HP25 keypad problems

[calc-calcs]

I have a HP25 that had the typical internal corrosion problems, but was easy to clean up internally at the usual spots and is now fully operational. Only remaining problem is with some of the keys, specially the 2 and 5, even tough they still have the right feel, but seem to have a lot of bounce and when pressed once usually cause multiple entries, or none at all. This behavior goes away after giving the key a bit of a workout and in general after that they start behaving normally again. But if left unused for more than a day or two the bouncy behavior returns. My guess is that there might still be some active corrosion inside the contacts and/or dirt getting in the way.

I know its not an easy task to disassemble the keyboard in this series, and even less to put it back together. So far I tried to get a drop or two of IPA 91% between the key and the front, but it only seemed to help marginally if anything at all. My guess there is a build-in splash barrier preventing the alcohol to even get close to the actual contacts.

Has anybody fixed a keypad problem like this one in any hp2x without necessarily having to disassemble it for the contact cleaning? I'd be willing to try anything reasonable before embarking on taking it apart, although guess the process might have a favorable outcome as long as I can be put back together afterwards and still remain reliable.

Thank for any ideas.

[Dave Britten]

(06-23-2020 03:25 PM)calc-calcs Wrote:  I have a HP25 that had the typical internal corrosion problems, but was easy to clean up internally at the usual spots and is now fully operational. Only remaining problem is with some of the keys, specially the 2 and 5, even tough they still have the right feel, but seem to have a lot of bounce and when pressed once usually cause multiple entries, or none at all. This behavior goes away after giving the key a bit of a workout and in general after that they start behaving normally again. But if left unused for more than a day or two the bouncy behavior returns. My guess is that there might still be some active corrosion inside the contacts and/or dirt getting in the way.

I know its not an easy task to disassemble the keyboard in this series, and even less to put it back together. So far I tried to get a drop or two of IPA 91% between the key and the front, but it only seemed to help marginally if anything at all. My guess there is a build-in splash barrier preventing the alcohol to even get close to the actual contacts.

Has anybody fixed a keypad problem like this one in any hp2x without necessarily having to disassemble it for the contact cleaning? I'd be willing to try anything reasonable before embarking on taking it apart, although guess the process might have a favorable outcome as long as I can be put back together afterwards and still remain reliable.

Thank for any ideas.

You shouldn't have to do anything extreme like fully disassembling the keyboard, but you'll need to open up the calculator and remove the circuit board, which is very easy to do on Woodstocks (there are plenty of pictures and instructions floating around if you need them).

Get some Deoxit D100L:

https://www.amazon.com/CAIG-LABORATORIES...B0000YH6F8

Put a very small drop of this into the tiny hole in the PCB behind the affected keys and work it in by exercising the key. I've used this technique to fix bouncy/hesitant keys on my 25 and a couple 41s, and it works great. Note that the key may feel and sound a bit different for the first day or two, until the Deoxit dries.

[Massimo Gnerucci]

Yes, as Dave said.
You may wish to have a look at this thread!

[GreyUser]

Use the de-oxit d100 and apply more than once if needed. If, after two applications, it still isn’t right, you can fashion a miniature wire brush and clean the contacts with it as described in post #2 of this very old thread in the archives. Substitute D100 for the isopropyl. This is the last resort before busting it open, something you should not rush into, it will never be the same.

https://www.hpmuseum.org/cgi-sys/cgiwrap...5511#55511

[calc-calcs]

Thank you everyone for the very good information. Will reopen the calculator sometime this week as time permits and hope my keypad has the small holes behind the keys to apply deoxit red. Fortunately I currently have a small tube of it as do I use it with some frequency on switch contacts.

Will report back with hopefully positive results. Thanks again.

[GreyUser]

(06-24-2020 06:18 AM)calc-calcs Wrote:  Thank you everyone for the very good information. Will reopen the calculator sometime this week as time permits and hope my keypad has the small holes behind the keys to apply deoxit red. Fortunately I currently have a small tube of it as do I use it with some frequency on switch contacts.

Will report back with hopefully positive results. Thanks again.

The chances of having a sealed keyboard is extremely low. They were used for a short period of time on new 20 series units, late in the Woodstock’s life cycle and AFAIK, it was only on the 29C.

I don’t know if they were ever used in the repair department. I suspect so but the number of units out there in the wild was probably very small. My own 25C went back to Corvallis more than once, getting a new keyboard in about 1983 and it was the original design with metal strips.

The sealed Woodstock keyboard was based on the new design for the 30 series units that did away with metal strips, replacing them with conductive plastic domes. In a rush to get to that new product to the market quickly, they did not wait for the usual accelerated life cycle testing results. With product having been shipping for some time, the results of those tests came in and HP knew they had a problem. The keyboards were failing much sooner and in greater numbers than was expected. This was due to the design having one half of the connections passing through the domes in the form of a flexible, conductive trace. These traces would crack with the flexing of normal use and break the connections to the keyboard scanner.

The result was a redesign for the Spice series with the round metal domes that were the basis of the HP41 keyboard. I don’t the history as to which was first, was the 41 being prototyped and the round domes were already in use or did the redesign result get recycled for use in the 41.

I think there is an article in archives somewhere from an HP employee that related the story of the keyboard design failure in their own words. Hopefully someone will remember it and can provide a link, it was an interesting read.

[calc-calcs]

Previous picture that I found taken some months ago during initial disassembly. So I do indeed have a keyboard with holes.

[calc-calcs]

So since it was only 1:30am decided to open her up, and after putting Deoxit in the 4 and 5 key holes it almost instantly cured the problem. The 5 key required a second shot, but is since working reliably. Then I started testing all the other numeric keys and found that the 2 was also quite unresponsive at times, and 3 was a bit sluggish as well requiring a bit more firm press, and the divide key was a bit bouncy as well. But all now apparently cured. Also discovered that the minus was sometimes also acting two (or more) times on a single press, so that one got its D100 shot as well. Then in the end as a matter of course did all the lower half white keys, as it appeared that they might have been affected in varying degree by the former battery corrosion issue. Will now let her sit overnight and test again tomorrow (later today) and hopefully all will still be good.

But if otherwise just in case left the two feet out to spare them any additional undue stress if I have to reopen for another squirt. These feet are really gummy, and have lost their original consistency, so I try to treat them gently, and had to put a piece of magic tape on their outer face to avoid abrasion and stickiness. Its the only hp2x in my collection that has this problem, all the others are still in perfect shape. Yes I know there is a solution for the gummy feet by cutting up a piece of rubber eraser to match their shape, but so far it has not really been necessary.

Thanks again to everyone for the key restoration tip with Deoxit. BTW need to order more, less than 0.5ml left.

Quick update: after about 45 min the number two key still requires a very slightly firmer pressure than the other keys in order to work, but is not bouncing. With normal pressure it clicks alright but no action. Not sure this might be caused by the Deoxit not having fully dried yet, some dirt still in the way, or perhaps the metal dome is just worn. Will check again later during the day.

[calc-calcs]

Another quick update on the number 2 key situation. When I push the key on the lower edge with normal pressure it works perfectly every time, but if I push it only on the upper edge it clicks but no action on most tries unless I apply slightly more pressure and then it also works every time. If I push squarely onto the face of the key with normal pressure it works about 50% of the time. Any ideas?

[Steve Simpkin]

(06-25-2020 06:33 AM)calc-calcs Wrote:  Another quick update on the number 2 key situation. When I push the key on the lower edge with normal pressure it works perfectly every time, but if I push it only on the upper edge it clicks but no action on most tries unless I apply slightly more pressure and then it also works every time. If I push squarely onto the face of the key with normal pressure it works about 50% of the time. Any ideas?

That key might need the miniature wire brush treatment mentioned in post #4.

[calc-calcs]

After letting it sit for a number of hours I just tested again the keys and all seem to still work fine, no more bounces. The number 2 key seems to have improved by just leaving it alone and is behaving much better than before and really now only misses a much smaller percentage of the pushes then before. Probably the Deoxit doing its magic soaking in. Will leave it another 24-48 hours and if it still misses by then will try the mini wire brush trick to further work in the Deoxit.

[calc-calcs]

Just another quick update, the calculator sat now for about 10 days with zero use and after retesting all the faulty keys every one of them seems to be still working perfectly. Perhaps only for the 2 key I can still find a certain push angle that sometimes makes it slightly unresponsive, but its now more the exception than the rule and I would consider it perfectly usable.