HP Forums
HP97 The journey begins - Printable Version

+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: General Forum (/forum-4.html)
+--- Thread: HP97 The journey begins (/thread-12156.html)

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

RE: HP97 The journey begins - Dreato - 09-20-2021 03:47 PM

Looks like great progress! I want at least 2 when they're ready for prime time!

Keep on truckin'

Best Regards,


RE: HP97 The journey begins - teenix - 09-21-2021 03:45 AM


The journey might be ended.

While testing the printer, I noticed some variation in the printing width and while trying to figure out why, I realized that I miss read the data sheet for the motor driver IC that I chose. It is for DC motor control in battery powered equipment and I thought it compensated for motor torque variations but it doesn't - only for battery voltage changes to try and maintain a constant rpm.

Unfortunately, to add this extra functionality means more testing and then hopefully a working PCB re-design, and then software changes.

Because it is hard to prototype with surface mount bits and pieces, I have already spent quite a lot on development and even though I never wanted a profit out of the project, I am not sure I can continue to throw any more at it. I will have to sit down and have a think. :-(



RE: HP97 The journey begins - Dreato - 09-22-2021 03:37 PM


On behalf of many folks worldwide, I'd like to thank you for all your work to date, and encourage you to continue. We can send $$ to you if needed. However, we can't send time, or help with surface mount components.

I have found your HP67 motherboards to be a godsend, with much increased speed and storage capacity, and the handy bluetooth connection. In addition, I anticipated taking the 'blocks' of 43 program from the HP 67 to the HP97.

So please take some time for "havin' a think", and remember we're pulling for you!

Best Regards,

RE: HP97 The journey begins - Paul Dale - 09-23-2021 03:09 AM

Pass a hat around and I expect you'll get enough donations to continue.
There seems to be quite an interest for the 97 replacement.


RE: HP97 The journey begins - teenix - 09-24-2021 12:51 PM

Hi all,

I was talking to my plumber son the other day and he gave me a bit of work as a gopher, go for this go for that type of thing, so I was able to make a few $ that I otherwise wouldn't have had. This should enable a bit more experimenting.

I did do some formal training years ago, but I am not an electronics engineer by any standard especially with analog. I was having some thoughts on how I could salvage what I have so far without reinventing the wheel and have to deal with even more problems.

The DC brush motor driver that I chose is a DRV8832 IC which among other things has printers listed in its battery powered applications and the drive capability and supply voltages are good. Great I thought. One of its main tasks is to monitor battery voltage and adjust the PWM motor drive signal and so keep the motor spinning at a constant speed.

It also monitors the voltage across the motor terminals and integrates that to get an error offset, so I thought that would compensate for torque as well. Unfortunately, as mentioned, it does not seem to do this.

Ideally, the printer head mechanism should have relatively constant resistance as it moves from right to left while printing. HP obviously discovered otherwise and hence had to incorporate speed control based on the voltage produced by a free wheeling dc motor and some wizardry inside the PIK chip. This setup is there to account for battery voltage and torque changes.

As I understand it, a change in load on the motor changes the speed. This will increase or decrease the motor current, leading to pretty much a short circuit if the armature stops.

I was thinking of putting a low value resistor, somewhere around 0.2 ohm as a start, in series with the ICs supply pin. The chip is designed to cope with with varying battery voltages anyway, so the idea is that as the motor current varies under load, there will be slight changes in the sensed battery voltage. This will cause the chip to vary the PWM duty cycle to try and compensate. Also if the battery voltage changes, then that is catered for as well.

Obviously there will be some fiddling so that the chip doesn't under/over compensate, but it might work as a good enough solution, and if so only some time and a few cents. A good quality current sense resistor should suffice.

I'm open to ideas :-)



RE: HP97 The journey begins - mfleming - 09-24-2021 01:30 PM

This seems like a case where direct contact with the manufacturer (T.I.) or a post to one of their support forums may be appropriate. T.I. also has a specialized version of Spice that includes a library of their parts and example circuits. I'll take a look at that later today to see if there is anything helpful.

I had a choice between analog and digital upon graduation, and ran like hell away from analog Smile

RE: HP97 The journey begins - teenix - 09-26-2021 02:14 AM

Hi all,

The modification did not make any difference to the printer drive.

Attached is an image showing the different print line widths. The paper is crooked, not the lines. The line spacing is uneven because the old rollers are not gripping the paper properly.

You can also see the effect of the software controlled print intensity adjust.

It would seem that I will have to recreate the original motor driver setup, although it would probably be better and more compact with a newer H-Bridge device. The trouble at the moment is, that the semi-conductor shortage is having large delays in delivery for some of these items.

The motor chip that I am using has direct PWM input so I can try driving it directly from the PIC and then try and add the closed loop control as per the original circuit. Somewhat of a chore with tiny parts.



RE: HP97 The journey begins - AndiGer - 09-26-2021 06:26 AM

Hi Tony,
Studying the printed lines it seems to me that the more intensive lines 1 and 4 are of approx. same length whereas the lighter lines are longer.
The first 5 digits of line 4 look less "italic".
May it be related to print intensity?
And maybe old blue paper?
Wish you the best

RE: HP97 The journey begins - teenix - 09-26-2021 07:29 AM

Hi Andy

Thanks for the assist, very appreciative.

The CPU board is operating in "remote mode" away from the calculator and the paper is just sitting in the bare printer so it is a bit loose without the clear plastic screen. I think the paper is moving about a little bit while the head moves. The oscilloscope pixel output looks even and clean.

Well, I got to thinking again :-)

I didn't think much of what actually popped into my head, but on the off chance, I thought why not get out the 97 service PDF and find out the oil points for the printer mechanism. I applied a small amount of oil that you use for hair clippers, nice and light, and allowed it to work its way in. I then cleaned up any residue and tried the printer again.

Immediately, the [ 0.00 *** ] print line doubled in length. I printed about 15 lines and all were exact same length. I reduced the print line length timing in software by 50%, and printed [1.23456789.0] about 10 times and every line was the same length.

I'm not jumping for joy yet, but it would seem that the mechanism needed a bit of oily love. If this is indeed the case, then I will reduce the PWM hardware reference for the motor FWD speed and try it out again.

It would seem that this type of simple printer maintenance is well worth while.

It is a late cold Sunday afternoon in what is apparently the worlds longest lockdown here, so maybe tomorrow. I think a wee dabble in some brown frothy suds might be in order for now.



RE: HP97 The journey begins - AndiGer - 09-26-2021 08:27 AM

Hi Tony,

Well done! I thought you had cared about the printer mechanism before. Obviously not till today :-).
I once had a HP 82143A printer that had a dense print output. I put some oil to the spindle - et voilĂ  - it worked fine. A pure mechanic problem.

Have a nice Sunday evening - and hopefully tomorrow you can report final success.

RE: HP97 The journey begins - teenix - 10-01-2021 06:35 AM

Hi all,

Since oiling the printer it seems to be settling down. I received an email with another original 97 CPU and printer experiencing the same changing line lengths. Some oiling there fixed the problem too, so I'm guessing that there is not a lot of torque control for the original setup even though it has speed monitoring. Once the PWM maxes out there is not a lot more it can do.

Attached is a new printout. The first 2 lines show a user programmable change in intensity and the last line has the user programmable speed decreased.

I have gone over the software again and have remodeled the printer ROM data, plus a few software changes and have enabled some extra PIC hardware. Now I'm pretty much 100% sure the whole setup can be run with a single processor.

Depending on what is happening the processor may have to do these functions all at the same time...

Execute HP code
Update the LED display
Read the keyboard and switches
Control the keyboard buffer
Service the card reader
Service the printer, including shutting it down asap in case of code or motor failure
Monitor for external commands
Control the 97S functions (still in progress)

Quite a bit for it to do.



RE: HP97 The journey begins - teenix - 10-01-2021 11:26 AM

Hi all,

Three PIC processors running here.

The joys of prototyping :-)



RE: HP97 The journey begins - Paul Dale - 10-01-2021 12:01 PM

Three seems excessive Smile


RE: HP97 The journey begins - teenix - 10-01-2021 12:10 PM

(10-01-2021 12:01 PM)Paul Dale Wrote:  Three seems excessive Smile


You bet, two is bad enough.

The third is there helping out to prove the point that I can get away with a single processor. It is making it possible to figure it out just by tapping into the main board and not having to modify it - which is hard.

So far all is well and it is behaving the way I want, so two of them will soon go.



RE: HP97 The journey begins - teenix - 10-03-2021 12:56 AM


Murphy is still alive and mean as ever on this project.

One foot forward, ten steps back this time :-(

Everything was going ok, but yesterday the laptop I was using to design the circuit board wouldn't start up - dead as a doornail. I measured the power supply and it was showing zero volts. I had another supply which worked, so that was tried but still no go. I suspect either the laptop or power supply killed one another, or Murphy did.

More Arrrggghhh!!

This means I now have to migrate to a different PCB designer with all the associated problems. I started using KiCAD but understandably it only imports the gerber files. The data is imported as vias and tracks only, no components, pads etc. This means the schematic has to be redrawn, new components designed and then the PCB needs all the project components re-designed and the layout done from scratch. Many many hours of work. As I am unfamiliar with what this software creates, the final board would be a gamble and hope all the dimensions and a few important component placements stay the same or the board will be junk.



RE: HP97 The journey begins - ThomasF - 10-03-2021 06:02 AM

Hi Tony!

Sorry to hear, but would it not be possible to mount the drive from the laptop (eg as a USB-drive) to another laptop/computer and extract the files?
I have saved many programs and data that way from failing laptops (completely dead ones or with cracked screens).

My 2 cents, but maybe I'm just kicking on open doors...

Best wishes from the other pole!


RE: HP97 The journey begins - teenix - 10-03-2021 07:10 AM

Hi Thomas,

The files are ok, I have copies on a backup drive. The software that uses them is unworkable on anything including and beyond Win7. I was hoping to get through this project before having to upgrade anything for use on Win10.



RE: HP97 The journey begins - ThomasF - 10-03-2021 07:36 AM

Ah, then the penny dropped!
Yes, then I see your problem, so you need a laptop with Win98 or XP ...

Have you tried VirtualBox? Eg. as shown here?


RE: HP97 The journey begins - teenix - 10-03-2021 11:29 AM

(10-03-2021 07:36 AM)ThomasF Wrote:  Ah, then the penny dropped!
Yes, then I see your problem, so you need a laptop with Win98 or XP ...

Have you tried VirtualBox? Eg. as shown here?


Tried it, after 3 hours it seems buggy, GPF faults every second go at using it and I cant seem to put the directories/files I need into the virtual space.



RE: HP97 The journey begins - ThomasF - 10-03-2021 03:32 PM

Ok, sorry for taking your time ... Sad
I've used it mainly for running Linux under Windows and that has worked fine.
So, I presumed this could have worked as well as described.

Hope you find another way to resolve it!

Best regards,