HP-65 - card reader with WB wave noisy - FIXED and RESTORED

[albertofenini]

Dear All
we tried in the past to repair one of our HP-65 with partial success
The unit is working, can read cards but cannot writes them and we discovered that WB wave is very noisy.
It was all described in this thread :
HP-65 more on the card reader
All the single components work if used in conjunction with other HP65 hardware but when put together in this unit we have this problem.
Since we received few days ago some parts including a keyboard PCB we wanted to try this one.
The keyboard is without Anode and Cathode driver which we have used to repair another HP65
However, we have connected the card reader assembly including the battery compartment,
we have connected the logic board and turned on.
Of course the display is off, but we can feed cards.
Using an ohmeter We have checked that PRGM switch works and so work the motor, HDS, WPS switches
Using the scope we have seen normal waves on both RA and RB channels
But WA and WB when the unit is set to PRGM/W and we feed a blank card are FLAT
Does anyone know if Anode and Cathode driver have any influence on recording a card ?
More generally, has anyone ever fixed a situation like this ?
As always thanks for help !!!!

[teenix]

(05-05-2019 09:30 PM)albertofenini Wrote:  Using the scope we have seen normal waves on both RA and RB channels
But WA and WB when the unit is set to PRGM/W and we feed a blank card are FLAT
Does anyone know if Anode and Cathode driver have any influence on recording a card ?
More generally, has anyone ever fixed a situation like this ?
As always thanks for help !!!!

Hi Alberto,

Does it give any signals with a recorded card?

Are WA WB connections between hybrid and card reader ok?

I wouldn't think the display drivers have any influence on the reader.

cheers

Tony

[albertofenini]

ciao Tony,
with a pre-recorded card from Standard Pac we have only looked at RA and RB and these are ok,
it's a fact that with the previous keyboard PCB when recording WA was good and WB was noisy,
we are measuring RA, RB, WA and WB both at the Hybrid pins and at the card reader pins.
We have not measured WA and WB when reading, nor with this keyboard nor with the previous
Will do it and report results !
Also we will check all the connections but they seems to be good.
I also don't think that the presence of the anode and cathode drivers should make any difference
however the calculator behaves different when using a keyboard complete or a keyboard driver-less
thanks again !

[albertofenini]

Ciao Tony
We played a little with the unit but if used with a “driverless” keyboard pcb
WA and WB remain flat, no matter the position of the PRGM switch or the type of card

We remounted the keyboard pcb with the led drivers and we returned to the initial situation, it reads but WB is noisy

We played a little with the capacitors on the card reader
If we detach the capacitor on the motor the unit will read
but when recording it will display a 00 00
way before the card is ejected
I understand that the capacitor on the motor has an influence on the writing behavior
Then we changed with a new 3.3 microF and it worked well in reading, took the right time to write but still WB too noisy
So we changed also the other capacitors on the card reader pcb but nothing changed

What drive us crazy is that the logic board works great in any another unit of the HP65 we have producing good waves and recording well

So do the card readers if mounted on other units but when joint together they behave weird

We have two card readers units both inclusive of the battery compartment and then
With the card reader switches on the back
Both behave equally bad
Could both have a damaged amplifier ??
Could be the heads ?

We have a third back panel which we will try tomorrow trying to recover an additional head and will see what happens

As always any idea is welcome !
Take care !!!

[teenix]

Hi Alberto,

Not sure what "driveless" means

A long shot maybe, but it was mentioned in a thread some time ago with a card reader not writing well. I think it was for a 67, but the same principal applies.

There are capacitors across Vbsw and GND in the power supply. 15uF and 60uF, maybe changing one or both of these might help.

cheers

Tony

[albertofenini]

ciao Tony,

with "driverless" we mean a keyboard PCB without Anode and Cathode drivers,
thus "driverless"

The test showed that whether there are or there are not the Anode and Cathode drivers on the keyboard PCB it makes a difference in the WA and WB behavior

We can surely replace the capacitors on the logic board and see what happen,
I also had read that post.

However, do you know how can we check the goodness of the magnetic head ?
At the moment we can not measure the impedance but we could measure the resistance values.
Do you know between which cables it should be done ?

Do you think that connecting a capacitor from the WB pin of the hybrid and the ground would help ?

Thanks for help !

[teenix]

(05-06-2019 11:10 PM)albertofenini Wrote:  ciao Tony,

with "driverless" we mean a keyboard PCB without Anode and Cathode drivers,
thus "driverless"

The test showed that whether there are or there are not the Anode and Cathode drivers on the keyboard PCB it makes a difference in the WA and WB behavior

We can surely replace the capacitors on the logic board and see what happen,
I also had read that post.

However, do you know how can we check the goodness of the magnetic head ?
At the moment we can not measure the impedance but we could measure the resistance values.
Do you know between which cables it should be done ?

Do you think that connecting a capacitor from the WB pin of the hybrid and the ground would help ?

Thanks for help !

Ahh ok.

Aren't the Phi1 Phi2 system clock signals generated from the anode driver? If that is missing, then I would assume the CPU won't work.

I would assume the
orange/blue pair
then
yellow/red pair

They are just small coils so should have some inductance/resistance to measure from a DC source. The impedance will also depend on a signal frequency through the coils.

A capacitor there might just distort the square wave signals.

The reader chip seems to have a bit of black magic in there

cheers

Tony

[albertofenini]

Guys we fixed it ...
We "borrowed" an Anode and Cathode drivers from an HP45
We wanted to have one more keyboard to test and we recovered the one received with some spare parts a couple of week ago,
basically is the one from which we removed the drivers because it was heavily corroded.
We didn't transplanted back those chips since we wanted to keep at least a proven working keyboard.
We had cleaned it with water and vinegar and it was came out fairly cleaned.
Once we have installed the Anode and Cathode drivers and assembled the unit we had it working, reading and writing.
Apparently the two keyboards are very similar but have different version numbers
The one working is marked 00065-60201 and has the following numbers 4321 C-1314-61
The one not working (WB noisy) is marked 00065-60201 with the following numbers 3351 B-1314-61
We have inspected the latter very carefully and there are no evidences of broken traces, oxidation or anything else
Seems absurd bat it was in very worst condition the one that we received and now is in place
So I'm assuming that with old hybrid (this machine is a 1333A serie) only a certain type of keyboards are fully compatible.
We have now to do all the aesthetic restoration and we will try do document it as much as possible
Thanks again for help !!!!

Below you will see the keyboard we used, before the cleaning process
and the two keyboard, the one with the gold domes was the one giving the erratic behavior

   
   
   

[albertofenini]

And now for the restoration process :
The unit was showing moderate wear, but for the fun of it we have separated every piece and :
1. polished the red plastic for the display using a common car polish and some cotton
2. refined the chrome trim line
3. sprayed the keyboard with the matte transparent finish
4. swapped the green top frame with a HP67's one since this was damaged
5. flattened as much as possible the original back label and glued back in place (this is process that we must improve although the actual result is not so bad)

Here are the pictures ..

   
   
   
   
   

Thanks everyone for the support !!!