On 3/30/10, Vincent Slyngstad <vrs at msn.com> wrote: I might just have to do that. I have several ways to read papertapes directly into pre-Omnibus PDP-8s, but I'd have to think a bit on the best way to attach something to a "modern machine" to read tapes for uploading. I have an RS-232<->20mA adapter lying around, but no docs and I haven't yet reverse-engineered the schematic to ensure it's using the alternate EIA pins for current loop. That's one option. I'm sure I have an IBM 5150 or 5160 with an original serial card (that could support current loop), but it'd be at the bottom of a stack in storage. Building a current loop adapter from scratch might be the fastest path. I have an ASR33 and a DEC PRS/04 reader (both current loop, one 110 bps, one faster, IIRC). I'm implying that I don't know for certain that the tape I have is that exact tape. I have *a* tape. I haven't looked at it this year, so I'm allowing for fuzzy memory. I might have a tape containing a later version of the program, if there happened to ever be one. I have not. I just did a few minutes digging (beyond a one-line drive-by googling) to help the original questor. Doug would be an authoritative source, but in case he doesn't have a 40+-year-old program lying around in easy-to-access machine readable form, one could pursue the DECUS route as well. I don't have all that much of the older stuff myself - perhaps a handful of papertapes (less than 10) and perhaps a floppy or two. My only exposure to the vast majority of those programs is those same blurbs that are already well circulated. -ethan

On 3/30/10, Chris Elmquist <chrise at pobox.com> wrote: I have more than one reader - only the ASR-33s are mechanical. The one that's the best shot to adapt at this point is the PRS/04, a portable reader (it even has a handle) that was sold with machines of the PDP-11/34 vintage when diagnostics still came on papertape but customers might have bought a VT52 instead of a teletype. It is a current loop device with plugs to sit between the console terminal and the CPU on the same loop. It has a large (i.e. gentle curve) toothed feed roller (vs pinch roller), but is optical read. I'd post a link to a picture, but I can't seem to find any examples on the 'net. It's a little larger than an external 5.25" floppy enclosure with long (more than 2m) cables, white pebbled finish, small handle... Nice. -ethan

I restored one of thoise a couple of years back -- or actually the -A version which has simpler electronics (but much the same functionality). I got the HP9800 interface for it (HP11202 Opt A01 IIRC) on E-bay. As you probably know the outputs of the 2748 are pilled up to 12V, but through 10k resisotrs, whcih means they will feed the 180R/390R-terminated TTL inputs of the 11202 with no prolems. Of course the -B model that you have has clamp diodes anyway so you can force the outputs to TTL levels by jumpering 2 pins on the connector. What was wrong with yous? Mine had a bad mbearing in the magnetic clutch. T oreplace that I had to make special tools to press the clutch unit aprt and put it togetehr again. The ball race itself was a standard size and I had no problems getting one. Of course I also had to make the alignment mandrel to ge the clutch parts centred up wehn I put the motor unit back on. There are a few photos in my flickr account (tony_duell) of the internals of the read head and of taking the motor/clutch apart. -tony

THe -A model doesn't have the clamp diodes or jumper... HP connected the outputs of the 2748 stragiht into the TTL level inputs of a 11202 interface. The reson nothing got fried is that the 11202 has a 180 Ohm pull-up (to +5V) and a 390 Ohm pull-down (to ground) on each input while the output colelctor resistors in the 2748 (to +12V) are 10k. If you work it out, the votlage at the IC input is well within TTL specs. ~~~~~~~~~~~~~~~ I might have guessed.... Mind you, when I wanted RS232 output from my 2748, I used a set-up with a few _hundred_ ICs... The fact that there was an HP9830 calculator involved has something to do with that... Just after I got mine working, somebody listed on E-bay a 'DEC 2748 head'. It turend out to be the read head module for this reader. I bid on it, and fortunately nobody else was interested.... If you d oever take the read head apart, you will need a couple of alignment pis to get it back together. The reason is that the metal block that holds the photrasistors is not ground, it's connected to a supply rail (-5V IIRC). It therefoe sits on insulators. To gt everything alighed you put the alignment pins into a coule of blind holes inside the read head. slot on the insulator sheets, then the phototranssitor block, and then fit and tighten the 3 mouting screws (with insulating bushes). Then with everythinc clamped up you remvoe the pins. These pins are just 1" (or so) lengths of 1/8" diameter metal rod. A slight bevel on the end (trivial if you have access to a lathe) makes it slightly easier to assemble, but is not essential. There are pictures of this on flickr. And note this is not in any HP manual that I've seen... I still prefer the Trend USR and HSR readers with their self-setting read thresholds. -tony

What the heck is a digitised paper tape? Paper tape is digital anyway. -tony

OK... I was worried for a moment that somebdy had tried to archive paper tapes by scanning them a foot at a time on a flatbed scanner or something equally daft -tony

Once you get it digitized, would you mind if we suggested storing that file on paper tape?