Threaded Mode | Linear Mode

albertofenini · (This post was last modified: 10-28-2019 03:18 PM by albertofenini.)

Hi All,
this unit came with the HP-65 1333A19931 that we just restored (HP-65 1333A19931 Full restoration)
It had a sticker saying "Top Row Dead"
While we were expecting to find a row of led segments not working, it had and issue with all
the top keys (A,B, ...) and randomly the RTN key.
What happened was that pressing the A key, as example, rather the computing the 1/X function was sending the unit
in an unknown non responsive state with the display off.
It was then necessary to turn it off and on to bring it back to life.
So we decided to open it and clean it, other then rebuild the card reader.
It must said that when we put this very logic board into the HP-65 1333A19931 it was not presenting the issue.
Though we didn't try to read or write cards when this logic board was in the HP-65 1333A19931 chassis.
You can follow the work here : HP-65 1601A03250 Full restoration
We have not done anything cosmetic if not washing the keys and the chassis since this was not the goal for now
We re-assembled the issue with the top row keys was gone.
However, trying to use the card reader was generating random weird behavior
Some times it was reading, sometimes not
However as of now we haven't been able to write any card.
Then we thought of a possible incompatibility, has we had already experience, between the
keyboard logic and the other boards : card reader and logic board.
Since we had a spare keyboard logic of a previous generation unit we tried to swap
Don't pay attention to the low light led for the 2nd and 3rd digit, this keyboard already had this issue.
However, we didn't solve anything, since also this combination gives the same issues.
Surprisingly when the unit looks locked in some unknown state, it still can read cards ...
The clips are done with the spare keyboard logic ...
Any idea will be very welcome .... thanks in advance !!

teenix · 10-07-2019, 10:51 PM

(10-07-2019 09:24 PM)albertofenini Wrote: ...
We re-assembled the issue with the top row keys was gone.
...
Surprisingly when the unit looks locked in some unknown state, it still can read cards

Possibly a poor connection which was cleared when changing the board around. Pin 18 of the CPU to keyboard connector controls the keyboard row for the A - E keys from the CTC chip in the hybrid circuit (pin 22).

The card control is by hardware in the 65. As long as the CPU board is functional, card access should work even though the microcode is not working as it should. This couldn't happen in the 67 as card access is hardware and software controlled.

This makes the 65 more "finicky" with card access because the card reader motor speed has to be closely matched to the CPU clock rate.

cheers

Tony
cheers

Tony

albertofenini · 10-07-2019, 11:05 PM

ciao Tony,
what do you suggest to troubleshoot the behavior of this unit ?
take care !

teenix · 10-08-2019, 02:00 AM

(10-07-2019 11:05 PM)albertofenini Wrote: ciao Tony,
what do you suggest to troubleshoot the behavior of this unit ?
take care !

Hard to say.

I would use your oscilloscope to check the keyboard row outputs and make sure they change state when the appropriate keys are pressed.

I have a simplified HP65 schematic in this zip file which may help.

http://www.teenix.org/classSchm.zip

The card read/write signals can be confirmed with the oscilloscope also.

cheers

Tony

albertofenini · 10-08-2019, 06:03 AM

Hi Tony
that's a good start, can you please help us in telling what should we measure ?
I mean between which points should we put the probes and what to expect when a key is pressed ?
We have never done it and it's a good thing to learn !
thanks in advance !

teenix · (This post was last modified: 10-08-2019 06:42 AM by teenix.)

(10-08-2019 06:03 AM)albertofenini Wrote: Hi Tony
that's a good start, can you please help us in telling what should we measure ?
I mean between which points should we put the probes and what to expect when a key is pressed ?
We have never done it and it's a good thing to learn !
thanks in advance !

I'm pretty sure each of the 8 row outputs go logic low in a sequential manner. These pins are marked on the hp65 schematic. Then the 5 columns are tested and if any are logic low, one or more keys are pressed. This is a typical multiplexing arrangement which reduces the pin count for chips and connections..

Place an oscilloscope probe on row 1 (pin 18 of keyboard connector), for monitoring the A B C D and E keys. When you see a low signal, the CTC is now checking these to see if any of them are pressed.

If any or all of these keys are pressed, you should see a matching logic low signal appear on the corresponding column input.

For example, when row 1 is low and the A key is pressed, column 1 should also go low.

cheers

Tony

albertofenini · 10-08-2019, 09:48 PM

Tony ... we fixed it !!
Look at the last three pictures of the album we published yesterday
First we changed the motor on the card reader and it didn’t improve
Then we mounted an older card reader on the keyboard logic that came with unit
And it worked ...
We will publish the versions of the boards but I’m convinced
that there are combinations that work and some that don’t work

Also I think this unit had been tampered
The case has a serial number 1601A but the CPU board is older
The switches were one in gold finish and the other in steel finish like those found on the HP67
So I think someone had mixed up things a bit ...

However we will measure the signals with t he scope for the keyboard
We want to learn something more about the HP65 ...
Thanks for help !!!

teenix · 10-08-2019, 11:28 PM

(10-08-2019 09:48 PM)albertofenini Wrote: Tony ... we fixed it !!
Look at the last three pictures of the album we published yesterday
First we changed the motor on the card reader and it didn’t improve
Then we mounted an older card reader on the keyboard logic that came with unit
And it worked ...
We will publish the versions of the boards but I’m convinced
that there are combinations that work and some that don’t work

Also I think this unit had been tampered
The case has a serial number 1601A but the CPU board is older
The switches were one in gold finish and the other in steel finish like those found on the HP67
So I think someone had mixed up things a bit ...

However we will measure the signals with t he scope for the keyboard
We want to learn something more about the HP65 ...
Thanks for help !!!

Good news, well done

Are the part numbers on the card RW ICs the same on the different boards?

cheers

Tony

albertofenini · 10-13-2019, 06:50 PM

Hi Tony
The part numbers are different
The sense amplifier is different
If you look at the last four pictures of the album you can see the differences
The original one came with the unit
The final one we had as a spare and is older (we think)
Definitely pcb release is different and the sense amplifier is different
Since we didn’t take any measure we cannot say if the original magnetic head was working or not ...
Take care !!!

teenix · (This post was last modified: 10-13-2019 08:37 PM by teenix.)

(10-13-2019 06:50 PM)albertofenini Wrote: Hi Tony
The part numbers are different
The sense amplifier is different
Take care !!!

Thanks

Looks like the

1826-0158 is for the HP-65

and the

1826-0322 is for HP-67, HP-97, HP-41 82104A card reader, but seems to work on your HP-65.

They might be different because of the different ways these calculators read/write the cards.
Maybe the 1601A03250 HP-65 had a replacement done at some stage.

cheers

Tony

albertofenini · 10-13-2019, 08:28 PM

Tony did you checkout the pictures ?
Let us know if you want us to take more pictures or need any measure
Take care !

teenix · 10-13-2019, 08:49 PM

(10-13-2019 08:28 PM)albertofenini Wrote: Tony did you checkout the pictures ?
Let us know if you want us to take more pictures or need any measure
Take care !

I did thanks :-)

cheers

Tony