HP 82104A card reader, fork-shaped brass tongues, REPAIR PROBLEM

Hello HP-calc-communuity,

I hope the collected experience of this outstanding forum can give me some information to solve a serious problem:

I recently repaired the "gummi wheel issue" of my HP 82104A card reader, using the "O-ring method" described in several documents here. That worked very fine, the movement of the magnetic card through the slit is now smooth and regular.

BUT: While opening and disassembling the card reader's case the two fork-shaped brass tongues and their respective small(in fact very small!) plastic counterparts FELL OUT.
I afterwards found out that the uppermost (i.e. far from display) tongue serves as a simple SWITCH to turn the motor on and off while inserting and propulsing the card through the reader.
But what purpose is the second, lower (i.e. nearer to the display) tongue? What movement must this part perform while a card is read, and when? The small plastic counterpart of this tongue is different from the other one, it is even smaller, smoother and has no distinct "nose" or cam. I don't get this tongue to bend and move upward while a card is read in the reader, in no stage of the process... The other tongue bends up very well when a card is inserted and always closes the (funny construction...) electric traces on the printed circuit covering the whole installation, and it opens again when the work is done.
BUT the other strip seems to have nearly no effect. It does NOT move up in any stage of the card movement process. This is why I always get a "MALFUNCTION"-message. I think the signal from the second circuit is responsible for this error.

I hope I could describe my problem precise enough for You to know what my issue is. So I can only beg You to PLEASE help me with any information You can give me! What can I do, what must I observe, what method of repairing and testing do You propose? If You need more information I could of course publish some photos with additional text showing the parts I mean.

Thank You very much in advance for any support
and best wishes from Bamberg, Germany!

Markus Sanke


Hello Markus,
the second tongue enables the magnetic head to read/write the card. If this circuit is not closed by the contact of the tongue with the upper contact, you will always have an error message.
It's somewhat strange and unusual that the two contacts and two cams have gone amiss. After this maybe you have made an error in reassembling the contact and one cam.
If you can, post some image of the actual assembly so we could see if there is any error in the reassembly.




Hello Ignazio,

thank You very much for Your quick answer! I already thought that the second, not working switch is for starting and stopping the reading process or something like that.
I made some deatiled photos of the module as it is now, and I wrote some additional text into some of the pictures.

Here are the links, please take a moment and look:

The assembled Module The lowest picture shows the change when a card is gently put in the slit.

The assembled Module (w/text)

The opened Module

The opened Module (w/text)

The Printed Circuit, loose (w/text)

The ends of the fork shaped brass tongues

The ends of the fork shaped brass tongues (w/text)

The two cams that should move when the card is pulled through the reader, but only the upper one works (w/text)

Testing functionality

Testing functionality (w/text)

It would be of graet help for me if You could post me any idea. Maybe You see one part that is broken, not on the right place or otherwise misused. Thanks in advance for Your answer!

Best wishes and warmest greetings to Italy (a country I love so much since I am a professional archaeologist...)


Edited: 8 Apr 2010, 11:52 a.m.


Having only seen the inside of the HP 97 reader I can't be sure, but it looks like you are missing the tiny nylon balls that are supposed to be in the cavities to the right.


Hi Markus,

I've taken a look at your pictures and think I may shed some light with your Card Reader issue.

First, to answer your question regarding the glass tipped part in the circuit: it is a phototransistor, it's used by the Write Protect detection circuit. Therefore it has nothing to do with the reading trouble.

The shape of the actuators (the plastic balancing pieces) is indeed diferent as they should. The smaller one is for the head activation circuit. While the larger is for the motor driver circuit. They both look good in your pictures.

The contact blades ("tongues") don't look that good though... the faulty one shows a slight bending which, in my opinion, may prevent it to work properly. Both blades are identical, so you can swap them to test if the actuators are working.

Should this was the case, my advice at this point is that you try to gently unbend it in order to turn it back to the shape of the working one. You may be able to to this just with your finger tips.

Hope this helps.

Best wishes from the Caribbean Sea.


Edited: 9 Apr 2010, 12:04 a.m.


Hello Markus,
Very good pictures, and the issue is clear now.
The Diego's analysis it's everything you need.
Regards and good luck.


Hi Diego, are you fine? Regards from Sardinia's island.


Hello Ed.
The card reader of the 41C don't have the same system of 97 and 65/67. In place of the spheres there are two small cams, operated by the card when it's inserted.




Hello to You all (wow, around the whole globe, as I see...)!

Thank You Ignazio, Diego and Ed for Your answers; I'm happy to hear that my parts are at least not too damaged.

Two short informations would help me to finally solve my problem:

#1 While moving through the slit, what is the exact time-scheme the two contact blades should operate ("up" and "down" in relation to time or card-way). I description or diagram would help!

#2 Could You describe the correct form of the blades? Must they be absolutely flat except for the two side-parts bending upward? Or do they have a kind of "pre-bending"?

I will try the changing of the two blades again, but I probably already did it without any improvement. I have the impression that the small actuator for the defective blade is not moved/rotated enough by the card. Maybe this is also due to a wrong pressure or distance of the two module halves?

Thanks again! Maybe one of You finds time to answer again!
(I am very impressed of the kind helpfulness in this forum!!!!)

Best wishes



Hi Markus, Ignazio, all...

(Ignazio, thanks for asking, I'm fine and hope you're also doing well. Currently in the Dominican Republic taking some weeks of rest... well, more or less... ;-)

To your questions Markus:

The first actuator to operate is the (so called) motor driver. In fact this is properly a Card Insertion detect switch by means of which the internal circuit is waked up and, if necessary, the 41 CPU will also be waked up. It is the CPU that in the end instructs the Card Reader circuit to turn the motor on.

Once the card is running through its path, the head actuator should make contact at the point in which the card reaches the head tip.

Here is an image of how the contact blade should look like. Is it obviously not to scale, and drawn from my head so please take it with a grain of salt... :-)

The long blade should be *very slightly* (less than pictured) bent upwards. Just enough to settle in place with very little pressure on its ends. This bending should look convexe along the blade. In your pictures a small concave bending shows up just beneath the actuator hinge and I think this must be corrected.

Note that this is not the "factory" shape, but I found it easier to produce that small banding than get it back to full straight when a repair is needed.

The actuator could in fact has gone so badly worn out that it can no longer tilt enough to activate the switch. This looks quite improbable to me though. So let's try the easier and if it doesn't fix the trouble will look further.

Hope you can get your Card Reader back to work soon.


Edited: 9 Apr 2010, 11:18 a.m.


Hi Markus,
this is my best picture of a switch contact profile:





Again, a (good) picture is worth a thousand words... :-))

In this case, showing the slightly convexe bending I tried to draw and describe.

Thank you Ignazio!


Edited: 10 Apr 2010, 2:26 p.m.


Thank You again Ignazio and Diego! I spent another two nights in attempting to repair my reader with Your friendly help. The effect was as follows:

I now got the two switch contacts into the shape of Diego's drawing and Ignazio's photo. They both had to be reshaped, using slight pressure of the fingers and finally with the help of tweezers.

Now the situation is this: When the upper board, bearing the brass contacts and their actuators, is isolated (i.e. NOT mounted onto the lower half of the slit assembly), BOTH contacts now move upward when touched by a slightly moving finger. Only the sligtest pressure is sufficient. The movement of the contact blades is satisfactory, the center part moves upward for more then 1 mm.

BUT if I re-assemble both halves of the mechanism, again only the first switch works. Te second one (that with the smaller actuator) is kind of blocked or disabled, no real movement occurs when a card moves along the actuator.

I experimented with different distances of the two halves, by screwing up and down the screws. Last night I got the card reader working for about one hour, demonstrating that principially everything is still all right (Writing and re-reading a program over two magnetic cards). But after a few tests the "Malfunction"-message appeared again, now mixed with "Card Error"- and "Checksum Error"-messages.

I think the problem now lyes more in the way of re-assembling the two halves. The distance of both parts seems to be a crucial issue. If too tight, the motor stops when a card is inserted. If too wide, no regular movement occurs and the above error messages appear.

May be one of You could again check my description and offer me some helpful ideas! If photos are neeed, I can again publish some!

Best wishes to all of You


Edited: 11 Apr 2010, 9:48 a.m.


Hi Markus,
Let us see other photo, thanks.



I agree. At this point some pics, similar to the one posted by Ignazio, from your two contact blades will help a lot.

It is quite unlike that the spacing between the PCB and the reader's plastic block require any adjustment at all.



Hello again,

I made a short video showing the upper half of the assembly with the two contact blades. When I slightly glide on the lower side with a finger(simulating a magnetic card), the first contact blade always moves up for a good amount. The second one often, but not always moves the same way. The axis-part of the second blade is very loose, sometimes allowing to "jump" up and down (see video, from 0:36 on).

Link to Video

If I reassemble the two halves, only the first contact works; the second one does not move up and close the switch.

I don't believe that the form of the two brass stripes is responsible for the error. If I change them, again contact #1 is good, #2 is bad. I show You the actual form of the two blades in detail in this photo; I treid to form them as near as possible to the photo You posted here a few days ago:

Link to Photo

I am short before giving up because I don't know what else can be done to repair the card reader... This is especially frustrating because I know that "principially" the machine works, which I could see a few days ago - but only for an hour or so.

Any hints and helping suggestions are VERY welcome!

Best wishes


Edited: 15 Apr 2010, 12:55 p.m.


Hi again Markus,

I'm affraid I have to disagree, the shape of the blades IS the cause of the malfunction. This is clear after watching at your last picture.

Just in case you have not the tools or the experience to straighen them back to the suggested shape, and before you thru away a "basically" operative Card Reader. You may want to ask for service.

I would happily do it for you, but I'm currently far from home and will still be away for several weeks.

My advice is to give it another try and do your best in order to remove any irregularities in the blades, mainly these small concave bending curves that appear in your picture.

Best from the Carribean Sea and don't give up! :-)


Possibly Related Threads…
Thread Author Replies Views Last Post
  Reasonable price for a 41C, Quad Mem, Time, Navigation and Card reader module? Eelco Rouw 10 4,766 12-12-2013, 01:52 AM
Last Post: Eelco Rouw
  hp-67 card reader problem. Marcel Samek 5 3,233 12-09-2013, 09:09 PM
Last Post: Paul Berger (Canada)
  Strange HP 25 Problem (Repair) Onur Ilkorur 5 2,885 12-06-2013, 05:13 PM
Last Post: Onur Ilkorur
  HP-97 card reader pinch roller axle davorin 2 1,629 09-15-2013, 08:47 AM
Last Post: aj04062
  HP-97 Card Reader davorin 3 1,910 09-13-2013, 04:21 PM
Last Post: Stephan Matthys
  HP-67 Card Reader Repair Question aj04062 1 1,517 08-08-2013, 11:15 AM
Last Post: Randy
  Hp-67 card reader: tantalum capacitor identification Ignacio S├ínchez 8 3,270 06-02-2013, 07:26 PM
Last Post: aj04062
  Testing card reader motor for 65 Michae Altmann 2 1,642 03-17-2013, 09:24 PM
Last Post: Randy
  Repaired HP-97 card reader doesn't show any signs of life jbssm 6 2,673 02-17-2013, 03:15 AM
Last Post: Alberto Fenini
  HP 65 card reader problem H Fischer 3 1,546 11-24-2012, 12:11 PM
Last Post: H Fischer

Forum Jump: