Re: Mike Loewen / was: Re: [midatlanticretro] Can anyone read a CDC 9877 disk pack?
- Hi all,
This is primarily directed at Ian, but just in case anyone else can help to:
So I actually acquired a 9762 disk drive (one of the super early models apparently - the install date on it is 6/7/1976!). I cleaned up all of the foam stuff (you were right, that stuff is a mess!) and pretty thoroughly cleaned out the drive. I also cleaned the filters and things, and then let it run for a while, just to make sure that no dust or anything was getting into the drive cavity.
My drive unfortunately had a short on one of the logic cards on the +20V rail, but I was able to swap it out for another identical card, and then the fault light went away.
I also have a TB-216 tester, which I'm attempting to use to test the drive with (the actual drive only has two buttons on the front of it - 'start' and 'fault clear', as well as corresponding LEDs for those two, and then a third 'ready' LED). I have the tester cabled to the drive, as well as a terminator plug thing on the empty port on the drive, but I can't actually get the tester to recognize the drive. So, a few questions:
1. Any reasonable strategies for debugging what appears to be a problem with the unit-select hardware? Also, there are some options for selecting different 'flavors' of the 9762 (NRZ or MFM, standard or 'mux' cables, index/sector on A or B cables) . . . any idea how I can tell which drive I have?
2. Do you happen to know of any sources of documentation for these types of drives other than bitsavers.org? The documentation up on there is actually for later models of the 9762 which has completely different electronics.
Any advice/help would be welcome.
-ChrisOn Tue, Jan 4, 2011 at 8:39 AM, Mr Ian Primus <ian_primus@...> wrote:--- On Tue, 1/4/11, Evan Koblentz <evan@...> wrote:The host computer is irrelevant. What you *need* is the disk drive. These CDC drives have a standard SMD interface (usually). You'd probably want to be using a machine that'll have some other means of communication with the outside world, like a Vax or something, so you can get the data off the pack and onto other media. And if you're trying to archive data from an important pack, you don't want to go tossing it into an unknown drive. These disk drives are fragile, and require a lot of service.
> > http://www.bitsavers.org/pdf/cdc/discs/brochures/ProductLine_May73.pdf
> DEC? No, we're talking about a Data General Eclipse
> S/130, circa 1977:
I have a few 300mb CDC drives - let me tell you, there is no way a drive that's just been sitting for 20 years is going to fire up and read a pack without crashing. This is a hard drive we're talking about - the inside of the drive where the pack goes needs to be immaculate - no dust, hairs or particles. Any junk in there will surely cause a head crash.
To make matters worse, in the their infinite wisdom, Control Data used that sound-deadening foam, you know, the stuff that turns to crumbly crud? It's lining the entire inside of the cabinet - but even worse, it's inside the cover that goes over the head stack. The entire drive will need to be cleaned, and all that foam removed, before it's going to operate reliably.
Granted, my experience is with the 300mb drives - the 80 meg drives might be different, I don't have one to compare. But the general procedure for resurrecting one is:
-Clean drive, remove foam, clean, clean, clean.
-Replace filters - while you've got them out, clean all the air circulation ducting.
-Power up the drive and let it run for 24 hours to "purge" - the air circulation runs, and any loose stuff gets caught by the filters.
-Put a scratch pack in, start it spinning, and let it spin for a good long while - hours - you want to purge the crud off the pack too
-cross your fingers and hit the head load switch.
And, powering the drive is going to be a problem too - the 300mb drives at least require 220v.