My original HP-25, ailing

[Peter Klein]

I took my original vintage 1977 HP-25 out of storage and tried it out. The last time I used it, the calculator worked fine, but the memory (storage registers and program steps) did not. A couple of number keys were intermittent. This time, the calculator would sometimes display the opening "0.00", and I could enter a few digits, but then the display would start to go crazy, as if each LED digit was rapidly vibrating on and off at random, and/or baubling on and off one after another from left to right. And eventually the display would go completely dark, and stay that way until I disconnected battery and reconnected the battery pack (running off two Emloops NiMH AA batteries). I also tried a 3v power supply, same symptoms.

I'm torn. I could buy a Panamatik kit for $95 U.S. and attempt to fix it. Or I could sell it off for parts and move on. It would be nice to revive the calculator that started me on my HP journey. But I'm not sure if I want to sink time and money into a 46 year-old calculator, considering that I have a bunch of more advanced models that work well, and that I think either a DM32 or DM42 is in my future.

I have an oscilloscope, so I can measure the timing signals as suggested in the Panamatik manual. If I decide on fixing it, I can run that test before I commit to the kit.

Could someone point me to an article or post that explains how to clean the keyboard buttons? And how to remove the rubber feet that conceal the fastening screws without damaging the feet?

[teenix]

I think it would be worth trying to fix. If you sell it, once it's gone it's gone, and you might be left with that nagging feeling, I wish I kept it. Besides, even if it is not working now, it may get a touch up in the future, and sitting idle in a cupboard doesn't cost anything.

It might be something simple like a failed capacitor in the power supply, or something worse like a failing ROM or ACT. Check for signs of corrosion and clean, you may be surprised.

You could search "cleaning woodstock keys", and get some hits.

https://www.keesvandersanden.nl/calculat...o_open.pdf

I'm probably a million miles away down under, but I would have been happy to give it a look over.

cheers

Tony

[John Garza (3665)]

I definitely agree with Teenix. Better to keep it even if you don't have an immediate solution.

You would be surprised what corrosion can do.
Ensure your contacts are scrupulously clean, then use full strength deoxit.
When a battery or corrosion is suspect NEVER use the AC charger.
In fact, it's good policy to never use the charger and always charge the batteries externally, just in case.

If it's not corrosion, it will likely need a new chip from a donor machine or a Panamatik upgrade.

I was in a similar situation a while back. I had an HP-29C that would not take a program. I installed a RAM chip from a donor HP-25C. Then I installed a Panamatik in the HP-25C. Both are working fine now.

To Disassemble:
remove battery
remove two top rubber feet by inserting a small (little greenie) flatblade screwdriver to pry them up.
unscrew the screws that were hidden under the feet.
press on one of the charger contacts with the screwdriver to move the keyboard away from the bottom. Gently pry the halves apart.

Check for noise on the power supply (bad capacitor)
And check the timing signals as per the Panamatik docs.

Access to the keyboard can take 1 of 2 paths:
1: If you can see holes in the back of the keyboard that is the easy path
Spray deoxit in the holes and scrub with some fine stranded copper wire in each hole. Flush any debris out with spray deoxit, Let dry.
2: No holes requires you to shave off the tops of all the little heat stakes.
Then scrub under the metal ribbon strips with deoxit and a homemade plastic or cardstock tool as described in Teenix's doc. Reassembly will need Gorilla glue (the type needing water to activate). Lay the keyboard back on the faceplate with the heat stakes lined up in the holes. Drop of water on each stake, then Gorilla glue, then spread it around in a little circle about 2mm diameter with a toothpick. Do this for each heatstake. Let it dry overnight. This duplicates the 'mushroom top' of the original heatstakes and is very strong. The water activated glue expands during the curing process to make the wide top.

Also, while you have the keyboard apart, clean and re-lube the slide switches with silicone grease.

Good luck !

-J

[BobVA]

If you decide to repair it, also take a look at Panamatik's Woodstock "LP" kit. The convenience of a 25 that runs for months from lithium AA's (at least in my use case) and has continuous memory makes it a very practical daily driver. And the clock is pretty cool!
The LP kit will also revive a 25 that is suffering from more faults than a New ACT replacement can fix.

25LP vs a DM42? Touch choice! I believe the forum wisdom in such cases is "buy both" :-)

Keep us posted!

[Duane Hess]

Peter:

Not sure if the behavior is the same as a 27 I have. An ultra-cheap, known defective item. When powered on all digits lit at once and "shimmied." Or could be referred to as vibrating. Don't know what baubling means, but recall a VERY faint left to right intensity ripple.

Kept looking at it and noticed a pattern I could not really see. Was too close to a window. Each segment of all digits shimmied VERY rapidly simultaneously. Hence, it appeared the entire display was lit with shimmering 8's. In a significantly dimmer location I saw the pattern. Within the shimmer all digits were performing a count (up/down-forget) in unison. All 0's, then 1',s, etc. to all 9's (or the reverse direction) and continually repeated. The count image had an incredibly short duration. i.e. a ghost-like flash amungst all the shimmering.

The display going completely blank didn't happen to me.

Since sold as known defective, I wasn't surprised. Opening it up, it looked like most components were resoldered on the power supply portion of the circuit board. i.e. if you look at the PCB from the front, most every component to the left of the power pack hole was re-done.

For whatever reason, I was checking continuity in that area. (this was a good 5-7 years ago, forget reasoning; if any) I noticed a trace was weird. i.e. measured a value (close to zero) between 2 points. Then had largely varying values elsewhere. I was measuring the trace itself, so should have been small everywhere. Held one probe stationary and slid the other along the trace, watching the ohmmeter. As approaching an area which looked like a crack occured, about 1/4" of the tracing fell off the PCB.

Please reference the link:
https://www.keesvandersanden.nl/calculat...inside.php

At the top are pictures of the front/back of the PCB. Look at the left image. The tracing that fell out was the long one, immediately left of the battery "hole." (don't recall if I checked the 27's behavior at this point)

Don't recall how many connections the trace has. By checking all the images (both sides) you can see it goes to one of the 3 holes at the very bottom of the PCB. The one closest to the middle of the calc. It goes to a chip at the top of the PCB. Believe the left chip labeled 7551LX (upside down) in the row/column driver picture. As I recall it was the 3rd or 4th pin from the left (right?) on the top edge of the chip.

Running a copper wire between those 2 points resulted in the 27 working 100%.

Moral of the story?
- ..... dern if I know whether there is similarity to your situation. Doesn't really sound the same. Mine had no functionality at all.
- never trust a cracked trace you might replace the wrong thing?? I have no electronics training. Usually can only handle the obviously broken situations.
- I don't know, that's just what happened to me.

For what its worth.

[Peter Klein]

Tonight I opened up my HP-25. I cleaned the battery terminals, which had quite a bit of corrosion on them. I also cleaned some oxidation on the power supply traces at the left of the main board (see picture). My keyboard is the type that has holes on the back of the board, so I flooded them with contact cleaner and brushed the "4" key out with a bit of stranded wire (it was barely working before), then flooded again and dried.

Results were dramatic. "It lives!" The RAM was already dead, so no registers and no programming. Everything else is working again. The "4" key has improved, but is still intermittent. So is the "8" key. So I'll have to clean them again. There is one quirk: The calculator shows low battery with a pair of Eneloop rechargeable NiMH AA batteries supplying 2.6v. Another, at 2.8 volts, works, but gets a low battery warning with 8 digits displayed. The calculator happily runs with a pair of standard AA alkaline batteries supplying about 3.05 volts, or a 3v regulated wall wart that supplies about 3.16 volts under load.

I suspect that the electrolytic capacitor in the internal switching power supply has degraded, even though it is not visibly leaking. See picture below, it's the silver cylinder near the bottom. It's 60 microfarad, 6 volts. I have a 22 microfarad capacitor that I may be able to fit in. A local electronics store has 100 uf 10v capacitors. Would those work? Failing that, does anyone know where I can get a replacement in the U.S. that's slender enough to fit between the main board and the keyboard?

I ran the tests on p. 12 of the Panmatik manual (http://panamatik.de/ACTManual.pdf).
Pin 1 varied between 5.75 and 6.1 v.
Pin 2: -12 v.
Pins 16 and 17: 176.8 khz, probably close enough to 180 spec.
Pin 20, Sync: approx 1.9 khz, with some variation. Some noise was evident, but I don't know if this was pickup from the test leads, or internal noise.

Pictures below. Further comments and advice welcome.

   

   

   

[teenix]

Just a quick look, I found...

Element 14 Part No. 1165416

Mouser Part No. 75-500D606M025CB2A

Digikey Part No. 500D606M025CB2A-ND

It would be good if you could get one local to save expensive delivery.

A more common 68uF would probably do ok.

The scope signals look ok.

cheers

Tony

[John Garza (3665)]

Also... ensure you have excellent conductivity with the battery. Even with corrosion removed, sometimes the metal plating flakes off. And Woodstocks only have one point of contact (the tangent of the circular disk contact) as opposed to two contacts against the v-shaped battery contacts in Classics.

On the capacitor - it is likely a filter cap, so you may want to check the power lines for noise on the scope.

-J

[Peter Klein]

I just realized that the capacitor may be tantalum, not electrolytic. Tantalums are much less prone to go bad. I'm also wondering if maybe the dead RAM chip is loading down the circuit. This could explain why the calculator needs slightly higher voltage than it should. If I remove the RAM chip, would the calculator still work?

The objective here is to get the calculator working well enough to be reasonably sure that I won't be wasting my money getting the Panamatik kit.

[teenix]

If the RAM chip was loading the power supply I would expect it to get warm. It may not though.

If the RAM does not work, then removing it should not cause problems.

The RAM chip is not accessed at start up except for a [Clear Data Registers] instruction which initializes all RAM to zero, so the calculator should start ok.

cheers

Tony

[Peter Klein]

Thanks, Tony. I cut the -12v pin on the dead RAM chip. The calculator immediately started working properly with the 2.6-2.8v rechargeable battery pairs. It starts to show the low battery indication at about 2.1 v. Now I'll have to decide whether to get the Panamatik kit or just leave things as is.

[teenix]

Nice one.

Quote:Now I'll have to decide whether to get the Panamatik kit or just leave things as is.

...or find a cheap donor :-)

I've been working on a few chip replacements, firstly a ROM 0 chip for my HP-92 which I accidently damaged.

I did have it mostly working, but finding time is the difficult part.

cheers

Tony

[Peter Klein]

I decided to get the Panamatik ACT repair kit. I thought about the more expensive LP (Low Power) version that replaces everything but the keyboard, case and LEDs. But since I have a Voyager and several Pioneers with most of the LP features I'd use, I decided that I'd prefer the standard ACT. It retains the 25's original character, adding a few welcome features including "constant memory."

I also figured out an intermittent battery issue that has been plaguing me from the start. The following description assumes you are looking at the calculator and battery holder upside down, as they would be when you remove or insert the battery. Years ago, I removed the cover of the original battery holder with a razor blade so I could put my own rechargeable AAs in. I then tape it shut with scotch tape.

There was just enough space or play in the above arrangement that when the battery contact pushed the positive "cap" of the right hand battery (+ side) back against the spring tension, it could slide up. A bit of the cap would stay caught behind the hooked tabs that go into the slots above the calculator battery contacts. These contacts normally touch the battery near the bottom of the positive cap, not the center. With the cap shifted up and back, there was intermittent contact to no contact at all.

The fix was to make an approximately 1mm thick shim out of the end of a plastic garbage bag fastener. I taped said shim to the inside of the lid of the battery case, in the curved area just behind the battery's positive terminal. Problem solved.

[John Garza (3665)]

I've found the best way to update the Woodstock battery packs is NOT to cut off the back plate. Instead, cut out the middle plastic bar, when you are looking at the batteries. There is enough flexibility in the plastic to pop in batteries, rechargeables or alkalines. No need to replace the bar.

-J

[rprosperi]

(11-10-2023 02:22 PM)John Garza (3665) Wrote:  I've found the best way to update the Woodstock battery packs is NOT to cut off the back plate. Instead, cut out the middle plastic bar, when you are looking at the batteries. There is enough flexibility in the plastic to pop in batteries, rechargeables or alkalines. No need to replace the bar.

-J

You don't even have to remove the middle bar. Just cut the bar (e.g with an X-acto knife) at the end where the contacts are and you can still pretty easily replace the 2 AA cells (use flat top NiCD - alkalines will work but the extra length of the buttons make the pack very hard to remove). As John says, the plastic will flex enough to let you remove or insert batteries.

And if you forget polarity (which is very important for Woodstock), it's actually etched on the inside of the plastic pack, on the inner sides of the long upper portions that hold the cells inside.