The 2748A/B paper tape reader uses the "8-bit duplex register interface". That would lead me to believe it's an 8 bit interface, not a 16 bit one. I have full docs (& schematics I believe) for both the 2748B and the 8-bit duplex register interface board. I have two of those units presently running, so I can get the cable pinout too if you'd like (but I believe the cable diagram is in the interface manual). Let me know if you still need them. One word of caution about the facit 4070 punch... as I found out when I got mine - the facit had a board in it that was replaced depending on what electrical spec the system you were attaching it to used. I seem to recall that you had to specify which (of many) interfaces you wanted on the punch when you ordered it. So just because you find a 4070 punch doesn't mean it has the right board in it to hook up to the 8-bit duplex register board (hp punches and readers both used the 8-bit duplex register board). Matter of fact, unless it was found with an HP system, it probably DOESN'T have the right board for the HP. You can't tell from the outside either, as I've seen the RS-232 interface board in the punch presents a DB25 externally - just the same as the HP version of the card does (which is decidedly NOT RS-232). Jay West ----- Original Message ----- From: Tony Duell <ard(a)p850ug1.demon.co.uk> To: <classiccmp(a)classiccmp.org> Sent: Friday, June 29, 2001 12:41 PM Subject: Re: HP2748A papertape reader : connector ? have 82032 character-based came Facit the

Tony Duell wrote: This is incorrect! The HP tape reader uses a HP tape reader board that is unique, it is NOT the HP 8 bit duplex register board. If need be, I can run downstairs and get the acutal part number. I've jsut recently gotten my 2114A booting HP basic from my new HP paper tape reader emulator hardware. I'm pretty familiar with the tape reader interfaces. What I need, are the pin-outs for my old Decitek tape reader!

I walked out my door this afternoon to find an Apple Thermal Transfer printer sitting next to the garbage can outside of our building. It's model# A9M0306 and while I've seen plenty of Imagewriter's and other Apple printers, this is the first time I've come across one of these. I almost passed it up because at first glance it appeared to be just another 80 column dot matrix. Two things that caught my eye were the old multi-color Apple logo and the serial cable attacbed to it. It's missing whatever head/ribbon would be installed and I rescued it just minutes before a thunderstorm moved through. If anyone is interested in it, you're welcome to it for the cost of shipping. Otherwise I'll likely hold on to it. Jeff -- Collector of Classic Microcomputers and Video Game Systems: Home of the TRS-80 Model 2000 FAQ File http://www.geocities.com/siliconvalley/lakes/6757

Some thermals use the heat of the printhead to blacken the thermal paper, others use the heat of the head to melt off the coating on a plastic ribbon (like the old Commodore color printers) - some can do either, such as Zebra, DataMax, etc. The serial printers that UPS accounts use to make shipping labels have thermal labels. The Zebra/Datamax types that I've worked and used work well with either a ribbon or thermal labels, mostly for barcodes. -> -----Original Message----- -> From: owner-classiccmp(a)classiccmp.org -> [mailto:owner-classiccmp@classiccmp.org]On Behalf Of Edwin P. Groot -> Sent: Wednesday, July 04, 2001 2:15 PM -> To: classiccmp(a)classiccmp.org -> Subject: Re: Apple Thermal Transfer printer -> -> -> Good rescue! Why would a thermal printer need a ribbon? All the -> printers for my HP stuff just need thermal paper rolls. -> Edwin -> -> At 05:40 PM 6/30/2001 -0400, you wrote: -> > I walked out my door this afternoon to find an Apple Thermal -> >Transfer printer sitting next to the garbage can outside of our -> .... -> >serial cable attacbed to it. It's missing whatever head/ribbon would -> >be installed and I rescued it just minutes before a thunderstorm -> >moved through. If anyone is interested in it, you're welcome to it -> > -> > Jeff ->

Also the horrid IBM quietwriter series I, II and III. If the head wear is excessive also worn brake pincher, the ribbon will snap on these IBM quietwriter series. Prints out laser-quality text and graphics especially on III. The speed I spell " SLOW" and table is quake hazard for anything sitting on it where that printer is placed. :-) Yesterday I saw 8 color text IBM dot matix wide printer w/ parallel interface. Anything special about it and driver available on most OSes? What about finding 4 color ink ribbon cartridges? Cheers, Wizard

Not that kind of impacters. This print head is removeable has lot of electrical contact pads uses finely spaced traces that presses on the carbon-based ribbon. The short circuits it creates allows print to leave behind on paper because heat melts the ribbon. Whole "print head" thing has 2 motors, one to keep ribbon moving in short bursts, one to move the head in and leave there (during printing) That motor also lifts printhead off during non-printing also for reservsing back to start of line or to print again on specific areas. This motor doesn't only moves the head, it also unpinches both roller ribbon pinch and the brake as well when head is lifted. Printing mechanicals is *complex* and heavy that why printer lurches back and forth while it whips the assembly back and forth in 1 or 2 inches on certain characters before going on one direction then lifts off, advance the paper, whip the thing back to the start of new line to print. It prints one direction only. The contact pads is appox 50 by guess. The daisy wheels doesn't leave laser-quality characters. This one did which is amazing at that time back then. Cheers, Wizard

Bob Shannon wrote... the HP Bob is incorrect for the 2100 as well as the entire 21MX M/E/F (and HP1000) lines. He could very well be right on the 2114/5/6 cpu's though. The HP 2748B paper tape readers were connected to an HP2100 or HP21MX M/E/F system via the 8-bit duplex register board. No other interface was used, at least for sale in HP configurations. Matter of fact, if you bought the paper tape reader kit (the part # escapes my mind) which included the paper tape reader, the HOST interface board, cables, manuals, and diagnostic, there was only one interface you would ever get. The 8-bit duplex register. This was the HP designated interface for the 2748B. I can only assume Bob is referring to the older 2114 (not actually an HP machine other than the label), 2115, and 2116 machines which I'm not as familiar with. I have however, seen 2748B paper tape readers hooked up to a 2116 with an 8 bit duplex register so I know that to say in general terms that the HP tape reader board is unique and not the 8 bit duplex register board isn't accurate. But I can't speak from experience on the 2114 or 2115. Of course, the 2114/5/6 are NOT 2100 machines, and of course they're not 21MX M/E/F systems either. So, perhaps early 2114/5/6's had a modified paper tape reader and a unique interface at that time, but certainly not after the 2116. Jay West

Bob wrote... documentation In system. No Kidding!? That's truely interesting. You have my curiosity up, I'd love to know what the deal is for my own future reference. I do have a lot of sales literature and marketing collateral for these systems, and all of them show the 2748B paper tape reader as using an 8-bit duplex register board. Hummmmmmm.... You know what - is it possible that the ones that use this "tape reader interface board" that you've seen are the 2748A reader and the B version uses the 8-bit duplex register board? All the literature and docs I have shows the 2748B, I have nothing that relates to the 2748A which is why I'm conjecturing this. One wouldn't think that HP would release a board with the same electrical specs as the 8-bit duplex board, so wouldn't you think the electronics in the reader would have to be different for the two boards? Most interesting... could you check and see if your reader is the A one that uses this board Bob? Or maybe show us a pinout of the old style reader interface vs. the new style? small I knew the first 21xx machine was already designed and put together by another company, and HP just absorbed the company. When they saw how sucessfull it was, they then came out with the other two. I was thinking this first one was the 2114, and the HP made ones were the 2115 and 2116. I stand humbly corrected - must be the 2116 that was absorbed and they made the other two. The 2100 did come later. Interesting tidbit - one of HP's largest users of the 2100 system was a company called "MeasureX". This company had it's own engineering staff and redesigned the 2100 to be a single cpu card. They sold this design back to HP and it became the 21MX line. The MX in 21MX stands for MeasureX. I got this directly from a phone call with the former VP of engineering of MeasureX, so I suspect it is accurate historically. I'm just wild guessing here, but is it possible that the 2100A that you got had the tape reader (old style) added later and wasn't actually sold that way from HP? The reason I'm wondering this is I have the sales/ordering guide for the 2100A and the 2100S, and both only specify the 8-bit duplex register. Most Strange! My earlier comments were based on the HP sales/ordering guide, and in retrospect you're quite right, those guides were almost certainly revised from time to time and earlier ones than mine might indeed use the old style reader board. Again, I'm wondering if it's not the A reader that used it? interface site wrong or drawing incorrect conclusions): The 2100 machine is a 2100 machine. 2100 was not a family or class of machines, it was a specific machine (2100 with option pack A or S). There is no difference in the A and S versions except the S version included options as standard which were also available not referred to as "2100's", they were referred to as 2114/5/6. Part of my basis for this is logical - they wouldn't name the 2100 (which came later) a 2100 if that was the generic family name for the 2114/5/6 machines I'm guessing. The early nomenclature of HP's mini products is *SO* confusing... cpu's that were called 21MX's were later redesignated 1000 machines, etc. Plus the whole 2000 thing that doesn't refer to a computer, but rather a specific configuration of both hardware and software (for TSB). What a mess. What's your take on this Bob? Regards, Jay West

Tony wrote.... and also interesting? Several items of note.... The facit 4070 punch, when used on an HP 2100 or 21mx M/E/F system is called (rebadged) an HP 2895B. It's entirely possible that the 4070 was used on other later (different architecture, non 21MX) HP's with a completely different interface and thus those might very well use the standard 4070 jumper board when used with the non 21mx hp systems. When used on 2100&21mx systems, the 4070 is fitted with a board that is not a simple jumper board. The HP-oriented board inside the 4070 was definitely a full lenth board, and there were electronics on it. A very sparse set as I recall, it could be hobbiest-made quite easily. I think a handfull of transistors and perhaps a few caps. If you're curious, I can pull the card out and photograph it for you. It's definitely not "UART-ish". Jay West

This is interesting! I have never seen a 2748 interfaced to a 8-bit duplex register board. This would take a special cable, quite different from my reader cable. The HP tape reader board I've always seen connected to the photo tape readers is P/N 02116-8002. Its simply labelled "Tape Reader", and has a two row of 11 DIP packages, 8 two-transistor data receivers, and a 7 transistor driver-receiver section for flag and control. Because the HP processor has no need to send data to the reader, the more complex and expensive 8 bit duplex register is overkill. But the signal levels may be correct, if you have the right version of the duplex register board... The HP Tape Reader 02116-8002 board can only send the state of the control bit to the 2748 (or 2847) reader. It accepts 8 data bits on pins 1-4 and A-D, which is not compatible with the 8-bit duplex registers pin assignments. Jay West wrote:

On Fri, 29 Jun 2001, you wrote: The HP2748 I rescued has a very old style 50 pins connector, in 4 rows 13-12-13-12 pins. Only 11 pins are used, so it is bound to be an 8 bit interface. With the info give here that it is 12V negative logic, I should be able to find out the connections and wire up an interface board to the PDP/8E. I did not take the Factit 4700 because I already have 2... They come indeed with many different boards. One has a RS232 interface, the other has a 8 bit parallel interface that I hacked to be centronics -like. This way I could punch some tapes from the PC to be used on the 8/E Regards, Jos Dreese

A non-text attachment was scrubbed... Name: not available Type: text Size: 1087 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/20010701/94dcf18e/at…

I have a HP7970E. Where are you located Jay? Jay West wrote:

More info.... Just was digging for the repair/maintenance manual on the 2100 power supply (see previous message *GRIN*)... Came across a manual I had forgotten I had... The "1970 HP measurement, analysis, and computation reference catalog". It has the 2114/5/6 B version systems in it. It does specify there is an 8-bit duplex register board, so that board existed at the time. The paper tape reader isn't a 2748A/B variety, it's listed as a 2737A punched tape reader using (included) a 12532A interface. The 12532A obviously isn't a dup reg board :) (the kit listed at $2100 by the way). This catalog also has the 2000A timeshared basic system (using a 2116) and it specifies a paper tape reader is included, it's also the 2737A with a 12532A interface. Is this 12532A interface that I've never heard of the original tape reader interface you're referring to Bob? I'm not familiar with the 2737A reader, is it possible the 2737A used the old style interface and the 2748 used the later one? Surely someone knows this for sure! Here's some rather historically interesting l970 list prices for those fellow HP'ers on the list: 2116B cpu, $24,000 (8K) 2115A cpu, $14,500 (4k) or $19,500 (8K) 2114B cpu, $8,500 (4K) or $13,000 (8K) 2773A drum memory with 2776A power supply, $23,500 (393,216 16bit words, transfer rate 120Kw/sec, access time 8.7ms average) For those on the list into the 9100 calculator/computers, 1970 list price for the 9100A was $4400 and the 9100B was $4900. Jay West