Which resistance controls the card reader speed on a HP-97

[Francois Lanciault]

Hi,

I just finished refurbishment of one of my HP-97 card reader but this time I have a problem. I cannot get rid of the ERROR message when reading a card.

I have changed the rubber rollers and cleaned everything. I have check the alignment of the nylon ball switches and verified that they work as intended with a voltmeter. The clutch is working well also.

The card goes thru smoothly but I always get an error message when I try to read cards. Write operation do not produce error messages but I don't think any check is performed on write operation and reading card written this way also produce an error.

So the only thing that looks not normal is the speed of the card going thru the reader. It is faster than what it should be. I have adjusted the eccentric nylon shaft to all possible position but the speed is always too fast.

I had the same problem once with a HP-67 and I added a trimmer to adjust the speed and eventually it work as intended. The information on where to install the trimer was available somewhere on the net... but I cannot find this information for the HP-97.

Sure it can also be some defective chips, but first I would like to get the speed right before giving up.

Anybody knows ??

Thank you.

François

[teenix]

There is some card reader maintenance information in the HP-97 Service Manual, including how to check speed.

cheers

Tony

[Kees Bouw]

Hi Francois,

What a coincidence! I also finished restoration of my HP-97 this weekend and I also could not get the card reader to work. The exact same problem as you describe: card runs through smoothly but always results in "Error". I removed resistor R1 and soldered wires so I could test resistor values outside the calculator. Unfortunately, it made no difference at all. The original resistor was 4.42K and I tried many values between 3K an 9K. The higher the value, the faster the card goes through. But the error was always there. I don't really know what to do next, I am still thinking about it. It is difficult to test because you can't see what is happening and you can't adjust the eccentric cam unless you take the calculator apart. I think the service technicians back in the day must have had special test equipment to fine tune the card reader outside of the calculator. Please let me know if changing the resistor works for you.

Kees.

[brickviking]

(11-07-2017 12:04 AM)Bouw Wrote:  Hi Francois,

What a coincidence! I also finished restoration of my HP-97 this weekend and I also could not get the card reader to work. The exact same problem as you describe: card runs through smoothly but always results in "Error". I removed resistor R1 and soldered wires so I could test resistor values outside the calculator. Unfortunately, it made no difference at all. The original resistor was 4.33K and I tried many values between 3K an 9K. The higher the value, the faster the card goes through. But the error was always there. ...

Kees.

I was thinking (if you can find such a thing) you could use one of those variable resistors that you turn with a little screwdriver, I think they have about ten turns to go from bottom to top of range. Stick that in series with a 3.3kohm resistor, and see whether you have any further luck. You could conceivably stick the variable resistor in parallel with a fixed resistor (say 8.2kohm) and combine the pair with a 3.3kohm series resistor, but that is getting a bit more refined. You may even need a 1% value resistor.

Just my NZ$0.02 worth...

(Post 128)

[teenix]

The manual mentions having some special stuff (Magna-See) to see the length of the magnetic data tracks to determine which resistor to use to trim the motor speed, so that the track length is within tolerances. The faster the card passes the longer the data track appears on the card therefore it will probably have missing data.

Slow card feeders have tracks that are too short will probably have data bits too distorted to be read properly by the calculator hardware.

The card is supposed to move at a nominal 6cm per second +/- 5%, so you might be able to time the passing of the card and adjust the resistor to suit.

A smaller resistor value should slow the motor down and shorten the data tracks.

The minimum specified is around 2800 ohm (2K8).

The median resistance for the adjustment resistor is around 4700 ohm (4K7)

As mentioned in another post a variable resistor would help. A miniature Bournes type multi-turn 100Kohm would do with a 6800ohm (6K8) resistor in series to limit the minimum resistance. Both components are commonly available. Connect the 6K8 resistor to one of the outer connectors on the Bournes. Then make connections from the other end of series resistor and the wiper of the Bournes (Centre connector).

If you clip these 2 wires directly across the speed resistor, that will give a range of around 2K7ohm to 4K5ohm. This will avoid having to desolder the original part for testing.

If this doesn't work, then there might be other issues.

cheers

Tony

[Francois Lanciault]

Thank you all for your contribution.

I will try playing with the value of resistance R1 (using a variable resistance) and report the result. Now I only need to find where that resistance is located on the board. I will check the maintenance manual.

However, I just find out that the cards written by the problematic HP-97 can be read on a HP-67... That means that the speed is not so much different after all...

François

[Geoff Quickfall]

Ah, important info, that means the 97 write circuit is correct.

Starting to sound like the read circuit is the problem in the 97. Have you written a card in the 67 and tried reading it in the 97.

Geoff

I have had quite a few read only or write only readers and have corrected all with shaft adjustment. I tweak the shaft turning only a few degrees at a time and testing. Less then a 1/32 of an turn per test.

[AndiGer]

@Bouw
You can test the reader without assembling and disassembling every time.
Look at that repair guide, picture 19 + description:
http://www.rskey.org/CMS/index.php/the-library/102
"To test the card reader, assemble the frame and case as shown. Don't forget to connect the keyboard and display connector. Then, pull the ribbon cable through the frame and attach the card reader. This makes it easy to test without assembly and disassembly. Adjust slotted shaft until the card reads and writes properly."
Andi
BTW: Recently I encountered the same problem too. So we are 3.

[AndiGer]

Today tried another card reader in the machine. Worked - but read only one card before the coupler of the replacement reader went to "to-be-repaired".
This proofs that the problem is inside the original card reader. Don't know yet what its problem really is. Suggestions or even solutions are welcome.
Andi

[Kees Bouw]

@AndiGer, thank you for your tip regarding testing the card reader without disassembling the calculator. This made it a lot easier. I have attached a picture of my setup with the failing card reader :-)

I have tried reading a card with all resistor values listed in the service manual, I have tried 4 different rollers: square o-rings, round o-rings and two sizes of silicon tubing, I have also tried several positions of the eccentric cam. But nothing worked. At best, I could read one particular card (diagnostics) some of the time. All other cards resulted invariably in "Error". So I gave up - for now. Fortunately I have another HP-97, also original and not restored. I decided to restore the card reader from that calculator in exactly the same way and I attached this card reader to the calculator I was working on. From the first try it worked, and it still works, flawlessly. I have not seen a single card that gave an error. Writing also works fine. The motor current is with slightly over 180 mA within range. It still has the original resistance. This encourages my opinion that the problem with the first card reader is not the resistance or the roller or the cam, it must be something else. Maybe there is still some tiny piece of dirt on the magnetic head, or maybe there is something wrong with the flat cable (it is in poor condition, the flat cable of the second card reader seems to be of better quality), or maybe the motor does not give a constant speed, or maybe there is a weak component on the pcb. Just some thoughts. I will try again later. In the service manual there are some pictures of scope signals, maybe I should get a scope and see if that helps to pinpoint the problem. For now, I have one fully working HP-97 and I am very happy with it!