HPIL junk wanted (with e-mail address)
#1

Hello, folks;

forgive-me posting the messae again; that's just because I forgot the e-mail address for answers.

I suddenly can use an HPIL in the HP41. Amazing! After more than 18 years waiting for it, I finally got one. Thanks (many thanks) to one of the contributors in here.

I became ambitious (in the very best sense) because I remember reading a lot about the HPIL and delving into my own brainstorms (WOW!).

At that time, the HPIL Kit was a must to developers. I know it no longer exists, and even the ones sold were used, of course.

I just want, from any sort of HPIL junk, both:

IN/OUT connectors
Pulse Transformer

I believe many things can be done with these...

PLease, if you have something, e-mail me.

#2

A kludged way of getting the connectors it to cut an HPIL cable in half and solder the cut ends to your circuity. Not as neat as the proper panel-mount connector, but it'll work.
As regards the transformer, electrically-identcial transformers were used in _every_ HPIL device I've worked on, even things like the HP71 IL module. So you can raid a broken HP71 IL module (or whatever) for the transformer. One day I must disect the dead transformer in my junk box (shorted turns) and count the turns, etc, so I can attempt to make a replacement.
Do you have the 1LB3 HPIL interface chip, or are you planning on managing without it? HPIL is not trivial at the hardware level, so I'd try to get a few of those chips as well.

#3

Hello, Tony; thank you for your answer.

Tony wrote:
<< A kludged way of getting the connectors it to cut an HPIL cable in half and solder the cut ends to your circuitry. Not as neat as the proper panel-mount connector, but it'll work.>>

Yes, I'm lucky having a spare cable, but I'll try keeping it intact (cutting it will still allow me to connect more than one only IL device in the loop, I know; I just want to keep it intact). If I can find an unused I/O connector, better.

<< As regards the transformer, electrically-identical transformers were used in _every_ HPIL device I've worked on, even things like the HP71 IL module. So you can raid a broken HP71 IL module (or whatever) for the transformer. One day I must dissect the dead transformer in my junk box (shorted turns) and count the turns, etc, so I can attempt to make a replacement.>>

I know about a guy that built an HPIL transformer. I'm no longer in contact with him (I saw him by the last time in Petrópolis, Rio de Janeiro, in 1989), but I remember he told me he used three toroydal ferrite cores and computed the number of turns based on the IL signal's frequency and waveform. I was not smart enough to ask him for the "recipe", so I'll try to do my best. Anyway, if you count them down...

Hewlett-Packard Journal, January 1983 issue, is a very good source of information. It has a brief, concise description of the electrical parameters and a simplified schematics of the IN/OUT IL stage. There are indeed three transformers, and their primary-secondary relation (2:1 / 5:43 input, 3:1 output) is mentioned. If I have at least one number of turns for each...

<< Do you have the 1LB3 HPIL interface chip, or are you planning on managing without it? HPIL is not trivial at the hardware level, so I'd try to get a few of those chips as well.>>

Yes, I know. In the same HP Journal issue we can see the internals of the HPIL IC, the ILB3. Somewhat complex fellow, han? Well, I believe for the time it appeared there was no other "easy" way to achieve an HPIL communication. But I'd like to know if there are no other controllers nowadays that would allow the IL protocol implementation easier. I thought about ATMel controllers (if I am not wrong, Pavel Korensky is/was working with them), or some Microchip PIC's, I don't know. My starting "project" would be simply a dummy buffer, that would capture any group of bytes from a starting code (byte) to an ending code (byte). Just a single binary, ASCII-coded character sequence that would not harm any other device in the loop. I believe it's easy to arrange, mostly because I have an Extended IO module.

This buffer would allow a dot-matrix, graphic LCD to show images and text sent from an HP41. Basically a record-only RAM and a microcontroller dedicated to IL comms, being continuously scanned by the LCD's controller.

BTW: I was told you, Tony, have some programs that allow LIF disks to be read (manipulated too?) by Lynux-based machines. Do you have them? Are they available to download? If not, can I have a copy of them? What to do?

Well, I wrote too much. Is there anything else somebody would like to add? Please?

Thank you.

#4

Tony said in another thread that HP2225A HPIB printers have an HPIL chip inside. I remember noticing a couple of years ago, when I was looking for a serial Thinkjet on Ebay, that the the HPIB ones sold very cheaply.

#5

You should try to obtain copies of the documentation that came with the 82166C kit. I think most of it is on the MoHPC CD-ROMs. There's a very detailed spec of HPIL (complete with all the state transition diagrams), a data sheet for the 1LB3 chip (which includes the application circuit), another 'user guide' for the components in the kit (which essentially gives the schematics of the 82166A module) and a guide to writing the device's firmware.
Alas the spec of the transformers isn't in any of those documents (well, the fact that there are 3 transformers, that some windings are centre-tapped, and how to connect them up is there). So thanks for the turns ratios -- it'll be a start.
My suggestion of cutting an HPIL cable in half was simply made because (at least over here) HPIL cables are much more common that spare I/O panel connectors. Don't wreck your only HPIL cable, though.
COncerning LIF disks on linux. Yes, I have written programs to do this. They are very much incomplete (I add more when I get the chance!), and there are almost certainly bugs in them (again, I fix those when I find them -- I do use these programs myself). I have no idea if they will work on any version of linux (they certainly depend on a PC's disk controller, so they are _not_ going to work on non-PC linux systems), but you can try them.
The programs are available as GPLed C sources (and a Makefile) from the HPCC web site (http://www.hpcc.org/) -- I think following the links for HP71 or HP75 will find them (!). If you have trouble, I'll send the latest version to you.

#6

Re Tony's comment on the subject of HP-IL transformers being the same in terms of electrical function are there not different versions for the different IL Chips.

The 1LB3 seems to use the 3 toroid design on a sort of DIL header as in 82163,5,6 18169 (most common) but in the HP3421 and 3468 the HP-IL chip is the 1LR4 and the transformer is in a reddish potted block. The schematic shows it to contain single transformers for Tx & Rx. They all seem to be made in Mexico and although they are proprietary HP parts the 3468 quotes Pulse Eng Inc as the manufacturer. In my 82169C the chip is the 1LB4 and the transformer is again in potted form and made in Hong Kong. Are there any other variants?

Among other items the 82166C Kit contains a module for the HP41 to assist in HP-IL development and provides a 'scope' function whereby messages going round the loop are delayed and displayed on the 41 for a time and then passed on.

How critical are the transformers? I did a quick experiment and made a 1:1 (10t:10t) txfmr using a 2 hole ferrite core from the junk box and inserted it in the loop between a 71 and 9114. Loop integrity was maintained and files could be read and written to the disc (2m cables)

#7

Hello, Nigel;

thanks for the information. Would you be as kind as for posting the txfmr core dimensions and wire specs? Did you use each hole for each coil? Did you try one single-hole txfmr? Sorry for bothering you with as many questions. With the specs for the ferrite core and wires I'll try the single hole transformer with two coils.

The ten turns for each side is a very good starting reference. Was it a guess or you considered frequency, waveform and electrical parameters?

Let me (us) know if anyone else is doing something on this area. HPIL protocol is somewhat complex, and for the control level it allows to, it is not expensive. It's (now-a-days) just slow.

Thank you all again.



Possibly Related Threads...
Thread Author Replies Views Last Post
  An easy course in using the HP16c-wanted- Frank 4 1,388 10-29-2013, 11:50 AM
Last Post: Mark Hardman
  New website address for Meindert Kuipers' MLDL2000? Garth Wilson 2 925 08-11-2013, 01:43 PM
Last Post: Meindert Kuipers
  HP41 Functions Address Table (F.A.T.) Antoine M. Couëtte 6 1,522 07-21-2013, 02:48 AM
Last Post: Antoine M. Couëtte
  HELP WANTED ON ALGORITHMS Joerg Woerner 19 3,325 04-27-2013, 12:56 PM
Last Post: Eric Smith
  HP 39gii Help Wanted: Inputting Strings in a Program Eddie W. Shore 0 579 02-09-2013, 02:17 PM
Last Post: Eddie W. Shore
  Apologies on mail delay. Diego Diaz 0 519 11-07-2012, 02:27 PM
Last Post: Diego Diaz
  CAS beta testers wanted Oliver Unter Ecker 5 1,470 07-30-2012, 07:11 AM
Last Post: Oliver Unter Ecker
  HP75D with HPIL Karel Kupka 2 894 07-13-2012, 02:57 AM
Last Post: Karel Kupka
  Help wanted SR-52 nina scholz 29 5,147 05-16-2012, 07:48 AM
Last Post: Anders Persson
  Hard or wanted CAS stuff Oliver Unter Ecker 9 1,992 04-16-2012, 04:32 AM
Last Post: Mike (Stgt)

Forum Jump: