loose. After removing each board in the system and re-seating those IC's that were in sockets (I just pushed them down all the way), the boot-up problem was cured. This solution is similar to Ed's, in that the probable cause was insufficient contact between parts, with the solution being to re-establish good contact. Just thought I'd pass this along. By the way (flame-retardant suit on), I felt that George's problem was probably common to all micros so I have cross-posted this article, with the solutions used by Ed and me, to (what I considered to be) all the comp.sys.* groups involving micros. I appologize in advance if I have offended anyone in doing so. Please edit the Newsgroups: line if you wish to followup to this article in a system-specific way. - paul -- Paul Sutcliffe, Jr. UUCP (smart): paul@devon.UUCP UUCP (dumb): ...{seismo,ihnp4,cbosgd}!bpa!vu-vlsi!devon!paul