HP15C- LE bad button



#17

I purchased 6 HP15C-LE calculators and one of them had a "4" key that would not register when pressed normally. It clicked nicely but would take a little extra pressure to register. I took it apart to find the cause.

The back is easy to remove with 6 screws. Four screws under the feet and 2 in the battery compartment. There is a tiny spring that connects the metal plates and is probably for ESD. The PCB is heat staked in place. There's a little slack under the stake heads when pressing the PCB towards the front panel.

I thought loose heat stakes might be the problem. I pressed the PCB and it did not correct the problem key. I used a large sharp drill bit to trim off the heat stakes flush with the PCB. I removed the board from the front panel. There is a rubber sheet between the plastic keys and the PCB dome switch assembly. My guess is the sheet makes the keys feel good and protects the switch assembly.

So now I have it apart and can press directly on the switch assembly. The "4" key is more reliable but still fails about 1/2 of the time. On a failed press a little extra pressure makes it register. If I lay the rubber sheet over the switch assembly and press the key the reliability goes down.

Next I peeled back the plastic that holds the dome springs in place. The dome springs lift up with the plastic sheet so I can see the PCB contacts. There are small dents in the contact pads that look like the board is tested with spring probes or a flying probe.

Near the dent on the center of the pad for the "4" key there was a very small piece of clear to white plastic. Just a flake, but I suspect it prevented the dome spring from contacting the center pad since it is positioned almost dead center. I remove the plastic flake and put back the plastic cover and the problem is gone. I have 100% reliable presses directly against the plastic and through the rubber sheet.

This seems to be the fix for my calculator however the problem could be wide spread. So HP, check for contamination under the dome switches.


#18

Great information, thank you! Now, if only HP hadn't used heat-stakes but screws, we all could fix our problems without ending with a loose calculator... :(

Cristian


#19

This would explain why some units have keys that are extremely hard to press and don't click at all: contamination with some plastic residue that blocks the path of the key.

Are there tiny holes below the domes so that we can clean the keyboard from the back without cutting the heat stakes? There was such a mention for some other calculator in the past.

#20

Great analysis!

May I ask you, how you re-applied the heat-stakes after you "cut" them off? As I'm having the same issue as you, I'm thinking of self-repairing the calculator myself as well...


#21

I plan on using thread forming screws or small machine screws to reassemble the calculator.

Keep in mind this repair is a sample size of one. Other calculators could have other issues.


#22

You are correct, but if I can reassemble the calculator afterwards, at least it should not make the problem worse :-)

I have not seen the internals of the 15c LE yet, but what you think, would it also be possible to just apply some glue on the cut heat-stakes? E.g. Some good two component glue or something?

#23

Would you keep us posted about your success in reassembling the calculator?

Thanks in advance.

#24

Djj, now that you have seen and solved the problem do you think it would have been possible to shake the calculator hard enough to move the small piece of plastic out of the way?


John


#25

How about putting a strong vacuum cleaner hose to the key and create enough air movement inside to move the flakes out of place?

#26

It would be terrible if that works. How many 'flakes' must there be to have a chance of one of them in a neuralgic position each 10th calculator? :-/


#27

When I purchased my 50g a few years ago I couldn't believe the amount of debris (too large to be described as dust) between the LCD and plastic cover. It was so bad I had to pull the plastic cover off and clean it out. The pieces were large enough to block contact between the snap domes and PCB.

#28

I doubt shaking the calculator could have dislodged the plastic. Vacuum won't work because the plastic film that holds the dome springs in place seals the assembly.
I don't know if particles under the dome switches is the widespread problem. HP will have to figure that out.

#29

an excellent analysis.

i worked on a product several years ago that used a very similar keyboard arrangement. there was a relatively small number of units produced per year (amounting to around 10 per day) selling to the customer for about $5k each. hence assembly was quite 'manual' orientated with many QC checks.

the white plastic sheet has adhesive applied with a stencil. there is a dot of adhesive (perhaps 3/16th of an inch across) that holds each of the domes in place, as well as adhesive in all areas well clear of the domes. in our case, a 'path' was also left clear of adhesive such that no closed air pockets were formed around keys.

as supplied to us, we received the white plastic sheet with the domes pre-applied. a waxed-paper (light blue coloured) backing sheet was also there to cover (just) the areas with exposed adhesive. the application process in the factory consisted of peeling off the blue backing sheet - being careful to not disturb the domes - and using an alignment jig to apply the white plastic sheet + attached domes to the keyboard PCB.

the possible failure modes were:

1. a dome could fall off, rendering a key completely non-functional



2. a dome could be moved out of position slightly, producing a key that would often operate ok but wouldn't 'click' due to the button pressing it other than in the dead centre



3. a stray scrap of the white plastic sheet, blue backing sheet, or a piece of loose adhesive, could get between dome and PCB during assembly, resulting in a key that would click but either not register, or only register if pressed hard.

the above three cases to seem to mirror closely the problems seen with the 12C/15C-LE.

#30

The silver lining in the 15c LE quality control cloud is that - they are repairable! Who would have expected that today?

#31

The 15C LE problems are reminding me of the 49g+ keyboard problems when they first came out in 2003.

#32

Excellent work! Now it seams that nothing can be done to fix a calculator without taking it apart and thus avoiding the (non-existent?) warranty?
Cheers,

On the other hand, HP should be aware by now about the problem; they must have have done and found the same as you did.

HP, any comments?

Edited: 9 Oct 2011, 5:02 p.m.


#33

Sorry, but HP is busy developing the new ePad.

HP Support Team.

#34

Djj - Got Photos of the teardown???


#35

I bought two types of screws to reassemble my HP15C-LE. The first was a small thread forming screw #00 x 3/16" and the second was a 0-80 machine screw. I was hoping the thread forming screw would fit. I tried to order the same size as the small screw from the battery compartment since it seemed to be a good fit. I didn't get the size right and ended up with screws too small. My past experience with thread forming screws is that they can be difficult to size correctly and the fit is important to prevent cracking the plastic. There is not much excess plastic to thread into or deform in the thin plastic posts.

The machine screw size was a good fit so I went forward with it even though machine screws are not typically threaded into ABS plastic. I used a bottoming tap to form threads in the plastic posts. Then I just reassembled the calculator. The LCD is a little fragile so I placed it in position then set the PCB in place. The used small plastic washers under the 0-80 machine screws. I ran the self tests and everything is good and the keyboard feels great. I know I completely voided the warranty but the problem was bugging me.

I bought the machine screws, washers, and tap from McMastercarr.com

The PN for the tap is 2522A811, the washers are 90295A313, and the 0-80 phillips machine screws are 91772A054. the total was 21.50 + shipping.


I put a few pictures on photobucket at:

http://s1093.photobucket.com/albums/i427/djj2011/


#36

Great job! Thanks a lot.

#37

Nicely executed and documented, Thanks for sharing!

#38

Excellent! Thanks for letting us know how this worked out.

I suspect that this will be useful to many of us, you should post this as an article here so that it doens't get lost.


Edited: 14 Oct 2011, 12:20 p.m.

#39

Great work, Djj!

As Katie suggested, please make an article of your project! I think that many of us will be doing the same mod to fix "bad keys" on the 15C LE.


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP Prime: the drama of the Apps button, shutdown, formatting, nothing solves Davi Ribeiro de Oliveira 0 349 11-17-2013, 07:23 AM
Last Post: Davi Ribeiro de Oliveira
  HP Prime: the drama of the Apps button, shutdown, formatting, nothing solves Davi Ribeiro de Oliveira 1 435 11-16-2013, 11:26 PM
Last Post: Michael de Estrada
  ENG button (like Casio calculators) on HP Prime uklo 3 654 11-04-2013, 09:45 PM
Last Post: LHH
  How to create a touch button [HP Prime] uklo 1 442 11-02-2013, 02:03 PM
Last Post: Marcus von Cube, Germany
  1984 HP15C rattles Footloose (Illinois) 4 639 10-15-2013, 09:43 PM
Last Post: BobVA
  HP Prime touch button SanS 3 524 10-10-2013, 12:24 PM
Last Post: Han
  HP15c continued fraction for Ln(Gamma) Tom Grydeland 0 349 09-30-2013, 05:48 AM
Last Post: Tom Grydeland
  Bad Flash in HP Prime Han 11 1,291 09-27-2013, 12:38 PM
Last Post: Han
  OT - A lucky find - Busicom Handy-LE LE-120A Cristian Arezzini 2 566 09-26-2013, 04:43 AM
Last Post: Cristian Arezzini
  Cheap hp32sii in bad condition on sale on Taobao Waon Shinyoe (China) 1 436 03-01-2013, 08:35 PM
Last Post: Raymond Del Tondo

Forum Jump: