HP-67 - 1612A03018 - Card Reader Advice Requested (was HP-67 - strange behavior)

[albertofenini]

Then we changed card reader with a HP67 style where we have changed all the capacitors
However the bakelite of this card reader (the part where the head is fixed) is very ruined
and we have trouble in inserting a blank card or a card from an HP65 standard pac.
We think that these three cards have probably different thickness and only the card from
the HP67 standard pac goes thru.
The card reader however produces a good RA and RB signal and some times read succesfully
the card.









Later we will try to combine the two card readers in a single working one,
will post the resutls here !

take care !

[albertofenini]

Hi Everyone, we fixed it !
As we said in our last post we have swapped the card reader bakelite and with this also
the magnetic head and the plate with the battery terminals.
We kept however the logic part of the card reader, where at the end we re-soldered the
original capacitor.
The replaced one was too big and didn't allow to close the case, we could have soldered
with cables but we gave a try to the old one and it was working fine.
We read many cards from the standard pac, both one sided and two sided and
we also recorded cards both one sided and two sided,
however, we were not able to read HP-65 standard pac cards.
Shouldn't this be possible for backward compatibility ?
Or was it between HP-41 and HP-67 ?
Never mind, in the next days we will try to repair the other HP-67 faulty CPU board
take care and thanks for all the support !!

New pictures and video here : HP-67 - 1612A03018 - Card reader Advice requested

[Thomas Okken]

(11-01-2019 10:01 PM)albertofenini Wrote:  however, we were not able to read HP-65 standard pac cards.
Shouldn't this be possible for backward compatibility ?
Or was it between HP-41 and HP-67 ?

I don't have a primary source at hand, but I'm pretty sure that cards written by the HP-65 cannot be read by the 67.

https://www.hpmuseum.org/cgi-sys/cgiwrap...ead=176632
https://www.hpmuseum.org/cgi-sys/cgiwrap...ead=172776

The HP-41 card reader can read cards written by the HP-67 and -97, but that's as far as magnetic card compatibility goes.

[albertofenini]

Thank you very much, we will finish and publish new pictures soon,
we still have to protect the layout with TS-80 and in this case we will
put an aftermarket label behind
take care Alberto

[AndiGer]

(10-20-2019 12:48 PM)albertofenini Wrote:  Write (apparently well) two sided card (we have been successfully only once)
but when we read back a double sided card, it goes well on the first side,
asks for the second side but is sounds "slippery" in the meaning that accelerate toward the end of the reading of the recorded second side and gives an Error message.

Did you find out what causes the "slippery" and the connected acceleration at the end of the card?
I had that as well several times but couldn't explain why ...
Maybe it is the card itself due to speed is data controlled?

In my cases some cards - often even only one side while the other side is ok - went thru smooth but with motor speed-up at the end quoted by an error message.

Thank you everyone
Andi

[albertofenini]

No, unluckily we didn't
I can tell you that it happens with this unit, even after the card reader swap, when we try to
read an HP-65 card.
We obviously get an error message after that.
Before we swapped the bakelite part (and the magnetic head of course) it was happening
pretty often even with the HP-67 cards.
My guess, now that it happens only with "non compatible" cards is that while reading is
missing a piece of information that tells the card reader about the length of the card and
that warns the card reader that the card is finished.
At that point the card reader keeps the motor running and because the card is physically over
we get that kind of acceleration noise.
This is just my two cents ... but any idea about why this happens is welcome
Take care !

[teenix]

(11-02-2019 07:27 PM)AndiGer Wrote:  Did you find out what causes the "slippery" and the connected acceleration at the end of the card?
I had that as well several times but couldn't explain why ...
Maybe it is the card itself due to speed is data controlled?

In my cases some cards - often even only one side while the other side is ok - went thru smooth but with motor speed-up at the end quoted by an error message.

Thank you everyone
Andi

The speed control is determined by battery voltage. If it was due to data bit spacing, the cards would go through the reader at different speeds which doesn't occur.

The motor speeding up briefly at the end of a faulty read is due to the way the HP67 microcode works.

The code evaluates the checksum before testing if the card has completed its travel through the reader. If a read error is detected, the program memory gets cleared also before testing if the card is out.

By this time the card has passed through the drive wheel which reduces the load on the motor. This effect must be outside the limits of the speed control and the motor speeds up. Shortly after, the code then tests if the card is out and because it is, turns off the motor.

cheers

Tony

[teenix]

(11-02-2019 10:15 PM)albertofenini Wrote:  My guess, now that it happens only with "non compatible" cards is that while reading is
missing a piece of information that tells the card reader about the length of the card and
that warns the card reader that the card is finished.

The card reader has no "smarts". Its operation relies solely on the switches and the CPU board plus the sense chip speed control. The only data that is on the cards is program info plus for the 67, extra data for checksum, card info and some calculator formatting.

cheers

Tony

[AndiGer]

(11-02-2019 11:22 PM)teenix Wrote:  The speed control is determined by battery voltage. If it was due to data bit spacing, the cards Would go through the reader at different speeds which doesn't occur.

The motor speeding up briefly at the end of a faulty read is due to the way the HP67 microcode works.

The code evaluates the checksum before testing if the card has completed its travel through the reader. If a read error is detected, the program memory gets cleared also before testing if the card is out.

By this time the card has passed through the drive wheel which reduces the load on the motor. This effect must be outside the limits of the speed control and the motor speeds up. Shortly after, the code then tests if the card is out and because it is, turns off the motor.

cheers

Tony

Hi Tony,
Thank you for clarification on speed control and in-sights to the ROM code.
Andi

[albertofenini]

Thanks to everyone for sharing the knowledge !

[AndiGer]

I guess the slipping cards and motor acceleration at the end of the card is due to faulty cards (due to age or storage?) ...

I have repaired a card reader inside a HP-97. Nearly every card had the slipping phenomen.
When I used another bunch of standard pac cards those cards were read ok.
So I re-wrote the unreadable cards and after re-write they now read well.

Hope this helps anybody running in that situation.

Andi