HP-9825 computer cartridges

[Al Van Duren]

I am doing physiology research and just discovered a cache of raw data that was collected in the 1970s but transferred to HP-9825 tape cartridges in 1992. The data would be very useful to me, but I have no idea how to transfer it to media I could use now. Has anyone else ever dealt with this situation and developed a successful solution?

[alm]

I would look for someone that has a working HP 9825, or something with a compatible tape drive, that can transfer it to a modern computer using something like HPIB/GPIB or RS-232. See these videos by CuriousMarc: part 1 part 2 for someone solving a similar problem with 8" floppies. It wouldn't hurt shooting them a message to see if they're interested, although obviously they are not a vintage data recovery service.

[srb1954]

Curious Marc also did an extensive series of videos on repairing a 9825 and some videos on restoring tapes for the 9825 and HP-85.

Many of the tape drives in older HP equipment are unusable due to some of the drive mechanism parts breaking down and dissolving back into primordial goo. Some people offer a restoration service/kit for the tape drives to replace the damaged parts.

The HP branded tapes weren't much better having a reputation for shedding large chunks of the oxide layer on older tapes.

[Stray Electron]

  I still own over a dozen HP-9825s.  I've rebuilt the tape drives in many of them but was extremely RARE to find a tape that is still readable even 15+ years ago.  Among many other tape sources, about 20 years ago I found brand sealed box of 100 tapes at an USAF base in Texas. I tested about 5 out of the box and NONE were usable. I sold the others to collectors. The ones that opened them and tested them also reported the they were no good. I won't go into details about all of the problems with the tapes but IMO you're wasting your time trying to read those old tapes. 

  Not to mention that the HP-9825 runs HPL and not Basic.  The size and resolution of the stored numbers is entirely different on a PC or any other (non-HP) computer.  The Math is also very different.  It would require a MAJOR rewrite of the HPL code to make it run a different system and if you didn't understand matrix math and rounding errors and numbers overflow and underflow and a lot of other topics that your new code would probably have serious bugs in it.

  FWIW they only conceivable way to transfer your data would be to get a working HP-9825 and LIST the data or program code and re-direct the output to the (optional) RS-232 port and link that to an RS-232 on a PC and redirect that input to a file there.

   I haven't looked at it in years but you can see if HP9825.com is still around.

[Al Van Duren]

Thanks very much. I'll review the videos later today.

[Al Van Duren]

Thanks. I had read about the problems with the capstan and the instability of the oxide layer on the tape, so I don't have high hopes; however, it would still be great to have the data in a usable form.

[Al Van Duren]

Thank you so much for the advice and cautions. If I am successful, I'll post a short summary of the process.

[Stray Electron]

The biggest problem with the old HP tapes is that they shed the brown magnetic media very easily and that exposes the clear plastic leader. (And of course, loses the data from the file located where the gaps is.)  The tape drive sees the media gap and thinks that is the End of Tape hole in the tape and immediately stops reading and refuses to read any files further down the tape.  So you lose ALL of the data from the gap location to the physical end of the tape.

  Development of those tapes was a joint effort between HP and one of the big mag tape companies (I don't remember who at the moment). I have found a few of the identical tapes labeled and sold by the tape company but they are just as bad and have all of the same problems as the HP tapes and HP tape drives.

   Frankly I would LOVE for someone to find a solution to this problem or even just a way to interface a 9825 to a more modern mass storage system so that newly written SW for the 9825 could be saved and used.

[alm]

A possible solution I could see, though no doubts a huge effort, is to basically build your own tape drive that as carefully as possible (low speed / tension) will read and digitize all of the signal it receives, and then do the demodulation in software. This would allow you to read past gaps and would allow you to vary the tape speed: just add an encoder and store rotation speed with the data. But I'm sure there are a lot of details to find devils in.

It wouldn't surprise me if there are more people with this problem of recovering data from these tapes.

I believe there's been some papers about doing something like this on old, low-density, hard drive platters.

[Stray Electron]

   Actually someone has worked on this.  I haven't looked at HP9825.com in over ten years but I looked it up this morning and it's still an active website.  And it has links to this article and it mentions someone (Larry Atherton) who is modifying the original HP 9825 tape drives to use a more reliable tape.  That doesn't solve the problem or reading the original tapes but it is a step in the right direction.  I think I may contact him and have him perform this modification on one of my 9825s.

http://www.hp9825.com/assets/applets/Oak_Trees_Small_Branches_and_the_HP_9825.pdf

  As far as original tapes are concerned I think the best approach would be to to carefully unwind them into one long continues length and check every inch of them over and then run the entire length through a especially built machine and read the bit pattern. And then duplicate that on a modern solid state memory (preferably) or on one of the new tape cartridges. When hand checking the tapes, you would have to find a way to deal with lost media and any GOO found on the tape.

  A have a question for anyone out there. Has anyone ever seen a floppy disk drive connected to a HP 9825?  I understand that it was possible using the right combination of drives (HP 9885?) and optional interfaces and the optional Mass Storage ROM but I've never heard of anyone that tried it.   I still have a pair of 9885s squirreled away!   

[cruff]

Lots of discussion about this on the VintHPcom groups.io list. There is a 9825 tape dumping tool that runs on 9845 computers described starting at this post: https://groups.io/g/VintHPcom/message/7268?p=%2C%2C%2C20%2C0%2C0%2C0%3A%3Arecentpostdate%2Fsticky%2C%2C9825%2C20%2C2%2C40%2C88771981

There are also lots of other topics about the tape hardware and methods to replace the tension bands on old tapes.

[woodchips]

From memory the old HP tape cartridges were the DC100 type, is this correct?

I think I have a very old, early 80's, tape drive that used DC100 cartridges, somewhere. The tape drive was connected using the floppy controller, again from memory, and I used it for several years until I bought a DC1000 drive, the DC100 cartridges were then unobtainable. Physically the DC100 and DC1000 were the same, again from memory, but it was the media that was improved.

[alm]

From memory the old HP tape cartridges were the DC100 type, is this correct?

Looks like you're correct, good memory!
http://www.hp9825.com/html/dc100_tape.html