HP-65 more on the card reader

[albertofenini]

Hi Everyone,

I would like to share few things on a card reader restoration (actually two) that are driving us nuts.
After successfully recovering two units, from garbage condition to being able to read ABD write cards,
yesterday we attempted this on two more units.
Both were working except for the card reader.
Basically we have three different type of materials for restoring the gummy wheel :
- o-ring
- square o-ring
- silicon tubing
Traditional o-rings are for me the best, the stay in place and give no problems.
Square -o-rings tend to separate, one remain close to base of the wheel, the other after few readings slips to the opposite side,
generating friction with the card reader structure and slowing down the motor
Silicon tubing (coming from TAS) is a little loose on the wheel (ID slightly too big) so we glued a little and it fits ok.
Also, opening this unit S/N 1511A..... we have found what was looking like a brand new motor,
with brand, model number and I have sent an email today to inquiry if it is still available.
This unit laso had a black "axle" differently from all the other we have where the axle is white.
(Is axle the correct term for this part ??)
It also features a very tiny point on the hard, wondering if this serves to position correctly the eccentric cam.
However, the worm gear although all intact was loose, and we swapped this motor with a different one
where the clutch has been somehow fixed and the strength is good.
I'm posting the first four pictures and I will continue with a second post in this same thread.

               

[albertofenini]

While repairing, it also happened us to break one of the contacts from the card reader to the main board.
Since it's a twin contact we hope this does not affect the functionalities of the card reader.
We tried to replace the gummy wheel with the following items :
- 2 traditional -o-ring
- 2 square o-ring
- only one square o-ring
- silicon tubing
In all of these cases we have not been able to record a single card.
Reading also is not 100% reliable
sometimes it reads all of the STANDARD PAC cards, sometimes it does not read on the first attempt,
but it will read on the following attempt
Sometime it reads, but when the program is launched it ends up in infinite loops with erratic results.
However most of the times it reads and it executes correctly the program.
It also can read cards recorded by the other two units we refurbished time ago.
When recording, it looks to us that the 00 00 digits will show up too early, when the card is still going thru the unit
We have also noticed that the noise that the motor does, it is different when the the unit reads a card fro when it try to records a card, does it make any sens ?
Lat, but may be not least, compared to the other units the card remains a little more inside the unit when read or recorded, like the motor was not able to pull it till the end.

Here are two Youtube links when the unit is reading and try to record

RECORDING https://www.youtube.com/watch?v=QAE8FesBPiI
READING https://www.youtube.com/watch?v=BjJuevQv6h0

What can we possibly try to get the machine more affordable in reading and successful in writing ?
We have already tried tens of different position with the eccentric cam without noticing any improvement

   
   
   
   

[albertofenini]

Last, but not least, we would like to improve our capability to renew the clutch of the motor
we have tried to follow this thread
HP 97 card reader clutch shaft repair
but we weren't able to find the right couple of cable as shown in the thread.
We have even tried with a old Apple Ipod earphone as indicated in message #7 but no way ...
As anyone found a good couple of cables ?
Thanks for any help you would like to share with us !!

[Geoff Quickfall]

The clutch is actually a dampening couple, not a clutch. It dampens the vibrations generated by the Swiss motor. Therefore the dampening material must absorb vibrations to prevent card reading / writing problems.

The original material was the natural rubber also used on the pinch roller and as such suffers the same rot.

I used, as indicated in the post you mention, two wire insulation sleeves from spare wires found in my hoard box. The white insulator is from defunct iPad ear buds. I used the thicker section of the iPad earbud insulation, which contains both left and right channel wires. The second red insulator is from a very small gauge which fits into the white insulator.

You will have to experiment or find similar gauge insulation sleeves.

Geoff

I sent an email to you.

[albertofenini]

(01-28-2019 11:03 PM)Geoff Quickfall Wrote:  The clutch is actually a dampening couple, not a clutch. It dampens the vibrations generated by the Swiss motor. Therefore the dampening material must absorb vibrations to prevent card reading / writing problems.

The original material was the natural rubber also used on the pinch roller and as such suffers the same rot.

I used, as indicated in the post you mention, two wire insulation sleeves from spare wires found in my hoard box. The white insulator is from defunct iPad ear buds. I used the thicker section of the iPad earbud insulation, which contains both left and right channel wires. The second red insulator is from a very small gauge which fits into the white insulator.

You will have to experiment or find similar gauge insulation sleeves.

Geoff

I sent an email to you.

Thanks Geoff,
we also used an old iPod earbud, we used the part close to the 3.5 mm jack, which contains
all the wires, but nevertheless it was too thin and was loose in the aluminium sleeve.
Unless you tell me that it get thicker and of the right size when we should put inside the smaller red tubing.
Is it this way ?

Also, what's your thought on our HP-65 not recording ? What could we check ?
I have read in some posts that it could be the card protect switch not working,
can you help us in understanding of the four gold flexible metal parts which one is the write protect one ?
Thanks in advance !!!

[Geoff Quickfall]

The red insulator fits inside the white insulator. The motor spindle is thinner then the worm gear spindle.

Therefore the red insulator slips over the motor spindle and the white insulator fits over the worm gear. The new dampening assembly fits inside the aluminum tube. The red insulator section fits inside the white insulator and only takes up half the length of the white insulator.

Check out this YouTube video on the Hap 41 c card reader repair: move to 22:43.

HP 41C card reader repair!

Cheers

[AndiGer]

Maybe Tony's repair he did lately helps?
http://www.hpmuseum.org/forum/thread-12156.html

[teenix]

(01-29-2019 05:01 AM)AndiGer Wrote:  Maybe Tony's repair he did lately helps?
http://www.hpmuseum.org/forum/thread-12156.html

I am waiting on a small tool that will enable me (hopefully) to cut a square shoulder on the inner drilled hole that will more closely match the worm drive shaft end. I don't expect much of a demand, but once I know it all works, and if anyone wants one they will be just for the cost of postage.

It is still working ok so far.

cheers

Tony

[wallyt]

I am having a somewhat similar problem with the HP65 card reader that I have been restoring. After cleaning the gunk out I used silcone tubing for the pinch roller and the reader transports the cards okay, but it does not seem to read or record. I tried all sorts of adjustments on the eccentric cam, no help. I measured the motor current and have been able to get it to 200mA by adjusting the cam. But the reader still does not read or record. When I attempt to read a card into the unit I get the same "endless loop" result that Alberto mentioned.

I got the idea the other day to give the sense amplifier a shot of freeze spray and then try the reader. WOW! It started to work fine. I was able to record a short program on a card, power cycled the unit, and read the card back in. Problem was that the good results were temporary, and I assume the area of the sense amp warmed back up.

I do not know if the sense amp was really faulty. The freeze spray also hit the adjacent fixed 5.11k ohm motor speed resistor and could have changed its value a bit. I decided to lift the resistor and replace it with a 10K pot pre-adjusted to 5.11K ohms. When I tried programming and reading a card back in the unit worked okay, but only for a couple of passes. I adjusted the pot slightly, up to about 5.8k and down to about 4.8k. I could hear the slight speed difference in the motor when feeding it a card. Unfortunately I could not reproduce the good results that I had earlier.

Very infrequently, when feeding a card I hear a screech from the unit. Today, I am going to repair the dampening "clutch". I don't know what else to try.

[Geoff Quickfall]

Heat or cold treatment, may cause a temporary connection on a cold solder joint or within an IC between a broken contact as you know.

I would remelt each solder joint in the amp circuit and check all other connections including the card reader PCA to Main PCA gold post connections.

The only fix for an intermittent connection on the IC is a replacement from a junked donor.

The dampening couple may be the source of the read write problem also. It is obviously failing as evidenced by the screeching noise. There should be no slippage here and the replacement material (see above links) should not slip but should absorb motor vibrations.

Resistors seldom fail so I don’t think it was that but introducing the pot resistor will aid in setting the card reader pass through speed. This speed is also controlled by the pressure exerted by the pinch roller which is adjusted by the eccentric cam.

Get a hold of the HP 97 tech manual (HPMuseum thumb drive) which gives the descri on how to adjust the eccentric cam plus the optic current draw as the cam is adjusted.

Manually adjust the cam once the entire assembly has been repaired and reassembled:

1. remelt the solder joints on the card reader PCA.

2. Replace the dampening couple material by reusing the aluminum sheath and inserting wire insulation of two different sizes. (See above YouTube links).

3. Clean card reader head (isopropyl alcohol).

4. I assume the ball bearing still resides in the card reader frame.

5. I assume the card reader sense fingers were not bent or accident readjusted under the card reader assembly and the balls, roller and spring guides are all there.

6. To manually check the eccentric cam: Once assembled run a known good preprogrammed card. If it doesn’t read adjust the cam by turning it five degrees. Continue to read and adjust until you hit the sweet spot.

Of course if it is the sense amp or the controller IC then a replacement will be needed.

Geoff

Pm sent.

[wallyt]

Geoff, thanks for the suggestions. I will try the solder reflow first. Worst case scenario would be a shot sense amp IC. I have a spare card reader from an HP41. I read somewhere that a fellow used the sense amp from it to replace the one in his HP65. Is this possible?

[Geoff Quickfall]

The HP65 / 67 / 97 have a lot in common but some of the hardware such as the worm gear are not interchangeable.

The HP 41 card reader is a complete reworking from the ground up. It was as if the HP65 plans were lost.

I haven’t checked out the compatibility of the sense amp between the two machines. It might make sense that the amp is the same but hardware wise many differences.

Can’t hurt or desolder and replace if all else fails.

Geoff

Email package sent! Let me know if it is helpful.

[albertofenini]

Dear all, a last update :

Last night we worked on a second unit, and we put some effort on rebuilding the so called "clutch"
At first we sacrificed an old iPod ear pod, but as you will see in pictures, the thickest withe cable is loose in the aluminium sleeve.
Then we read in a post that the WD40 red dispenser could be used but also that was loose.
We took them some measures, that you will see in the post.
We found that the white cable for the left or right cable is good as inner insulation, and we found a blue
cable from a 220V cable be good as external insulation.
The only thing is, that the two are too thick to also remount the aluminium, sleeve.
Question : is important to use the aluminium sleeve or it can be left out ?

   
   
   
   

[albertofenini]

At this point we had a strong coupling between the motor and the gear, but we must said that
when we have coupled the motor with the black wheel for the pinch the two became noisy, squealing ....
The motor alone even after the modification runs smooth and silent

Question : what can we do for that ?

However, we didn't improve the situation.
We measured the current consumption and it is 200ma when reading, 210ma when writing.
We re-soldered all the joints for the resistor, the capacitors and the controller.
We even swapped the controller with another one from a spare card reader
Nothing, it reads somehow, but it never ever writes ...
Contacts are clean, and are not bent in anyway.

The two HP-65 we are working on have different controllers.
The one mentioned at the beginning of this post has the same controller found on the HP67 the 1826-0322
This unit mounts the usual HP65 controller the 1826-0158 and last night we swapped this with another 1826-0158

So we have two different units, with two different types of controller, different types of pinch roller,
two different magnetic heads acting in the exact same way ... what could be wrong ?

I'm also posting a pictures for some possible capacitors replacement, what do you think ?
The ones mounted are marked as 15K and 22K, what kind of capacitors are these ? Tantalium ?
Can we use the one pictured ?

Thanks for any help !!!

   
   
   
   

[Geoff Quickfall]

Yes, you should use the aluminum sleeve. The sleeve h9lds the rubber / new insulation rigid. Without it the couple may distort. Also thesis sleeve is an excellent measure of the correct length for the assemble.

The new couple should fit inside the sleev and be no longer then the sleeve.

Also, check the pivot point for the worm gear on the calculator bracket; the point where the worm gear touches the bracket. Is the small bearing still there? It is pressed into the bracket but may have fallen out. This is why the new couple must b exactly the same length as the aluminum sleeve.

The WD40 sleeve is to rigid.

As an addendum the inner insulation is of a shorter length then the outer white insulation

xxxxxxxxxxxxxx sleeve
_____________ White
————— Red
————— Red
_____________ White
xxxxxxxxxxxxxx sleeve

The insulation should be a tight fit inside the sleeve and also be exactly the same length as the sleeve. The worm gear fits in the right side of the diagram as it is a large diameter. The motor spindle fits tightly in the left side of the diagram and is a smaller diameter.

Geoff

[albertofenini]

Dear Geoff, Hello everyone,

we were able to fix the worm gear of one motor, but nothing changed,
We also worked on two more units, with the same result.
They read 100% of the cards, even the cards recorded on a different HP-65 but none of them
can record a card.

More over, this two units have a bizarre behavior,
The first one, when reading shows a strange combination of 000 on the display, but when the card
is extracted it goes back to a normal 0.00, and the card has been read properly.
It can be seen here : https://youtu.be/TrJnuXr1H1o

The other unit, reads fine, but when recording the motor spins endlessly until the card is removed.
Hover, it doesn't record anything.
It can be seen here : https://youtu.be/NugramtopJI

Since four different units behave like this, it has to be a common mistake that we make
when restoring the card reader.

My feeling is that the write enable switch doesn't work on all of them.

How can we override this switch ?
Is there a why to force this condition ?
We have found on the schematics the pin, I believe is WE (write enable) but how can we force
this ? Should we ground this pin ? Or should we put a voltage there ?

We would like to override this just to see if we will be able to write on a card.

Thanks for help !!!

[Geoff Quickfall]

Hi, you say they all read but do not write.

Sorry, but have to ask; Are you using a clipped (read only card)?

If you are using read/write cards (unclipped) then have you checked the write sense spring on the reverse side of the card reader? This spring checks for a clipped section on a read only card.

Geoff

[albertofenini]

Hi Geoff,
thanks for your continuous help and support.
Yes, we are using read/write cards, un-clipped and even from different packages.
I also think that this consistent behavior across four different units, from different years can only be due to something trivial like a missing contact.
The question is, which one is the spring among the four that I have to check and more important
how can I check it given that once I reassemble the card reader the srping are hidden from the view ?
Thanks in advance !!

[Edit I] in all of the units, the springs are clean and the correspondent traces on the keyboard PCB are clean

[Edit II] I have attached a couple of diagramas (@moderator if this is some how inappropriate please remove the attachment)
These are for the CPU board and the card reader PCB.
I imagine that WA and WB stand for Write channel A, Write Channel B, as RA RB stand for Read Channel A, Read Channel B.
There is a GND pin for ground, a couple of Voltages.
There is then a WE pin, is this for Write Enable ?
I assume that if a card is recordable and the card protection switch works properly something must happen at this pin, right ?
What can I possibile expect and measure to see if the switch works ?
Is this pin normally open to ground and closes if the card is writable ?
Because the shape of the card is identical whether the card is protected or not
I assume that a non protected card will give as a first thing an "impulse" because of the presence of the upper left corner moving the switch
and this will the the CPU to enable the WA WB channels.
I'm missing however the scheme of the brass switches, which does what ?

Also in the scheme of the logic board the pin between WE and WB is marked as CRT/P while on the scheme of the card reader the same pin is marked ans N/C (Not Connected ?)

Thanks !!!

   
   

[teenix]

Just in case...

The first switch (I believe) turns the motor on, then as the card progresses into the slot, the RW heads are activated by the second switch. Card write will activate only if the PRGM/RUN switch is in he PRGM position AND the card is not sensed by the third switch to be write protected.

Just a matter of seeing the pattern of switches in the slot as the card comes into contact with the nylon balls. Obviously, the unit will have o be disassembled.

The WE pin looks to be an ACT output and would let the CRC chip know the PRGM/RUN switch position. The logic level of this line should change with the switch position.

cheers

Tony

[albertofenini]

Hi Tony,
look at this picture,
the bottom (usually curved) should be common to all
the inner or shorter (the middle one) is the first activated since is the first encountered by the card and I believe turns on the motor
If this is true, then the upper one (A in my picture) should the one the is triggered by the upper left corner of the card.
What are B and C doing ?
And, how can I check if a recordable card is treated like a record protected ?
Thanks for help !