HP97 The journey begins

[teenix]

Hi all,

I am getting through the help file. It is taking a bit longer than I expected.

I had to remodel a few things due to a fitment issue but the latest looks ok. I have it all assembled and am about to clam shell it back together. This is the squeamish part because you never know if everything is ok and if not, the shoulders and face tend to droop somewhat.

Anyway, not today, I am feeling a bit burnt out, so Sundee arvo off.

cheers

Tony

[teenix]

Hi all,

If anyone is interested in the HP-19C CPU board, I have posted the initial help PDF file here. Installation is a bit involved due to the way the HP-19C was manufactured, but as mentioned in the file I will help out as much as possible.

My writing skills aren't the best so comments welcome.

Still have some final hardware testing to do.

cheers

Tony

[Jonas Sandstedt]

Hello Tony,

That was a thorough description, well done! I understand that it has taken some time to accomplish.

It might have been mentioned somewhere previously but not finding it easily and being lazy I have some questions about the performance about of the “new” hp-19c.

How many memory registers and program steps are available - same as the original?
Are there any clock functions?
Are there any disadvantages running the calculator in “Turbo mode” except perhaps a higher current consumption?
Is it safe to charge the battery pack with the original charger?

All these questions are out of curiosity only, I think you have made an awesome work and produced these PCBs that can re-vive since long dead calculators. I will order at least two HP-19c kits when they are ready for shipping.

Best regards
Jonas

[teenix]

Hi Jonas,

Thankyou.

The memory is the same as the original, but I was working on an expansion as mentioned a few posts ago. I wasn't sure of how it would work in practice, so I shelved it for the time being. It is only code, so it can be re-flashed at a later date. Re-flashing can be done with Bluetooth on this model.

I didn't add a clock. I figured that the 19C might be more of a prized item and wouldn't be used as an "every dayer". This was also some of the reasoning for re-thinking extra memory.

Turbo mode won't increase power demands, the processor is "told" to execute the HP instructions as fast as it can, rather than wait around for the 280uS normal instruction time.

The new board has the original charger setup. I made mention in the help file about running it from 4 alkaline batteries, but I would discourage it as the motor and print head and their drivers are not designed for that voltage.

cheers

Tony

[Jonas Sandstedt]

Thanks for the quick reply Tony!
I just wonder if using the original charger might damage the ICs as in the original Woodstock design, i.e., should my re-chargeable batteries be charged outside the calculator?
Best regards Jonas

[teenix]

(12-07-2023 02:42 PM)Jonas Sandstedt Wrote:  Thanks for the quick reply Tony!
I just wonder if using the original charger might damage the ICs as in the original Woodstock design, i.e., should my re-chargeable batteries be charged outside the calculator?
Best regards Jonas

All the processor circuitry is isolated from the battery directly. The input to the CPU board should be ok up to about 16V. (Capacitor rating)

It may be a problem if the battery disconnects while on charge. There is up to 25 ohms resistance in series with the charger output, but I don't know the no load output voltage of the charger.

A 12V zener diode between the output of the high charge resistor and GND could be used to shunt excess voltage. It could be patched into the circuit board.

The safest bet is always external charging, but the decision is up to the owner.

cheers

Tony

[tangdfx]

Hi，Tony

I just give a samll suggestion that could it possible to add a HP97 mode to simulate as a mini HP97.

[teenix]

(12-08-2023 04:47 AM)tangdfx Wrote:  I just give a samll suggestion that could it possible to add a HP97 mode to simulate as a mini HP97.

It could be done of course but it will require a lot of work and I'm not sure if there is enough ROM space available. I'm not sure if the smaller 19C printer width will handle the extra line characters for the 97.

cheers

Tony

[teenix]

Hi all,

Still working on the 19C. My son has asked me to help him out nearly every day lately, so I haven't had much spare time.

I've implemented these new program steps. None of the HP microcode has been altered, it still does its thing, but my code monitors it while running and changes data here and there. The key sequences (or program steps as text) are listed on the display and the printer prints the new steps. [f][f] is used as a prefix because there are no spare key functions on the 19C.

[f][f][0] RAND
[f][f][1] CONS
[f][f][2] BEEP
[f][f][3] DATS
[f][f][4] FILE

RAND
generates a random number >=0 and < 1

CONS
allows constant storage and recall - 10 constants max

BEEP
sounds the beeper once

DATS
swaps the Primary and Secondary memory
14 registers total, Primary 0 (I register) and Primary 1 stay as is.

FILE allows 3 functions to access any of the 400 storage locations
load a program file from memory, then start from the program beginning or from LBL 0 - 9
load a data file into Primary memory
save Primary memory to a data file
this should provide a lot more steps and storage memory to play with.

There could be up to 8 errors generated during a FILE command, like trying to load an empty file etc. If one occurs, the HP microcode does its normal error thing and the error number is displayed as well, similar to Spice. Eg "Error 7"

cheers

Tony

[teenix]

Hi all,

With the 19C having the ability to save and transfer data files during a running program, I changed the memory chip to one that will serve this purpose a lot better. With this memory you can write to any file up to 10^13 times and save them for up to 10 years without power.

This way, if programs have to transfer data many times to reach an end goal, there should be no problem saving and recalling it.

cheers

Tony

[teenix]

Hi all,

Work has been slow with the 19C. I have been spending time assembling CPU boards and getting the Topcat ribbons out.

I did manage to install the new memory chip and rewrite the interface code, so the available read/write cycles should last the life of the board. I also added a few software tweaks and apart from final printer testing it should be ready. Testing the printer is fiddly because the calculator needs to be fully assembled. If there is a problem, then it may need to be opened up again, which I don't want to do if I can help it.

cheers

Tony

[teenix]

I needed a bit of a change of pace today and decided to add a kooky clock display to the Spice CPU board.

It displays the time as...

5 o'CLOC, MIDNIGHT, NOON, 5 PASt 7, 25 to 12, HF PAST 9, qtr TO 3, etc, with a PM indicator on the last digit.

The things you do :-)

cheers

Tony

[MinkLib]

A Kooky Spice Clock - it doesn't get better than that :-)
Now if only we can beep in our programs, like the HP19C.

I know, I know - spoilt children..

Happy New Year.

[teenix]

(12-31-2023 10:53 AM)MinkLib Wrote:  Now if only we can beep in our programs, like the HP19C.

Happy New Year.

The 34 program codes are all used. but the 33 and 38 have a few spare. Adding a new code is a bit involved but I might be able to do something in the new year.

cheers

Tony

[brouhaha]

(12-31-2023 11:20 AM)teenix Wrote:  The 34 program codes are all used.

The original 34C had one spare code, 0x2c. Do you mean that you have already used that?

[Paul Dale]

Eric's list lists 0x26 as spare: http://www.brouhaha.com/~eric/hpcalc/hp3...table.html

[teenix]

Long day, apologies, yes 0x26 is spare.

I checked again and I used it to create a random number.

cheers

Tony

[teenix]

(12-31-2023 10:53 AM)MinkLib Wrote:  A Kooky Spice Clock - it doesn't get better than that :-)
Now if only we can beep in our programs, like the HP19C

Done - for the 34C :-)

With only one program code available for use, I had to double up on how code 26 functions. It is accessed by pressing [g][0].

If the first digit in the X register is 1, the beeper will sound once, else a random number will be placed in X.

Example...

X = 999.89
Press [g][0]
Random Number in X

X = -1.234
Press [g][0]
Beep

[1]
[g][0]
Beep

The displayed code for this is 15 0, or in text mode, gFuncN. (N) because other functions may be added at some point.

I haven't uploaded yet, but will tomorrow.

cheers

Tony

[MinkLib]

Fantastic New Year Day's present!
You are the Spice Wizard, enjoy a cool one when there's a chance.

MinkLib.

[teenix]

Hi all,

I have the 19C fully assembled. After a bit of practice, if the board to board connector pins line up ok, then the cases come together quite easily. I had to add some code to format the new memory IC because it may have random data in it when factory fresh.

The attached image shows the printer output, which looks reasonable. It seems as though the paper is bending a little bit when the print head is near the middle of travel.

Even though the motor driver is supposed to compensate, just like the 97, the print carriage seems quite susceptible to changes in drag which alters the print line length a bit. The only way I could get the printout aligned like shown was to gently hold the paper down just above the print head. If the paper sticks up and rubs against the top of the print head, the line length changes a bit. Also, I need to gently apply a tiny bit of pressure to pull on the paper to get even line spacing, so I suspect the paper advance needs servicing.

I also noticed that if the rubber mounts for the print motor are not even, the motor sits a little bit out of alignment.

I can change the print speed and intensity values from the keyboard and the printer responds accordingly, darker, lighter, faster slower.

Unless I go back to square one, I don't think there is much more I can do with this mechanical marvel. Everything else seems to work ok. I tried writing a program that when it runs, loads new program and data files and it all seems to work.

I'm sure there will be bugs coming, but what a trip :-)

cheers

Tony