PDP-11/45: Data Recovery

Okay, the system is working well enough now to start attempting recovery and archive of the dozen or so RK05 packs that I have on-hand. These were all obtained (along with the RK05 drives, controller and power supply) in a surplus auction downstream of Stanford's Hansen Experimental Physics Lab, sometime in the early '90s.

The packs date from the mid-70's to early-80's, and the labels indicate contents related to experiments and research projects taking place at the lab at that time. One particular pack seems associated with the Stanford Gravity Wave Project, which built early resonant mass detectors. Other packs labeled "FEL" would be related to early free-electron laser research. Many of the names on the packs are are readily found on related scientific publications from the time.

The process for dealing with these packs involves opening them up for inspection and cleaning before mounting them, with hopes of avoiding beating up or destroying the drive heads and/or media with head crashes. Much has been written about pack cleaning on the classiccmp mailing lists and in the vcfed forums, but briefly the process involves some clean-room gloves, lint-free wipes, and anhydrous isopropyl alcohol. The outside of the pack is first cleaned of dust and grime, then the packs are opened and inspected and the disk surfaces given a scrub with the wipes and alcohol. If a pack seems in good enough shape to mount, it is spun up and run in a drive with head load disabled for about a half-hour. This gets a good air-flow over the disk to blow out remaining loose particulates and also lets the disk come up to thermal equilibrium. After that the heads are loaded, with a finger standing by on the unload switch in case there are any bad noises...

I've already been dealing with two of the packs extensively during the restoration: one is an RKDP diagnostics pack, and the other was a backup pack of same. I was able to capture a complete, error-free image of the RKDP pack using PDP11GUI. This seems to be an earlier version of this disk that what is already available on bitsavers; I've sent the bits to Al Kossow, but as I understand it his project has a big backlog at the moment so it may be a while before he can consider my submission. In the meantime, for those interested, the disk image is available here.

The RKDP backup disk was used as a test pack during the RK11/RK05 restoration work, and thus was overwritten by the RK05 diagnostic read/write tests. It now contains a bootable RT-11 image, written via PDP11GUI. Mixed results on the other packs so far: some have had severe head crashes (see pic below) or are otherwise damaged to the point that I am hesitant to mount them. Some have been mysteriously unreadable. It looks like I can expect about 50% recovery. Results so far are tabulated here. I hope to be able to make other recovered images available soon, but since they contain original research materials I am trying to contact the authors for permission first.