Disassembly of HP-19C

[Jonas Sandstedt]

Hello, I have acquired an HP19C calculator, and have detected some battery corrosion that needs to be removed. I have removed the five screws and have slightly lifted the top part without removing it, and now wonder how to proceed. I have understood that this calculator is painstaking to disassemble and re-assemble, and hope to get some helpful hints that ease the work and lower the risk that something is broken during the process.
Best regards
Jonas Sandstedt

[Don Williams]

Taking the calc apart is not the challenge. It is the reassembly that is the problem.

If I remember correctly there are two boards and two rows of contact pins between these boards. One row near the top of the boards and one near the bottom. When you have to reassemble the calc you will find it very difficult to determine if those boards are actually aligned. You cannot see all the pins when you attempt to mate the boards.The tips of these pins are easily damaged if things are not perfect. If you damage a pin good luck in finding a replacement. Even if you did have a replacement you then need to solder it in as perfectly as the other pins (which were machine aligned and soldered perfectly} before it will mate with the other board.

HP must have had some kind of assembly jig or tool on the production line to aid in the this task.

The first time I reassembled one it went as smooth as glass. Just luck though. That never happened again. All subsequent attempts resulted in broken pins.

Your experience may be different. Good luck.

[Jonas Sandstedt]

Thanks for the tips. Except the two rows of golden spring connectors there is a flat cable at the bottom, how is that one best handled? best regards Jonas

[Jonas Sandstedt]

I enclose a picture of the interior.

[Ren]

(03-31-2021 02:06 PM)Don Williams Wrote:  Taking the calc apart is not the challenge. It is the reassembly that is the problem.

Yes, it can be taken apart quite easily with a 8 pound sledge hammer!

Celebrating April 1st!

[Don Williams]

(04-01-2021 06:24 AM)Jonas Sandstedt Wrote:  Thanks for the tips. Except the two rows of golden spring connectors there is a flat cable at the bottom, how is that one best handled? best regards Jonas

That flex ribbon cable goes between printer board and the print head. Leave it attached between the board and the printer assembly. There appears to be a lot of corrosion on the main board. Time for a vinegar bath. I would not expect to accomplish that without separating the boards and doing them individually. Make it an ultrasonic vinegar bath if you have a ultrasonic jewelry cleaning device available.

Warning the reassembly is very difficult. The tips on those board connect pins are very fragile and will break or bend easily. There are actually three different board connection rows which you must magically align "in the blind" to get it all back together. Not a task for the faint of heart.

Good luck and keep us posted.

[Jonas Sandstedt]

Thanks for the advice, I will start working on the calculator when I have sufficent free time. I will keep you posted of what happens.
best regards Jonas

[Jonas Sandstedt]

Hello,
I take this in very small steps and have now separated the two housing parts as shown in the photos. In the bottom housing part is the printer and the power supply/printer driver board, and in the top housing part is the logic board that is attached to the key PCB via white ribbon cable.
Now to the questions.
I assume that the printer assembly has to be removed when removing the power supply/printer driver board such that these parts are held together via the flex cable that runs to the printhead (the printer assembly seems to be fastened with three relatively large screws). Is this correct?
But how is the power supply/printer driver board removed? The battery contact tabs protrude into the battery casing that is comprised in the housing and prevent an upward motion, and the board also seems to be prevented from lateral movement by the battery casing (from left to right in the two photos with the power supply/printer driver board).
Finally, can the logic board be removed from the key PCB by disconnecting the white ribbon cable?
I hope for helpful hints in this matter.
Best regards
Jonas

[Don Williams]

You can remove the three brass screws that secure the printer. Then lift the board and gently slide the board down until the battery tabs are clear of the slots in the case. Be careful with the printer flex cable. I never attempted to disconnect that cable.

The ribbon cable between the CPU board and keyboard is soldered. There is no way to separate the boards without unsoldering the cable.

[Jonas Sandstedt]

Thanks,
There is, however, a problem with two black plastic PCB holders protruding from the housing (being a part of the housing) at the side of the PCB where the battery connectors extend into the battery casing (one holder shown in the PCB's corner in the photo). The PCB requires a bend to go free of the holders, and at the other the end the PCB also requires a bend to go free of the battery casing, the PCB then being bent in a "banana shape" that seems quite risky. Since the PCB is removable, I must have overlooked something.
I look forward to more helpful hits.
Best regards
Jonas

[Jonas Sandstedt]

OK, it worked - but i disliked the procedure. How could this calculator have been be produced in a large scale?
I will come back with more information - and possibly more questions.
best regards
Jonas

[Don Williams]

(04-05-2021 04:40 PM)Jonas Sandstedt Wrote:  OK, it worked - but i disliked the procedure. How could this calculator have been be produced in a large scale?
I will come back with more information - and possibly more questions.
best regards
Jonas

How the assembly line worked is a mystery to me. I spent hours trying to discern the "magic" of re-assembling the top and bottom halves. I never figured it out. There must have been some tool or jig aid that was used on the production line. But I am not as mechanically inclined as some other individuals. Maybe you will discover the clever method they used. I didn't.

The corrosion on the bottom board is probably causing the fault. It needs to be thoroughly cleaned up. Check for circuit traces that may have been damaged by the corrosion.

I would attempt to keep the printer attached, but try to keep it away from the board cleaning process. Easy to say, but difficult to accomplish I realize.

Keep us posted on your progress.

[Jonas Sandstedt]

Yes indeed a mystery!
I remembered that Ignazio Cara taught me how to remove the flex cable from the HP-97 printer sub-assembly board by using a small tool several years ago. I re-used this knowledge here and it worked! I just cut a small plastic sheet that is relatively stiff and insert it on top of the flex cable such that the pressure is relieved of the flex cable conductors and then pull it all out of the connector.
When re-inserting, the tool should be positioned in the fold and then the flex cable is pushed back into place.
Ignazio used the metal cover of a floppy disc that was cut into shape, but a relatively stiff plastic sheet also works – see the picture.
I come back when something new happens.
best regards
Jonas

[Jonas Sandstedt]

Hello,
Now the bath is over, and I think this is as far as I will come - please let me know if you think otherwise. I have checked the printhead resistance and it is about 14 ohms for every printhead resistor, so I think it is OK. The plan is to check the transistors for short-circuit and the traces for discontinuity and then re-solder the PCB where needed. Thanks to PANAMATIK, useful tips have been provided for substitute components if needed.
Best regards Jonas

[[kby]]

Disassembly:
You have to (carefully) bend the power supply board. I will also suggest an alternative, but I haven’t tried it and drying may be more difficult.

The standard way: pull the bottom edge straight up, enough to just clear the two plastic pins. The thin area on the left side is quite long, so the amount of flex it can take is actually fairly high since it is distributed over a relatively long length and it isn’t that rigid because it isn’t very wide, either. At the same time you will have to lift the printer side enough to clear the battery compartment once both ends are raised, gently work the battery side up over the pins and towards the bottom edge. Your biggest danger her is actally breaking the plastic pin that holds the board in place. It can be repaired adequately if you breaki it, but it is much better not to do so. You will have to bend a bit more or less and maybe help the battery contacts come free of the case as they are not flat. On the printer side, once the board has been backed out a little, you’ll be able to have the printer side on to pf the battery compartment where you don’t need to monitor it. Sometimes you can work one corner/side out a bit more than the other so at least one corner will sit on top of the battery compartment; if you do it will be easier.

Sometimes at least one of the plastic pins are broken or weakened already. In that case you’ll have to glue it back. It’s actually easier to replace the board first anyplace the pin in place (if the bottom is not square, you should insert it in the hole in the PS board from the bottom (so you don’t break any part that is sticking out, after figuring the proper orientation. I then usually orient it after I’m sure things are working and glue it back by flowing cyanoacrulate UNDER the so board and holding in place. It’s good enough to keep the board locked in place from the pressure the batteries apply, but it’s likely to break again if you disassemble again.

Reassembly:
This first part of the reassembly is easier than the disassembly; just reverse what you did and things should pop back. Again, be careful not to put too much pressure on the plastic pins. See note above if you had to repair the pins. I find it’s the one on in the corner where the divide key is that’s usually damaged. Note that if there is a slight curvature to the board after all of this, that’s ok.

To get the case back together, first make sure the flex cable between the top circuit board and the keyboard is attached in all the places it should be to the keyboard. Then swing it downward (without detaching it from the keyboard so the bottom of the calculator (on the table in front of you in operating position), the to- circuit board will look kind of like a hand fan from the side with the keyboard on top and the front edge of the calculator as the “pivot” point.” In this configuration it isn’t too difficult to line up the two sets of pins (but you still have to be careful and it is easier if you have small hands). It’s usually easiest to connect the pins close to the display first, then worry about the ones along the side. If necesssry, you can remove the display to do that first fit together and then put it back in (but don’t forget!). At this point all of the electrical connections should be made. Remove your hand from between the keyboard and the cpu board (where it needed to be to do the above and snap the bottom edge of the calculator together. It will take a non-zero amount oof force to do this and there should be a fairly loud “snap” as it joins. You can then deal with the top edge of the calculator mostly by feel.

You may want to remove the printer cable from the power supply board; you may have to as it can slso harbor corrosion. Unfortunately it is a terrible design: The cable has a crease in it and the contacts in the connector are on the short tail end of the cable (so after the crease, electrically speaking. The cable plastic itself is reasonably robust but the circuit traces are not so strong and sometimes break over the crease. To remove the cable, look for some of teenix’ posts on the -97 printer. You need a thin shim (I use a piece of aluminum) you can insert into the connector between the connector and the folded edge (“short”) end of the cable. The shim needs to be the full width of the cable. You then gently pull out both together. You usually don’t need the shim to reinsert the cable.

You could try to just soak the ps board without removing it, but you won’t be able to inspect the underside of the board. It may also be hard to thoroughly dry the board. Make sure the printer itself doesn’t get wet.

[Jonas Sandstedt]

Thanks for this exhaustive description, I will continue to work on the calculator and report my hopeful progress.
Best regards
Jonas

[Jonas Sandstedt]

Hello,
I thought I should remove some old solder that seems oxidized and re-solder, buth the old solder just won't melt - even if I add som new solder. Should I just leave it as it is? On the picture are the solder joints of the nework 1810-0236 which provide an example of such old solder joints.
Best regards
Jonas

[teenix]

Assuming the soldering iron is hot enough and the tip size is adequate to supply the required heat, one way to get it to melt is to have some melted solder on the tip so heat transfer can take place and hold it (not too long) on the place you want to repair. You can add a bit more solder when the tip is in contact with the soldered PCB surface to help transfer the heat.

If it does melt and you have too much solder remaining on the joint you can remove some with solder wick.

If the solder "blobs" on the joint, it may be an indication of contamination (oxidation or dirt) in the old solder or component leads, or not enough heat from the soldering iron.

Avoid too much heat though, or you risk damaging the component, PCB traces or through hole vias.

Bear in mind also, that some components (IC's) are sensitive to static discharge so care should be taken to avoid damage when handling PCBs or during repairs.

cheers

Tony

[Jonas Sandstedt]

Hello and thanks for all tips. The new solder indeed "blobs" on the joint, and the joint won't melt to be sucked away bý either vacuum pump or desolder wick. The solder joints are most surely oxidized, having been rinsed from battery leakage with vinegar acid. Is there any other way to melt these joints without destroying PCB and components? Perhaps the safest way to continue is to try to re-assemble the calcualtor and see how it works before working more on the PCB.
best regards, Jonas

[[kby]]

Are you using a temperature-controlled soldering iron and interchangeable tips? While it’s certainly easy to just blast the board with too much heat, often time it’s the length of time you heat as much as the temperature that causes damage. This can be more problematic of too low a power soldering iron thst can’t transfer heat in fast enough to exceed the transfer rate out: the solder never gets to the melting point, but the whole surrounding area is maintained at a high enough temp to lift traces or cause other damage.

Make sure your tip is clean. Tin the tip well (as Teenix suggested in different words). You might try paste rosin flux and/or a larger tip (like chisel rather than point)being more careful about the time), and using the solder wick.

I’ve gone to using a portable desoldering tool (Hakka Frp-301), but thst is a fairly expensive solution (like a few hundred $$$).-kby

(04-12-2021 04:24 PM)Jonas Sandstedt Wrote:  Hello and thanks for all tips. The new solder indeed "blobs" on the joint, and the joint won't melt to be sucked away bý either vacuum pump or desolder wick. The solder joints are most surely oxidized, having been rinsed from battery leakage with vinegar acid. Is there any other way to melt these joints without destroying PCB and components? Perhaps the safest way to continue is to try to re-assemble the calcualtor and see how it works before working more on the PCB.
best regards, Jonas