Threaded Mode | Linear Mode

wallyt · (This post was last modified: 02-24-2019 03:25 PM by wallyt.)

Bob, Tony, Dave, Hans-Peter...thanks for the info! I got my HP65 working so now I will concentrate on the 97.

teenix · 03-24-2019, 01:23 AM

Pretty cool I think :-)

cheers

Tony

Paul Dale · 03-24-2019, 02:32 AM

Very nice.

I'm assuming we'll get the corresponding message on a 97 is due course Smile

Pauli

wallyt · (This post was last modified: 03-26-2019 01:25 PM by wallyt.)

After getting the card reader working (new pinch roller and clutch) I decided to repair the printer on my HP97. The printer would not advance the paper even though the gear train was in fine condition. I suspected the problem was in the condition of the two pinch rollers.

I was not brave enough to disassemble the printer enough to replace the two paper feed rollers so I elected to try a somewhat less invasive method. I first very carefully cleaned the two rollers in situ with a cotton swap and isopropyl alcohol. I then applied to the circumference of both rollers two wraps each of 1/8th inch wide automotive pin striping tape. The width of this tape is nearly a perfect match for these rollers and the tape is durable enough to withstand application on a car body. Working slowly and carefully I was able to do this without disassembly of the printer. As this tape has a very shinny surface, for good measure I carefully applied small amounts of Krylon Matte Finish spray coating, using a very tiny brush, to the surface of tape on each roller. Again, one must do this very slowly, applying the coating to a small area and allowing it to dry before moving on. To facilitate rotating the rollers while I applied the tape and Krylon I left a length of paper about one foot long in the printer. When I needed to rotate a roller I would gently pull the paper up through the printer an inch or so at a time.

After reassembly of the calculator I was very pleased to see the printer work as it should. I realize that the best fix would be a complete replacement of the pinch rollers but what the heck...we'll see how long this lasts!

Dreato · 03-26-2019, 05:11 PM

Clever fix! And you avoided the risk of all those clips and springs shooting everywhere!!

(03-26-2019 01:23 PM)wallyt Wrote: After getting the card reader working (new pinch roller and clutch) I decided to repair the printer on my HP97. The printer would not advance the paper even though the gear train was in fine condition. I suspected the problem was in the condition of the two pinch rollers.

I was not brave enough to disassemble the printer enough to replace the two paper feed rollers so I elected to try a somewhat less invasive method. I first very carefully cleaned the two rollers in situ with a cotton swap and isopropyl alcohol. I then applied to the circumference of both rollers two wraps each of 1/8th inch wide automotive pin striping tape. The width of this tape is nearly a perfect match for these rollers and the tape is durable enough to withstand application on a car body. Working slowly and carefully I was able to do this without disassembly of the printer. As this tape has a very shinny surface, for good measure I carefully applied small amounts of Krylon Matte Finish spray coating, using a very tiny brush, to the surface of tape on each roller. Again, one must do this very slowly, applying the coating to a small area and allowing it to dry before moving on. To facilitate rotating the rollers while I applied the tape and Krylon I left a length of paper about one foot long in the printer. When I needed to rotate a roller I would gently pull the paper up through the printer an inch or so at a time.

After reassembly of the calculator I was very pleased to see the printer work as it should. I realize that the best fix would be a complete replacement of the pinch rollers but what the heck...we'll see how long this lasts!

teenix · 03-26-2019, 10:43 PM

(03-26-2019 01:23 PM)wallyt Wrote: ....I carefully applied small amounts of Krylon Matte Finish spray coating, using a very tiny brush, to the surface of tape on each roller.

I was thinking of doing something similar by building up a few layers of liquid gum to each roller and let it dry but haven't had time to as yet.

I was also thinking that the idler rollers that these butt up against are a lot smaller and may take to some shrink wrap to build up the diameter a tiny bit which should achieve the same effect.

cheers

Tony

wallyt · 03-27-2019, 01:58 PM

My solution definitely shows that, in my case, the problem was the rollers. It is not a permanent solution. Disassembly of the paper feed mechanism and using O-rings seems the best way. Applying heat shrink tubing to the rear rollers sounds like a good idea since there would not be any interference with the print head or ratchet on the front rollers.

There is a liquid rubber solution that is used by watercolor artists to mask-off portions of their art work. After it dries it is removable but becomes more permanent the longer it stays put (it's also quite expensive). I was thinking of using this to build-up the front rollers before I hit on the idea of using the pin stripe tape.

teenix · 04-08-2019, 05:02 AM

Hi all,

I guess maybe a woo hoo!! is in order.

Probably the first time in history a magnetic card was written/read in a HP 67 from a home made CPU board. For me, the original CRC chip does still have a few hidden secrets but I'm pretty sure I've got it right. Anyway, the data on the logic trace looks pretty good. The data block stored on the card is well within the card ends and looks similar to the "magnasee" images in the HP journals.

The card read/write is still controlled by the same method as the original calculator hardware, which is directly from the HP67 original microcode, so I'm hoping all the other functionality like expecting card #2, data cards, pause to reload etc all work.

Still a lot of testing to go, but encouraging.

I thought of a new user option too (if it works) which is that the new cpu can time the data as it passes the read/write head and report the actual bit time. That way, eccentric cam adjustments might be easier.

cheers

Tony

teenix · 04-08-2019, 05:27 AM

(04-08-2019 05:02 AM)teenix Wrote: That way, eccentric cam adjustments might be easier.

Sorry, meant speed adjustments.

cheers

Tony

Paul Dale · 04-08-2019, 06:49 AM

Woo hoo!!

Massimo Gnerucci · 04-08-2019, 06:56 AM

(Woo hoo!!)²

**rprosperi** · 04-08-2019, 12:58 PM

WOO F^$#ing HOO!!

Amazing accomplishment Tony, well done. And I think it's safe to say it's certainly the first time this has been accomplished. Read/Write via original microcode succeeding is the ultimate test of compatibility.

Dreato · 04-08-2019, 03:02 PM

Great work, as always!

teenix · 04-10-2019, 02:39 AM

Hi all,

Unfortunately the woo hoo turned into a boo hoo :-(

It figures just when you think you are getting somewhere.

I had to measure a track on the keyboard PCB to clarify something and as careful as I have been with the hardware, when I reassembled everything I didn't realise that the brass power switch had slipped out of its carrier and was resting on the circuit board. It shorted some tracks and looks like the cathode driver has now failed.

This is a plea to see if anyone has a spare 1820-1749 for the 67.

If not, I think I can recreate the functionality on a different chip. It is basicly just a shift register that has to sink a single LED current, and it also has the battery voltage monitor.

cheers

Tony

Dreato · 04-10-2019, 05:19 AM

I'll PM you.

Harald · 04-10-2019, 08:03 AM

(04-10-2019 02:39 AM)teenix Wrote: Hi all,

Unfortunately the woo hoo turned into a boo hoo :-(

It figures just when you think you are getting somewhere.

I had to measure a track on the keyboard PCB to clarify something and as careful as I have been with the hardware, when I reassembled everything I didn't realise that the brass power switch had slipped out of its carrier and was resting on the circuit board. It shorted some tracks and looks like the cathode driver has now failed.

This is a plea to see if anyone has a spare 1820-1749 for the 67.

If not, I think I can recreate the functionality on a different chip. It is basicly just a shift register that has to sink a single LED current, and it also has the battery voltage monitor.

cheers

Tony

I am currently on a business trip. But when I get back home I will have a look!

Cheers,
Harald

Robert VM · 04-16-2019, 12:09 PM

Success to Tony with the 97 project!! Would be interested to hear how you go about recreating ACT (CRC) functionality with the correct timing.

If I may join the HP97 repair corner...
There's this item that came at a reasonable price but which seemed defective. After opening up, the card reader cable had obviously been put in shifted, shorting adjacent contacts in the connector.
After putting that right the calculator behaved erratically: occasional display of zeroes: "00" or "0.00",
almost no response to the keyboard. Later the printer printed out lines with mostly 'N's a single 'I' and '0's.
Then the idling gear lost some teeth.
To continue the search I disconnected the printer board: no activity at all...
That was some time ago.
I did transplant the 97 ACT into a HP-21 and that seemed to function well.

Just recently, looking through the schematics it became rather obvious why nothing functions when the printer board is disconnected. The oscillator elements for the ACT, that I couldn't find on the logic board, are situated on the printer board!
The interconnects for these elements are somewhat 'powdery white': oxidation?
Haven't yet done the extra cleaning of these contacts...
Does anyone has an idea why these oscillator components were put on the printer board and not near the ACT itself?
Likewise the card reader board and display board have some of the power supply components.

teenix · 04-16-2019, 10:05 PM

(04-16-2019 12:09 PM)Robert VM Wrote: Does anyone has an idea why these oscillator components were put on the printer board and not near the ACT itself?
Likewise the card reader board and display board have some of the power supply components.

I wondered that myself, there is plenty of room for the oscillator components on the cpu board. It seems they wanted the printer to be in circuit for the calculator to function properly which seems odd.

The printer state isn't tested during the microcode initialization process but if you hit any key that tries to start the printer when it is disconnected, the microcode will stay in a permanent loop until the cpu "sees" the print head is home. This effectively disables the calculator and makes it appear dead.

It makes sense to keep some of the power circuitry near the components that they will control.

cheers

Tony

teenix · (This post was last modified: 04-25-2019 01:53 AM by teenix.)

Hi all,

I've been making some more progress with the HP67 replacement CPU board.

It is really time consuming to debug software that is running other software that can also talk to a PC running more software :-).

I have been building code for the external program storage/recall plus a menu interface to access them. The menu is activated by holding the "h" key down for 1/2 second or longer. Then you can scroll through the items with the "+" and "-" keys and make a selection with "Enter".

I transferred the HP67 diagnostics "A" cards from the PC to the HP67 external program memory and loaded it from the new menu.

If anyone is interested here is a small video of the result. It shows the HP-67 PC emulator and a HP-67 with the new CPU board both running the same code.

http://www.teenix.org/HP67diag.mp4 (about 20Meg)

The PCBs arrived today with the modified design so I have to build one and try it out.

cheers

Tony

Massimo Gnerucci · 04-25-2019, 09:33 AM

(04-25-2019 01:47 AM)teenix Wrote: http://www.teenix.org/HP67diag.mp4 (about 20Meg)

Mesmerizing!

:-) and that's how I looked at it...