HP-19C Repair Odysee

[PANAMATIK]

Hi All,

I want to present you a small HP-19C repair story here in the forum, which I and a friend had experienced recently. I have provided it as .pdf file with images for download.

I hope you enjoy the text while reading. In any case we had a lot of fun at "one of the best birthday workshops ever".

Bernhard

[Sylvain Cote]

Very nice story Bernhard, thank you for sharing.
You convinced me to wait after my retirement to try to repair my broken Sting calculators stored in my defective boxes. (1x HP-19C & 1x HP-10)

[rprosperi]

Amazing and epic repair adventure Bernhard, like so many you've shared here. Even attempting such an adventure takes courage, daring and extreme patience plus of course immense skills. It is truly awesome the repairs you will undertake. Grafting the print-head from a 97 into the tiny 19C printer is something I could not imagine one might even consider.

Your 19C print module of course remains available, you need only beckon and it will be sent as agreed.

Indeed, properly closing a 19C is the most difficult task I've faced in the HP repair world! I'm also glad to see your comments that sometimes you lose patience and get angry and that makes it harder - good to know it's not only me that loses patience when attempting these intricate procedures....

Of course the horrible experience is not all bad - it may preserve the ability for the working print-head to remain near NY.

Thanks Bernhard!

[jjohnson873]

(03-01-2021 12:31 PM)PANAMATIK Wrote:  Hi All,

I want to present you a small HP-19C repair story here in the forum, which I and a friend had experienced recently. I have provided it as .pdf file with images for download.

Bernhard

Bernhard...thank you so much for the insightful document on HP-19c printhead repair! I have a few HP-19c calculators that are in my "repair bin", ready to be diagnosed and the printhead has been a real issue. It appears that there is no easy solution for a dead printhead pin and no ready replacement. It's unfortunate that HP didn't make a huge amount of spare printheads for the HP-19c. Perhaps they are sitting in a warehouse somewhere, waiting to be discovered. I can see that this will be a future challenge to see what can be done for a dead/failed printhead pin. ~ Jim J.~

[PANAMATIK]

(03-01-2021 02:55 PM)rprosperi Wrote:  Your 19C print module of course remains available, you need only beckon and it will be sent as agreed.

Of course the horrible experience is not all bad - it may preserve the ability for the working print-head to remain near NY.

Thanks Bernhard!

Hi Bob,

You don't need to fear that I want the printer module back soon. Although my corroded HP-19C meanwhile does work partially, the display lights up and shows "0,00". I could replace the totally destroyed original 15-digit display by a 12-digit Woodstock display and compress the display format with my new ACT. But I have to build a replacement for the PIK chip to get the keyboard and printer working.

The PIK chip is also present in the HP-9x machines. Would it be worth to develop a PIK chip? Is there a demand?

Bernhard

[Paul Berger (Canada)]

(03-01-2021 04:26 PM)PANAMATIK Wrote:  

(03-01-2021 02:55 PM)rprosperi Wrote:  Your 19C print module of course remains available, you need only beckon and it will be sent as agreed.

Of course the horrible experience is not all bad - it may preserve the ability for the working print-head to remain near NY.

Thanks Bernhard!

Hi Bob,

You don't need to fear that I want the printer module back soon. Although my corroded HP-19C meanwhile does work partially, the display lights up and shows "0,00". I could replace the totally destroyed original 15-digit display by a 12-digit Woodstock display and compress the display format with my new ACT. But I have to build a replacement for the PIK chip to get the keyboard and printer working.

The PIK chip is also present in the HP-9x machines. Would it be worth to develop a PIK chip? Is there a demand?

Bernhard

I have been looking for a PIK chip for a 97 you would need printer font files for each model, I discovered that when I installed the PIK from a 91 into a 97, it works but does not print the right symbols when listing a program.

Paul.

[PANAMATIK]

(03-01-2021 06:50 PM)Paul Berger (Canada) Wrote:  I have been looking for a PIK chip for a 97 you would need printer font files for each model, I discovered that when I installed the PIK from a 91 into a 97, it works but does not print the right symbols when listing a program.

Paul.

That is a good point. I didn't know they are different. But not unsolvable.

I know how the ACT processor sends a print line to the PIK, there are two print instructions, one only for numbers, the other for alpha characters, and with an appropriate program it would be possible to print any possible character on paper and re-engineer the 5x7 matrix font. Then do this with an HP-19C and HP-97 original PIK chip. But perhaps there is a documentation of these fonts in an early patent.

I think you would not be the only person, who needs an HP-97 PIK replacement chip. I have a defective HP-97, which cannot read the keyboard, because of a defective PIK.

Bernhard

[Ren]

When I read the part where you soldered two jumper wires to flex cable, I thought, I need to ask him to share his "Secret of Success" for doing it.
Then, I read a little farther...
B^( <---- (that is a sad, not mad, face)

Good Luck in any future work!
(I haven't placed the new ACT in the HP-22 yet. The wife wants my help in remodeling the kitchen!)

[Don Williams]

Well I wish I had not read this post because it brought back a lot of bad memories and now I need to confess as part of an attempt to heal.

I once had 2 HP 19 C’s. One pristine fully working example and one which gave random and unrepeatable symptoms. I finally decided to fix the the situation.

I started on the faulty unit and quickly discovered this is the calculator from hell. It is just exactly as Panamatic described. I would like to know how anyone could run an efficient production line with the way that calculator is assembled.

After many attempts and miserable failures at repair I deduced that that the randomness of the problem had to be a contact problem so I broke out me 28 gauge wire and started doing point to point soldereing. (I always do this with 41 series and Spice series with those pressed in IC’s. I stopped doing the DEOXIT dance long ago)

So the point to point wiring is finished and I decide to open the good calculator to do a final trace check before I power anything up. Both calculators are side by side connected to a bench supply and everything looks good. Success. Its late at night and I really don’t want to tackle the reassembly so I set it aside for a few days.

So before I finally get back to this project I realize that I did not check the printer functions. So just before I close them up I connect the power supply to check the printers.

Are you ready for it?

I connected the bench power supply going to both calculators BACKWARDS. Smoked them both. The only good thing out of it was at least I did not have try to reassemble them again. That was about a dozen years ago. They still sit in their final resting state. I taped everything down in that final condition and sort of created a small shrine to my stupidity. May they rest in peace. And may I never see a 19C again. Miserable design.

[teenix]

An interesting read Bernhard, thanks for sharing the experience, a great deal of knowledge learned.

I'm sure they would have had some sort of "magic fingers" tool to assemble these in the factory, but I wonder how the technicians of the day coped with repairs. Maybe most of them went prematurely bald.

This is a little excerpt from the Nov 76 HP Journal...

................
Because the [ribbon] cable connectors are copper and the
printhead interconnects are thin-film aluminum,
conventional solder bonding was not possible. Ultra
sonic-wire or stitch bonding provided the answer.
................

It may have been possible (but finnicky in the extreme) to solder to the stitch wires, but maybe they too are made out of material which will not take to solder.

I have been looking into creating a PIK chip replacement. While I have been developing the HP-97 replacement CPU board I have dug into a few details of its workings.

I already have the key scanning and 7 key buffer storage, plus ROM tables and code which assembles the print data generated by the HP microcode and then outputs it to the print heads with the required pixel timing to produce the original slanted 5 x 7 print format. That mouthful being said, it would still take some time to produce as a chip replacement. I myself would need more analysis of the instruction and data bus activity and one chip should do for all 9x models.

The replacement would not be able to print anything other than the standard characters for a particular model because those codes are still produced by the 9x operating microcode, although an updated ACT would add extras I'm sure :-)

I just had my 62nd birthday as well. I haven't played with a 19C so I still have my hair - well most of it, but coding for hardware interfaces has changed the color a bit ;-)

cheers

Tony

[teenix]

(03-01-2021 10:26 PM)Don Williams Wrote:  So before I finally get back to this project I realize that I did not check the printer functions. So just before I close them up I connect the power supply to check the printers.

Are you ready for it?

I connected the bench power supply going to both calculators BACKWARDS. Smoked them both. The only good thing out of it was at least I did not have try to reassemble them again. That was about a dozen years ago. They still sit in their final resting state. I taped everything down in that final condition and sort of created a small shrine to my stupidity. May they rest in peace. And may I never see a 19C again. Miserable design.

I feel the pain, although I didn't destroy such a valuable item - touch wood.

A reversed biased diode in parallel with the power supply via a fuse might have saved the day.

Did it actually destroy itself. The power supply has diodes stopping the reversed voltage appearing in some circuit areas, and through the power supply coil.

The RAM attached to the battery has diode protection as well although not meant for the purpose.

Some items like tantallum caps in the power supply may have fried/shorted and stopped it working.

You never know your luck in a big city.

cheers

Tony

[Don Williams]

I believe your observations are correct. There is polarity protection in the PSU. However I made a rather brilliant move during this procedure. I kept flipping the boards over and over again trying to trace the connections. I finally got tired of reattaching the power supply clips which leaped off the battery connections every time. My solution was to take some small jumpers cables that would stay firmly attached to the power and ground pins on the IC's. So I dialed up +Vcc and -Vcc and avoided those pesky protection diodes. I did all the trouble shooting in that mode and every thing went well until I made that last fatal test. PMOS and CMOS chips can really smoke when you reverse bias the power connections.

[Thomas Okken]

(03-01-2021 10:26 PM)Don Williams Wrote:  I connected the bench power supply going to both calculators BACKWARDS. Smoked them both. The only good thing out of it was at least I did not have try to reassemble them again. That was about a dozen years ago. They still sit in their final resting state. I taped everything down in that final condition and sort of created a small shrine to my stupidity. May they rest in peace. And may I never see a 19C again. Miserable design.

I sympathize, but hearing about dead 19Cs does make me a little sad as well. I used to own one back in the day, brand new, and after the HP-25, using the 19C was a real thrill. 98 steps, 30 registers (more than the 67!!!), indirect addressing, and a printer, all in a machine that you could just toss in a backpack. (Calling it a pocket calculator was a bit of a stretch.) Fond memories. I would like to have one again sometime, but I should probably just spend the extra $ and get one that is fully restored and guaranteed working.

[[kby]]

Very interesting read, Bernhard, and brings back many unfinished memories.

I caution the ambitious that the 19c case assembly is relatively simple compared to the HP-10 as it is impossible rather than just really difficult to get all of the gold fingers connected and close the case. I have posted elsewhere on this site my final heuristic technique.

I have run across the printer cable issue you mention. I have repaired some breaks at the fold with circuit board repair traces and solder, but it’s really difficult if you have adjacent traces to do. I have a few printheads that just need that cable repaired. I had not successfully removed the rubber cover. I had looked at the -97 solution but gave up when I noticed the cables don’t match. I haven’t had the time tio try again to repair the cables. It is annoying that the contacts are on the side of the cable AFTER the sharp fold, which is where the most damage occurs!-kby

[Paul Berger (Canada)]

(03-01-2021 07:52 PM)PANAMATIK Wrote:  

(03-01-2021 06:50 PM)Paul Berger (Canada) Wrote:  I have been looking for a PIK chip for a 97 you would need printer font files for each model, I discovered that when I installed the PIK from a 91 into a 97, it works but does not print the right symbols when listing a program.

Paul.

That is a good point. I didn't know they are different. But not unsolvable.

I know how the ACT processor sends a print line to the PIK, there are two print instructions, one only for numbers, the other for alpha characters, and with an appropriate program it would be possible to print any possible character on paper and re-engineer the 5x7 matrix font. Then do this with an HP-19C and HP-97 original PIK chip. But perhaps there is a documentation of these fonts in an early patent.

I think you would not be the only person, who needs an HP-97 PIK replacement chip. I have a defective HP-97, which cannot read the keyboard, because of a defective PIK.

Bernhard

Well I found my original PIK after a long time of looking for it intermittently, I was afraid I had pitched it out when I saw that the PIK out of the 91 seemed to work, that is until I went to list a program. The keyboard part of the original PIK works fine, but it needs to warm up for 1/2 - 1 hour before the printer works correctly, but it is a lot better than nothing.

Paul.

[teenix]

(03-02-2021 02:55 AM)Don Williams Wrote:  I believe your observations are correct. There is polarity protection in the PSU. However I made a rather brilliant move during this procedure. I kept flipping the boards over and over again trying to trace the connections. I finally got tired of reattaching the power supply clips which leaped off the battery connections every time. My solution was to take some small jumpers cables that would stay firmly attached to the power and ground pins on the IC's. So I dialed up +Vcc and -Vcc and avoided those pesky protection diodes. I did all the trouble shooting in that mode and every thing went well until I made that last fatal test. PMOS and CMOS chips can really smoke when you reverse bias the power connections.

Quoting Australian vernacular..

Bugger :-(

cheers

Tony

[PANAMATIK]

(03-01-2021 10:26 PM)Don Williams Wrote:  I once had 2 HP 19 C’s. ......

What a sad story! But perhaps not yet finished. What I found about the HP-19C printer PCB is that all components can be replaced by today available parts. This said and if I could manage to build a PIK chip replacement, also the processor PCB could be restored. The new ACT chip can drive the display and replaces all RAM and ROMs

I will ask teenix for sharing his PIK knowledge and add some of mine and perhaps at the end of the year there could be .... your story could become an happy end somehow?
No promise, but an attempt should be made.

Bernhard

[Don Williams]

(03-02-2021 07:33 AM)PANAMATIK Wrote:  

(03-01-2021 10:26 PM)Don Williams Wrote:  I once had 2 HP 19 C’s. ......

your story could become an happy end somehow?
No promise, but an attempt should be made.

When that board is ready, with the so easy to install that a half blind 74 year old can
do it instructions I am ready.

[PANAMATIK]

(03-02-2021 03:23 PM)Don Williams Wrote:  

(03-02-2021 07:33 AM)PANAMATIK Wrote:  your story could become an happy end somehow?
No promise, but an attempt should be made.

When that board is ready, with the so easy to install that a half blind 74 year old can
do it instructions I am ready.

PM sent.

Bernhard

[teenix]

Hi all,

I looked into a PIK chip replacement circuit just at an ad-hoc level and just parts including PCB should be around the $15 mark give or take. It is a bit of a hassle because most of the PIK signals are 0-6V and require level shifting to suit a modern processor.

The main thing is getting it going. One problem is the time involved and how this relates to demand which doesn't seem great from what I've seen so justification becomes an issue too.

The other problem is that this is a controller for hardware which is pretty much irreplaceable. There is a real chance of causing damage during development. Burnt print heads, motor drivers, stripped gears or a damaged printer mechanism are examples if the software crashed for some unforeseeable reason at a critical point. This is much more relevant than interfacing to the card reader or display.

Unfortunately, all of the printer hardware is totally reliant on software for correct operation, (assuming the driver circuits are ok) so it is quite a risk to take, a problem which I am still grappling with for the 97 CPU board replacement.

cheers

Tony