• Please review our updated Terms and Rules here

GPIB to Serial or parallel printer adapter with the PET

I have more information.

I received the first ADA1200 that I ordered. It came with a software tape from CMC. There are 6 "programs" on the tape.

It was a real trial to recover them from the tape (the dipoles must be fading) there were file name errors & load errors. But I used an old trick, by bending the spring arm in the cassette which forces the tape against the head harder (this increases the signal levels) I was able to recover all of the programs, they are:

CONTENTS
BAUD RATE
PLOT PROGRAM
FORMAT NUMBERS
SCREEN DUMP
DELAY ROUTINE.

The idea of the delay routine is to delay the serial data after a carriage return, long enough to allow the printer head to return to its home. This program is executed using SYS commands where the cassette buffers are.
 
The idea of the delay routine is to delay the serial data after a carriage return, long enough to allow the printer head to return to its home. This program is executed using SYS commands where the cassette buffers are.

Out of curiosity, does the delay routine send/have the option to send null characters? Some truly ancient teletype-style serial printers not only needed delays for carriage return, they wanted nulls to be constantly sent to keep the head advancing.
 
Out of curiosity, does the delay routine send/have the option to send null characters? Some truly ancient teletype-style serial printers not only needed delays for carriage return, they wanted nulls to be constantly sent to keep the head advancing.
As far as I can tell so far, there are no options, only to load it to the cassette memory area. There is an sys to turn it off or deactivate it. It just sits there, there is no on screen interactivity with it. But there might be a way. I still haven't tried printing anything..yet. The Imagewriter printer I bought got a real hammering (as things always seem to via the GSP) and it came with assorted broken plastic parts, so I'm fixing that up now.
 
More printer information and items have arrived now. I have repaired the Apple Imagwriter printer too. I have more documentation on the ADA-1200 too.

Also the interesting Taylor Wilson universal Pet printer adapter arrived. It has a serial and what looks like two sizes of Centronics printer connectors.

I opened it up for inspection, all British made with practically all IC's with an RS (Radio Spares components) logo on them, Tropical Fish capacitors in the power supply and all BA thread screws. It is in fact a 6502 computer with a crystal clock and quite a lot of support logic. It is going to take a while to study and document this creation. It clearly was designed specifically for the PET. One of the DIP switches is labeled PET/ASCII, presumably this is something to do with PETSCII ? (which I don't understand....yet)

I don't have the manual or schematic, sadly, but that hasn't stopped me in the past from copying the schematic out. I think this unit might just be the "master" PET printer adapter. Hopefully since it has its own "brain" and firmware in there, it won't require any support software.

(At the moment I'm working on another PET memory diagnostic project, so this Printer work will have to wait a little).

See attached photos:
 

Attachments

  • Print1.jpg
    Print1.jpg
    331.2 KB · Views: 19
  • print2.jpg
    print2.jpg
    254.8 KB · Views: 19
  • Print3.jpg
    Print3.jpg
    317.1 KB · Views: 19
Last edited:
It has a serial and what looks like two sizes of Centronics printer connectors.
One of the Centronics connectors has 24 pins. I'm thinking of an IEEE connector. My reasonings:
- I never saw a 24 pins printer connector. A 36 pins, yes. And the MSX has even smaller ones (IIRC).
- If the connectors were both printer connectors and therefore IMHO should have common pins, why do the cables end up at different sides of the board? And what about the different number of lines?

Have fun finding this one out :)
 
PETSCII / ASCII - enjoy!

Dave


For my part, I can document the schematic and the PCB, which could be ultimately be converted into new Gerbers. The power supply could be implemented with a modern SMPS.

Then it could be possible to recreate this device before (like so many other things) it gets "lost to the sands of time" and when somebody searches the net for it in 50 years it will apparently not exist. Unless we do something about it now.

So in the spirit of preserving vintage computer technology, I can do the hardware side of it. But it would require the firmware to be disassembled (not my wheelhouse) so as to create the original program, to figure out what they did to make it work. Could you do that part or Ruud can help?

If we (collectively) did that, we could also create a full manual for it, even though the original manual is probably already lost.
 
Last edited:
One of the Centronics connectors has 24 pins. I'm thinking of an IEEE connector. My reasonings:
- I never saw a 24 pins printer connector. A 36 pins, yes. And the MSX has even smaller ones (IIRC).
- If the connectors were both printer connectors and therefore IMHO should have common pins, why do the cables end up at different sides of the board? And what about the different number of lines?

Have fun finding this one out :)
Yes I think the small Centronics connector goes to the printer and the larger one must go (with a special custom cable that I don't have) to the port on the PET. I hope they stuck to a sensible scheme with the pin numbering.
 
Hugo,

Yes, what size of EPROM has it got so I can estimate the time it will take?

Dave

Doh, the writing on the 40 pin IC was difficult to read. Closer inspections shows it is a 6402 (a UART) not a 6502.......There is no ROM on the board. All 74LS logic and some CMOS, I expect the Xtal must be for the Baud rate circuit.

So there is no firmware to decode, only hardware.

I'm still somewhat baffled by the longer Centronics style connector, only 12 of its connections appear to be used, I thought initially it is the one that is supposed to be connected to the PET.. but The other Centronics style connector appears to match the standard Centronics printer cables I have seen which have the same connector on each end .

But it looks like that shorter 24 pin connector is connected mainly to the INPUTs if some 74LS14 gates, which makes it look as though this unit is trying to receive a Centronics style printer input and output it in serial or some other parallel format on the longer 36 pin connector, most of its pins are connected to the outputs of an LS244. But it is obvious to me now the 24 pin connector is for GPIB style inputs.

So it is now looking that the cable to connect this thing to the PET would be a 24 pin Centronics (IEEE-488) to an Edge connector type. I wonder where to find one of those ......(found one on ebay below)

The 36 pin connector would probably head off to the parallel printer with a 36 pin to 24 pin cable like this ? ( I'm very unfamiliar with these cables and connectors)




So it would be this sort of cable to connect it to the PET (costly is seems) :


(It is interesting that this Taylor Wilson unit is much more elaborate that any of the ADA units, it has 28 IC's in it (including the UART) !

So effectively it seems that this Taylor Wilson box is a IEEE-488 to IEEE-1284 adapter, with an additional IEEE-488 to serial adapter function.
 
Last edited:
Another thing I have found, unlike the ADA 1450, which has a male DB-25 connector (that will plug onto the the female connector of the Apple Imagewriter printer) for some odd reason the ADA 1200 has a female, and I don't have a gender changer handy so I will have to order one. So I have not been able to try out the ADA 1200 on the PET, yet.
 
Yes I think the small Centronics connector goes to the printer and the larger one must go (with a special custom cable that I don't have) to the port on the PET. I hope they stuck to a sensible scheme with the pin numbering.
Isn't that the other way around: 24 pins connector goes to the PET and 36 pins one to the printer?
 
Isn't that the other way around: 24 pins connector goes to the PET and 36 pins one to the printer?
Yes that is what I decided too.

I'm working withe the ADA 1200 for now. (as noted unlike the ADA1450, the ADA1200 has a female 25 pin serial output connector)

It works with the Imagwriter printer (with a gender changer) and I can print out program listings. The only serial printer I have is the imagewriter, and its connector input pin specs from its manual.

I also examined the way they had wired their female 25 pin female serial plug on the ADA1200:

The ADA1200's data output is on pin 3, which makes sense because the received data in on pin 3 at the printer.

Signal ground pin 7 as usual.

In the connector they have tied pin 20 pin 6 and pin 8 together, but this connects nowhere in the ADA unit. So DTR, DSR & DCT are connected together in the plug.

There is an input labelled DTR on the ADA1200's pcb, this connects to some gate inputs where their open collector outputs are in a wired OR config and connected to the NRFD connection of the PET IEEE connector. It looks to me that this would make the computer stand by and not send any more data unless the printer is ready for it.

This connection on the serial plug (labelled DTR on the pcb) is connected to pin 5 (which is CTS not DTR, so the pcb label and the pin name don't match). But also, pin 5 is not used for anything on the Apple Imagewriter printer and therefore it will assume a high state. But pin 4 in this printer, like a computer serial connector, is the output RTS from the printer.

Here is what I'm wondering; because if the ADA1200 was plugged onto a computer's serial link, instead of a printer, they have connected the "DTR input" for the ADA unit, onto the CTS input pin 5 of the computer serial connector, which seems odd.

For even a partial handshake RTS (printer to ADA unit) shouldn't they have connected what they have called "DTR " on the pcb to pin 4 of the serial connector to RTS at least ?

I wonder why they didn't connect what they labelled as DTR on the pcb to pin 20 of the connector, as one might expect, because at the printer end DTR is an output on pin 20 and at the ADA1200 end what they labelled as DTR on the pcb is an input. ?

( I know there are some oddball variations of serial cables both with gender and wiring that is part way between a Null modem and straight cable, it was like this with Votrax's Type'nTalk, so it could be another example of that sort of thing. I could probably figure is out if I had the schematics of other vintage serial computers to see what they use pin 5 for, because as I say there is no connection to that in the Apple Imagwriter and this ADA unit was designed to plug onto a printer's serial port (not sure which printers), not a Computer's serial port, but if it is, it still would pass data directly to it on pin 3 without the need for a null modem cable.)
 
Last edited:
I can confirm that the ADA1200 also successfully sends serial data to a modern computer (running TeraTerm) and it is possible to export a BASIC program listing, provided the PET has been poke'd into lower case (othewise it comes out as a mix of graphics symbols) but in the case of the printer, it prints correctly without having to do the poke . So the ADA1200 is a quick "one way" serial link.

( I believe making a fully functional bidirectional serial link with the 2001 PET requires firmware in ROM and the hardware interface)

But at least, even without the actual vintage serial printer the simple ADA1200 unit means its easy to get programs printed out via a modern computer.

The problem with this one though (unlike the ADA1450) it hogs the PET's IEEE connector, so at the moment I cannot use the SD2PET at the same time until I make an adapter.
 
Another Printer related question I have, in case anybody has any experience with this:

There are a number of good looking printers available, probably 1990's 2000 vintage, some NOS, with both serial and parallel inputs. Most come with software driver disks, containing drivers for "modern" operating platforms.

In the case of the vintage Apple Imagewriter.........it just prints the serial data you feed it on its serial input port and no support software is required. So it works fine with the PET and the ADA1200 adapter.

Would a modern serial input printer be likely to be able to just print serial data it was sent like this (assuming the baud rate was correct), or would it be expected to be stone dead in the water without its software driver running in a compatible host operating system ? Has anybody tried getting modern printers to print without their software drivers ?
 
With very rare exceptions a printer that has an RS-232 serial port is likely to be a pretty bog-standard character oriented printer little different from an Apple Imagewriter; throw text at it and it’ll print it. (You only need a printer-specific driver to print graphics, switch fonts, fancy format, whatever.) Epson and Okidata still make printers like this for industrial applications (tractor-feed multi-part forms are still a thing some places) and they usually emulate similar printers made in the 1980’s.

Parallel port printers are a little bit more of a crapshoot. Dot-matrix and some inkjets will, again, probably be capable of emulating a printer from the 1980’s and will be willing to just stamp out text from your PET. Laser printers and fancier inkjets might be either too smart (IE, they speak a more elaborate page description language) or too ‘dumb’ (some late-90’s-into-the-aught “Winprinters” predating USB rendering Centronics obsolete) to use, at least easily, with adapters like what you have.

For use on a PET… honestly, there’s little point upgrading? Ribbons for the ImageWriter and a lot of other ancient printers are readily available and the old printers are probably built more solidly.
 
… FWIW, it is *possible* to just dump dumb text at some laser printers and have it “just print”. HP Laserjets compatible with PCL will happily just default to filling a page with Courier 10 text and spit it out either after receiving about 60 lines (with the default margins) or after receiving a form feed code. (ASCII decimal 12, I think?) Probably less optimal with a PET than a line printer, but you might have some use for it.
 
Even with the 'really smart' printers - they still worked with a 'generic' serial or parallel printer driver.

In fact, this was the only way that some printers were supported by the Operating System. Either a 'new' Operating System required new drivers to be written by the printer supplier (so the Generic printer driver had to be used until they wrote a proper Printer Driver) or the printer manufacturer just stopped supporting the old operating system.

When my mother in law purchased a new computer and printer - this was the only way I could get the printer to work (by using the Generic printer driver) until such time as the printer supplier got around to producing the correct driver. Of course, all sorts of new features were then available with the correct driver - but that is another story...

Dave
 
When my mother in law purchased a new computer and printer - this was the only way I could get the printer to work (by using the Generic printer driver) until such time as the printer supplier got around to producing the correct driver.

FWIW, if this is recent Windows you’re talking about the “Generic” printer driver can mean Generic GDI driver, IE, the generic bitmap ”Winprinter” driver. And a pure native Winprinter is the one type of printer that’s uniformly useless on vintage computers. (Most Winprinters are USB, of course, but like I said, they did make some parallel ones into the mid-2000’s.) A pure Winprinter has no text mode, it only outputs bitmaps. Some have backup emulations like a low-level PCL3 mode that can do text, but you can’t count on it.

And to chuck out an example of a “too smart” example that might actually have a serial port, a pure Postscript printer also wouldn’t do useful things if you just spewed raw text at it. But “pure” postscript printers are kind of rare, I think even the Apple LaserWriter has a brain-damaged Diablo daisy wheel emulation that can handle plain text hidden in it. (Not positive it auto senses, however.)
 
Back
Top