HP-19C Printer

[teenix]

Hi all,

I'm trying to get a non working 19C going again for someone.

It had some battery corrosion mostly around the pins connecting the printer connections between the two circuit boards. I cleaned this up and it started working again but then stopped again. I had to clean out the plated through holes properly to get a better connection and it seems to be happy now.

I only get "Error" if I press [g] [PRx] to try and use the paper feed, either with paper in or out, with the print head in the home position. The Home switch is working right up to the PICK pin.

Does the 19C actually detect paper out? I cannot see any switches for this on the mechanism.

I am not familiar with the printer mechanism, but I would expect the print head to be attached to the slots in the worm gear, however the print head just slides back and forth. The motor works when connected to a bench supply, but the print head stays put.

Any ideas for the clueless ?

cheers

Tony

[Jonas Sandstedt]

Hi Tony,

First of all, the HP-19C does not have any Out Of Paper Switch (OOPS). The HP-19C printer is similar to the HP-97 printer, but is not adapted to return to the home position when the characters have been printed, always travelling all the way over the paper width by means of the double-threaded (cross-cut) worm gear. This means that the HP-97 service manual partly is applicable for the HP-19C printer, but the HP-19C printer has less components

I do not I have understood all the problems correctly, but as I understand the situation the motor does not turn on at all in the calculator and you get an Error message, and when externally energized, the worm gear rotates but the printhead does not move.

If this is correct, I think that any, or both, of the motor drive transistors may be broken. You can get tips from Bernhard’s homepage panamatik.de. Go to “Download” and download “HP taschenrechnersammlung 2020” and read from page 96 and forward (it is in English).
If the worm gear rotates but the printhead does not move, there is a small brass bit in the printhead that guides the printhead on the cross-cut worm gear, but it must be pressed towards the worm gear by a tiny spring that can be missing or broken. You find these components when carefully at least partly removing the print-head metal bracket that covers the part of the printhead facing a user. The spring and the underlying brass bit are held in place by the top part of the metal bracket.

I recently bought a HP-19C with the symptoms of the worm gear rotating and the printhead not moving. To my astonishment the spring was no there, it had been removed by a previous owner and probably mislaid. I replaced it with a small spring that I took from a small cylinder lock and now it works perfectly.

I hope the above helps, please come back if you have further queries that I at least can try to answer, or if I have misunderstood something.

Best regards
Jonas

[teenix]

Thanks for the reply Jonas, your assumptions were correct.

It is interesting that the 19C code has a paper out test instruction although this appears to correspond to the Power OK flag.

Maybe this is a clue to look for as this flag produces the same error when it is intentionally held in a faulty state.

I also just noticed that the Paper Out instruction is not a PICK instruction as I thought, it is internal to the ACT.

I thought there might of been a broken print head drive pin, but maybe the spring you mention (and/or maybe the pin) went AWOL if it was repaired before.

cheers

Tony

[teenix]

The printer seems to have been pulled apart in the past as the spring that holds the pin into the worm shaft has gone AWOL. Not much can be done except find a replacement.

Edit: Fixed.

I found an old miniature push button switch that was used for resetting my old mainly dead 486 PC. The switch came apart easily and the spring inside was the right diameter but too long. I cut it to size, installed it, and now the print head mechanics work :-)

Now to try to figure out why the printer doesn't work.

Edit2:

The LLD pin on the Cathode Driver is normally HI when it detects a good battery voltage. If the battery voltage goes below a certain threshold, this pin goes LO and usually lights a LED in the display. In the 19C this pin controls a flag that is tested in the Microcode and a low battery starts the decimal point led flashing.

When trying to print, the HP-19C microcode tests for this Logic LO condition for the duration of a timeout. If it is still LO after the timeout then an Error is displayed and printing aborts.

Also, before printing starts, the print head home switch is tested and if it is not home, the microcode will stay in a permanent loop until it is. If the printer is faulty and doesn't home, then the user will just see a blank display until the power is switched off then back on.

The 19C here seems to exhibit symptoms like the first scenario, as every time a print is started "Error" is displayed, however, the oscilloscope never sees a LO level appear on the LLD pin. From the code point of view, I cannot see any other reason for an error to be displayed. Also, the LLD flag input is independent from the other multiplexed flag circuitry for the other ACT flag input.

The motor drive signal from the PICK chip will never occur under these conditions, and indeed it doesn't.

The only thing I can think of is that the ACT LLD flag input is faulty.

It could be possible that the power supply capacitors have begun to fail so maybe changing them might offer a fix.

cheers

Tony

[Jonas Sandstedt]

Very interesting Tony,

I do not think I can offer any more help other than checking for voltages on the connector pins between the boards that I am sure you have already done. If you could have access to a working HP-19C, you could of course test the upper part of the calculator (logic board, keyboard, display) in a known good PSU/printer board to quickly find out in which part the malfunctions lies. That is of course not easy since it is a relatively rare calculator.

best regards
Jonas

[jjohnson873]

Hi Tony....I was going to suggest replacing the motor capacitor, but you're already onto that potential problem.
As stated earlier, there is no paper out detection on the HP-19c.
It's odd that an "ERROR" is displayed when trying to print. I can't say that I've seen that in any of my HP-19c repairs, unless the battery voltage is low. Are you running on a bench power supply or batteries?
At least you've gotten the mechanical portion repaired with the replacement of the spring in the lateral helix printhead mechanism. Once the motor starts, it will drive the print head back and forth, searching for the home position.
I've also noticed that the print head ribbon cable can get jammed if it's not routed correctly. It needs to move freely, or it can catch on the pc board.
Interesting that the motor just doesn't start at all. I'm baffled at what is causing the "ERROR" in the display. Is there excessive current drawn when you execute a print command?
There is a small drive train inside of the motor housing. I've noticed some motors can be a bit dodgy if that drive train is dirty. But I wouldn't put that issue at the top of your troubleshooting list.
Let us know what you find in your testing. ~ Jim J. ~

[rprosperi]

(03-14-2022 12:43 AM)jjohnson873 Wrote:  I'm baffled at what is causing the "ERROR" in the display.

Page 192 (App B) of the 19C/29C Owner's Handbook lists all the conditions that invoke the ERROR message, including the comment "The calculator will display ERROR when the calculator is first turned on if power to Continuous Memory has been interrupted", so maybe this is actually normal and correct under the circumstances?

[Jonas Sandstedt]

Hello Tony,

I have thought a little about his, and as I understand the problem, the calculator works normally until a print command is given – is that correct? If that is the case, my experience of these calculators is that if the printhead does not move when a print command is given, which is detected by the print head home switch, an Error message is provided. For one HP-19C with these symptoms this was due to a simple thing - one motor connector lead was broken. When the connection was fixed, the motor was running again and no Error message appeared.

What happens if you manually move the printhead away from the home position and then turn on the calculator?

The HP-97 repair manual states:

“2-29. When the PIK gives the command to move forward, transistors Q11, Q14,and Q9 turn on. Transistors Q12, Q13, and Q10 turn on to reverse the direction of head travel.”
Note: The three latter transistors are not used here since the DC motor only travels in one direction in the HP-19C.

“2-31. The printed line width and character-to-character spacing is determined by the speed of the dc motor. To control the speed of the motor, U4 samples the output voltage generated by the motor when the driver transistors are turned off and the motor is coasting.“

Have you checked if the PIK command is sent, and is so, where does the DC supply to the motor stop in that case? If there is an erroneous or non-existent PIK command, this could perhaps be due to faulty speed control in U4?

Hope some of this could be of help, perhaps you have already checked it all.

Best regards
Jonas

[teenix]

Thanks for the replies everyone.

Anyone know of a replacement PNP transistor for the motor drive. It appears to be open circuit. I imagine it will be a fairly sturdy device having to deal with PWM directly to the motor.

It is marked as 7123-393, so I guess HP part 1853-0393 which is not listed in the Xref sheets.

cheers

Tony

[Jonas Sandstedt]

Hello Tony,

I enclose a picture from Panamatik (Hope it is OK Bernhard) where some equivalents are marked. Hope it helps.    

Best regards
Jonas

[teenix]

Thanks Jonas,

I'll give it a try tomorrow.

cheers

Tony

[teenix]

I replaced the drive transistor with a BC327 and the motor worked and the carriage moved back and forward if I isolated the FWD line and applied Vss through a 1K resistor. This proves the drive circuitry works and the motor works.

However, if the motor is not homed at switch on, the motor does not home. If I manually home it, switch on and try to print I get a blank screen and nothing else which is what I saw should happen in the microcode. The home signal is appearing at the PICK chip and the hybrid package outputs the 1 volt feeding op amp U3a that senses the motor speed.

This sort of leans towards no motor drive signal coming from the PICK chip. I cannot see anything appearing on the scope either.

cheers

Tony

[Jonas Sandstedt]

Sounds tricky, Tony.

The HP-97 service manual states:
"Check pin 17 of the board for the FWD signal. If not present, replace U5 and U4 in
order until the 1.5 millisecond signal appears. Then go to step 10."

An illustration of the FWD waveform is shown in Figure 4-15, section 4-27.

In Appendix D, Service notes in the service manual, it is stated:

10. Doesn't print - Problem part: PICK.

So, it seems as if your suspicions are correct. I do not know if it helps, but in the HP Journal linked below, the HP-97 printer is described.

https://www.hpl.hp.com/hpjournal/pdfs/Is...976-11.pdf

Best of luck!

Jonas

[teenix]

Hi Jonas,

Yes, I have a copy of that journal, I've read through it a couple of times.

Seeing as how the calculator is functional, and now the printer electrics/mechanics are working, I'm pretty sure the PIK is faulty. I guess a ROM could be also, but that might be a stretch.

Oh well, I tried, thanks :-)

cheers

Tony

[Jonas Sandstedt]

Sad to hear, it would have been swell if the new transistor had done the trick.
It seems as if this particular calculator will have to wait for a donor, or even be a donor.

best regards
Jonas

[AndiGer]

Tony,

If it really the PIK you could replace with the PIK from a 97.
According to Eric Smith both 19C and 97 share the same PIK.

And as most of us wonder what's the meaning of PIK - Eric's page gives the solution:
Printer Interface/Keyboard

HTH
Andi

[teenix]

(03-15-2022 04:05 PM)AndiGer Wrote:  Tony,

If it really the PIK you could replace with the PIK from a 97.
According to Eric Smith both 19C and 97 share the same PIK.

And as most of us wonder what's the meaning of PIK - Eric's page gives the solution:
Printer Interface/Keyboard

HTH
Andi

...or Printer Interface Control and Keyboard buffer circuit. PICK
Ref: HP Journal Nov 1976 Pg 11

Tomaytoes - tomartoes :-)

cheers

Tony

[teenix]

Someone has definitely been into this one before. There are some new solder joints on the circuit boards and the rubber ring mounts for the printer are missing so it flops around ever though the 3 brass screws are reasonably secure.

I also notice while it was sitting on the bench and turned on that it would be working fine then on an odd occasion it would go bonkers and I have to turn it off to reset. Maybe if the PICK chip is faulty it may be causing a glitch or ... ???.

After reading about the dramas putting the 19C together again, it went back together ok. There are a couple of rectangular holes in the battery compartment that allow access under the power supply board. I made up some thick paper spacers (one L shaped) and slid these under the board to give it some support while the connecting fingers were mated to the holes as the case halves came together.

It wouldn't be too hard to make a CPU board replacement for this model - maybe one day :-)

cheers

Tony

[tangdfx]

Amazing news.

I am waiting for the HP 19C new CPU board.

It is a good candidate for the new CPU board just like new HP 97 CPU board.

I have 4 HP19Cs.

Thank you Tony.

[teenix]

(03-16-2022 06:50 AM)tangdfx Wrote:  Amazing news.

I am waiting for the HP 19C new CPU board.

It is a good candidate for the new CPU board just like new HP 97 CPU board.

I have 4 HP19Cs.

Thank you Tony.

Same problem as the other CPU boards, I need my own calculator to develop and maintain the boards. As mentioned - one day :-)

cheers

Tony