HP-65 1605A02841 - weird display and card reader issues

[albertofenini]

Hi Everyone,
this an update of an older post, basically this unit has an hybrid pcb and a keyboard/led pcb with this issue :

When used with its own hybrid the display turns on the requested segments for the requested digits PLUS all the decimal points and all the C segments

When used with any other different (known working) hybrid the display turns on the requested segments for the requested digits PLUS the decimal points
and the C segments for the digits actually used plus the one after the last one

I think the pictures will better describe the situation

All of the waveforms have been taken using a known working hybrid logic board

Quite surprising all the waveforms are of the correct form and value

What could it be the problem ?
There are no open coils but we don't know how to test if some coils are shorted together

Unluckily we have not taken the waveform for Anode segment C when it should not be on, like when the display is showing a 2

Thanks in advance for any advice !

Pictures : HP-65 1605A02841 - wrong display behavior

[teenix]

(11-14-2020 12:16 AM)albertofenini Wrote:  Hi Everyone,
this an update of an older post, basically this unit has an hybrid pcb and a keyboard/led pcb with this issue :

When used with its own hybrid the display turns on the requested segments for the requested digits PLUS all the decimal points and all the C segments

When used with any other different (known working) hybrid the display turns on the requested segments for the requested digits PLUS the decimal points
and the C segments for the digits actually used plus the one after the last one

I think the pictures will better describe the situation

All of the waveforms have been taken using a known working hybrid logic board

Quite surprising all the waveforms are of the correct form and value

What could it be the problem ?
There are no open coils but we don't know how to test if some coils are shorted together

Unluckily we have not taken the waveform for Anode segment C when it should not be on, like when the display is showing a 2

Thanks in advance for any advice !

Pictures : HP-65 1605A02841 - wrong display behavior

All DP LEDs on indicate a low battery voltage condition.

If the batteries and Vbsw on the anode driver are ok, then the 220K resistor near the anode driver may be open circuit or faulty.

cheers

Tony

[albertofenini]

Hi Tony
the fact that all decimal points and all the C segments are turned on only happens when its own logic board is used.
With a different logic board not all the decimal points and not all the C segments are lit, just the ones needed for the requested number to show plus the next one
What could it be ?
Will check the resistor anyway and let you know
take care !!

(11-14-2020 12:34 AM)teenix Wrote:  

(11-14-2020 12:16 AM)albertofenini Wrote:  Hi Everyone,
this an update of an older post, basically this unit has an hybrid pcb and a keyboard/led pcb with this issue :

When used with its own hybrid the display turns on the requested segments for the requested digits PLUS all the decimal points and all the C segments

When used with any other different (known working) hybrid the display turns on the requested segments for the requested digits PLUS the decimal points
and the C segments for the digits actually used plus the one after the last one

I think the pictures will better describe the situation

All of the waveforms have been taken using a known working hybrid logic board

Quite surprising all the waveforms are of the correct form and value

What could it be the problem ?
There are no open coils but we don't know how to test if some coils are shorted together

Unluckily we have not taken the waveform for Anode segment C when it should not be on, like when the display is showing a 2

Thanks in advance for any advice !

Pictures : HP-65 1605A02841 - wrong display behavior

All DP LEDs on indicate a low battery voltage condition.

If the batteries and Vbsw on the anode driver are ok, then the 220K resistor near the anode driver may be open circuit or faulty.

cheers

Tony

[teenix]

(11-14-2020 12:45 AM)albertofenini Wrote:  Hi Tony
the fact that all decimal points and all the C segments are turned on only happens when its own logic board is used.
With a different logic board not all the decimal points and not all the C segments are lit, just the ones needed for the requested number to show plus the next one
What could it be ?
Will check the resistor anyway and let you know
take care !!

Time has run out, I have to go and start training a new pilot, so will get back when I can :-)

cheers

Tony

[albertofenini]

Thanks for all the support you gave us !!!!

[teenix]

Does the fault move with the faulty logic board is placed in a known working calculator?

Is the Disp2 lines always HI? (Pin 3 of anode driver) This would cause the C segments to be on all the time.

I'm not sure why all decimal points are on other than low battery detection from the anode driver. Does the battery voltage drop when this board is working but in the faulty situation?

cheers

Tony

[AndiGer]

Is it ok the first 4 digits are less bright than the others? I ask as we have 3 blocks of 5 digits forming the display...

[albertofenini]

(11-14-2020 07:56 AM)AndiGer Wrote:  Is it ok the first 4 digits are less bright than the others? I ask as we have 3 blocks of 5 digits forming the display...

Hi Andi
It is not ok but we do not have spares at the moment to replace the first block
We have already found this problem in other units and it is a kind of degradation
Usually it is fixed by replacing the block
If we will not be able to fix the led board that doesn’t turn on (it’s another post of our) we will certainly have a donor for that
Take care !

[albertofenini]

Hi everyone, a couple of updates ...

Since we weren't able to fix a lcd/keyboard from another unit described in this post HP-65 1603S03504 - display not working at all
we decided to use one of those led block to replace the first one of this unit which was defective

After that the decimal point started to work correctly
However all the C segments C where still on.
We found out that this happens with this logic board and any led/keyboard is paired with.

On the contrary, when a working logic board is paired with this led/keyboard everything works fine, both the decimal point and the C segment behave like they should.

Then We have taken some snapshot of the Disp0, Disp1, Disp2, Disp3, Disp4 and RCD signals from both a working logic/led/keyboard and this unit looking for differences.
Waves in red are from this unit waves and waves in blu are from a working assembly.

We noticed that this logic board has a different Disp0 and a different RCD from a known working unit .

Any idea if this is fixable and how ?

Also we completed the restoration with a card reader
However, while it reads consistently well, RA and RB are clean as they should
it doesn't write correctly and in fact WA and WB are not in good shape.

What are the odds that the sense amplifier is faulty or the head is faulty ?

Please note that for the card reader, blu and red waves are from the same unit
Thanks in advance for any idea you may want two share

Updated pictures : HP-65 1605A02841 - wrong display behavior updated album

[teenix]

It looks like the Disp0 and RCD have problems with amplitude.

I have E-Z Hooks that I can connect the CPU board to the display board. I can then open one or more connections to see if that changes the amplitude of the signal. This may help to narrow down the problem

cheers

Tony

[albertofenini]

Hi Tony

Those are nice !!
I wonder if we can buy them over here ...

About the card reader what do you think ?
Could it be related to the the display issue ?

We have tested this logic board in another unit, and while the display behaves the same, the card reader is handled well both in reading and writing.

Will try to put together another card reader and see what happens

take care and thanks for the great help !!"

(11-15-2020 03:47 AM)teenix Wrote:  It looks like the Disp0 and RCD have problems with amplitude.

I have E-Z Hooks that I can connect the CPU board to the display board. I can then open one or more connections to see if that changes the amplitude of the signal. This may help to narrow down the problem

cheers

Tony

[albertofenini]

Hi Tony, Hi everyone

we replaced the following pieces and tested in between the changes

- the card reader amplifier
- the motor capacitor
- the 22microF capacitors on channels A and B on the card reader PCB
- the magnetic head

without getting a better result

while changing the amplifier we checked for hidden oxidation spots or
broken traces but everything is ok on the card reader pcb

we didn't change the motor .. but we will replace the clutch since it started to slip and fail reading cards

Take care !

[albertofenini]

Hi Tony

after reading the document and asking so many questions in the previous post,
we have uploaded two more images, HB and CB during a READ and during a WRITE

Also we have rebuilt the clutch in a new way that we will illustrate in a separate post

Hope this helps in finding what makes so dirty WB when writing, if it is the HYBRID or if it is the SENSE AMPLIFIER or something else ...

Take care !