HP-25 erratic behaviour


ciao a tutti ! (greetings to everybody), one of my HP25 is actually behaving quite funny. When turned on, random "0"s are shown on the display, keyboard is insensitive at all and there is no way to get further than that.

It can be the first led, the last, occasionally two zero's show up,
but the unit stays like that until switched off. When switched on again the digit that turns on change position, but that's totally randomic.

It has been cleaned, the three bottom ICs have been desoldered and resoldered, and the mother board has undergone a cleaning process with vinegar and water, but no improvements ....

Has anybody ever experienced something like this ?
As always, thanks in advance for advice and help,
take Alberto


Ciao Alberto,
The CPU of your beloved calculator is dead!
You could only replace it with one from a donor machine (all the Woodstock except 21 and 27).

Take care.



ciao Ignazio, how are you ?

Thanks for the reply, in your opinion, can I use a old CPU (the one with the gold plate on top) to replace what looks like a newer one (the IC body is all in gray plastic), is the pin-out the same ?

Take care and thanks again, Alberto


(Ciao, Alberto!)... an HP27 (not mine) behaving almost like your HP25, except that I could make it work when the battery pack was almost at the complete insertion point, I mean, with just a fraction of the pressure it makes when completely inserted. After being totally inside the compartment, the HP27 simply 'lost its e-mind'.

I checked all solder points and tracks and found one broken track in the lower part of the PCB, between the CPU and the ROM (the HP27 PCB layout closely matches the HP25´s). Eventually the broken track had its edges in touch depending on the battery position and pressure. It worked pretty fine after restoring the copper track.


Luiz (Brazil)

Edited: 30 Nov 2009, 9:26 a.m.


Hi Alberto,

Perhaps this is not the problem in your case (or perhaps you have already checked it), but when my 29C was exhibiting the exact symptoms you describe, poor electrical contact within the battery pack was the culprit (causing under-voltage to the calc). If you have re-celled the battery pack and no longer have that soldered strip between the cells, the spring in the far end of the pack must be kept really clean (I use a pencil eraser and alcohol swab) for good electrical contact. I give likewise treatment to the button contacts on the calculator, and the ends of the cells they contact. Lastly, after I insert the battery pack, I sometimes have to "set" it by pushing it as if to remove it, but stopping just shy of the release point, then easing off the pressure.

It's worth a try.

Best regards, Hal


First of all I'd like to thank everybody for the suggested tricks, that's what I ve done meanwhile : I have desoldered all the ICs and brushed the PCB to reveal the copper, I have then checked for broken traces, but found none. Resoldered the ICs recovered from a spare HP25 e cleaned with a soft brush and some alchool (isoprop...) to remove the soldering stuff. At this point, if power supply was attached with wires from a Woodstock battery packs, the unit didn't even light up,while if the board was mounted back in the shell and the battery pack inserted it worked !!! Then ... I did the most stupind thing I could have ever done ... I have attached an external charger (with the battery pack still inserted) and numbers have began to show randomly on the display. Since then, the unit looks like dead ...... shame on me .... However, I do have learned a lesson, and I'd like to thank everybody for the hints received. Take care Alberto


Once again, shame on HP for allowing such a marginal charging power supply design to be released with their otherwise outstanding product.
The term penny wise, pound foolish comes to mind.


Well, a functional keyboard, good case, spare battery pack and toss the charger. Wait for a functional, cosmetically crippled machine and replace.

TOSS THE CHARGER or keep it for posterity and your collection but never use it again!

Cut the battery pack central spine and insert fully charged NiCds using an external battery charger. Thats how I keep mine running. The 29C has been running for over a year that way and all my programs are still intact.

Cheers, Geoff


eeheheh, I had replaced the battery pack with NicD batteries, wich I usually recharge with the external battery charger, I admit it, it was a foolish thing, but it was also 2 AM in the morning, not that this excuse me in any way....


My approach is to use a duplicate model Woodstock as a charger for all my battery packs that I use in the other calcs. I never turn this calc on and only plug the charger into a live electrical outlet when I'm using it to charge a pack. Worst case I will eventually damage this calc in which case I will still have another one in perfect cosmetic and functional condition. I leave charged packs in my 25c and 29c, and swap them out when they get low. I have also found that flat top AA NiCads work better than high tops, apparently providing more consistent continuity with the calc's contacts and the springs.


to be very honest with you all, I did it twice (attaching the charger to the calculator) and the first time I have recovered the unit by resoldering (or at least re-passing) all the soldering points, I'll try later on and I'll keep you posted)


Hello everyone, the resoldering operation has worked out, and the unit is back to life !! I believe (but that's just my opinion) the most of the problems were due to a broken (even if not totally) link
between the two points of contact for the baterries and the main pcb !!! Anyway, lesson learned, I'll toss all the chargers ! Take care and thank you very much to all of you, Alberto


Ciao, Alberto;

I only connect the charger when I turn the calculator ON and it gives at least one single flash in the display, meaning that the batteries are in contact. If I see no sign of life, I remove the batt pack, charge it briefly, place it back and try again. If then I see no sign of life, I charge it with an external charger while trying to find the problem.

Laporte´s analysis is complete and perfect, should follow it with no remarks. Anyway, I prefer counting on the original calculator charger. Even if it is a Woodstock. I do not like too much to remove and reinsert battery packs, something may end up broken...


Luiz (Brazil)

Edited: 3 Dec 2009, 3:02 p.m.


My HP-29C exhibited the same problem, an intermittent contact between the battery contact rivet and the trace to the PCA.

Even after resoldering I had to desolder and soak the contact in some dilute acid to allow for a non-contaminated solder connection.

Great job, Cheers, Geoff


Hi, all;

most of you have already seen this, and I actually never thought of posting it based on the thread subject. Anyway, here it goes: HP21 power supply schematics and a suggestion for operation without batteries. NOTICE THAT THE POWER SUPPLY IS DIFFERENT OF THE ONE USED IN THE C-SERIES! (before editing: Not mine, do not remember who is the author, will post credits as soon as I find.)

Credits go to 'bh' (?); see below, please.

('head' extracted from original drawing, no changes)


Luiz (Brazil)

Edited: 2 Dec 2009, 4:57 p.m.


Hi Luiz,

How is the power supply in the C series different from the one for the regular Woodstocks?

I'm asking because I used to own an HP-25, which was transformed into an HP-25C during a just-out-of-warranty repair... I kept on using the same charger as before, and never had a problem.

- Thomas


Hi, Thomas;

the C-series must provide a continuous supply for the C-MOS RAM, and as we can see in the original HP21, all lines are down when the switch is OFF. The C-series must have something connected straight to the batteries, something like a diode + capacitor, that is not disconnected when the calculator is OFF (the capacitor should maintain the RAM data while changing batteries, for example). The same happens with the Spices: the power supply of the C-series is not the same as the E-series.

I know that the HP29C has a totally different power supply (separated board), and I guess the same with the HP19C (extra current for printing).


Luiz (Brazil)

Edited: 2 Dec 2009, 9:30 p.m.



At first glance, the schematic has some at least three mistakes of drawing around the output oscillator.
Be careful when you utilize the schematic.




Hi, Lyuka;

thank you!

Would you, please, point them out? I never needed to repair Woodstock´s power supply, so I never checked if it matches the actual circuit. In fact, I once noticed that the toroidal input and load coils were shorted in the circuit (forgot about that...), and I failed to remove the connection.

Thanks again!

Luiz (Brazil)

Edited: 3 Dec 2009, 12:23 a.m.


Off the top of my head, there are at least the following errors :

1) The oscillator and regulator transistors should be NPN not PNP. The part number 4-xxx is a shortened form of 1854-0xxx which indicates an NPN transistor. Not to mention the fact that a PNP transistor with its emitter connected to the -ve supply rail (ground in this case) is unusual to say the least. The circuit couldn't work with PNP transistors here. As an aside, the 6xx part of the transistor number is almost certainly a data code (1976), and is not important.

2) Both transformer windings are shown as shorted (this has already been mentioned).

3) But actually the transformer is either 1 tapped winding (autotransformer) or 3 windings with a common connection. If you desolder the transformer from the PCB you'll find DC continuity between all the pins.

4) The diode connected to the base of the regulator transistor is a zener diode (not suprisingly).

Is this the place to put in a plug for the HPCC schematics CD-ROM? It contains schematics of all woodstocks (and then some...)


And another one...

5) He specifies 'Capacitors in microfarads' and then labels the oscillator feedback capactior '2000K'. I can assure you that capacitor is not 2000uF, or 2000KuF (=2F!). It might well be 2000pF (2nF), but I would have to check.



Thanks Tony. What you mentioned are what I thought as errors.




Right, this circuit won't work at all.

The oscillator on the HP-25 uses a NPN 4-550 and the regulator (Zener at its base) uses another NPN 4-071 (HP parts 1854-0550 etc).

The cap is a 2nf in series with a 56 ohm resistor.


Jacques Laporte has drawn out the PSU of the HP25 and posted it on his website here, see Fig. 10 about halfway down.

Thank you Jaques for sharing your knowledge! :-)

Edited: 4 Dec 2009, 6:51 a.m.

Possibly Related Threads...
Thread Author Replies Views Last Post
  Strange HP 25 Problem (Repair) Onur Ilkorur 5 1,122 12-06-2013, 05:13 PM
Last Post: Onur Ilkorur
  HP-25 Wayne Selway 1 519 11-22-2013, 12:58 PM
Last Post: Michael de Estrada
  Interesting Base Conversions - Porting a 1975 HP 25 Program to the HP 35S Eddie W. Shore 1 559 10-13-2013, 07:49 PM
Last Post: BruceH
  HP-25/25C Potential Matt Agajanian 14 1,869 07-17-2013, 02:57 AM
Last Post: Matt Agajanian
  [WP34s] strange behaviour at number entry Dieter 12 1,613 06-06-2013, 05:36 PM
Last Post: Harald
  HP 39GII Undocumented behaviour Bunuel66 0 354 05-12-2013, 05:23 AM
Last Post: Bunuel66
  Recharging original batteries in HP 45, 65 and 25 Michae Altmann 1 511 03-05-2013, 12:21 AM
Last Post: DavidShenk
  HP-25 keyboard Brad Barton 5 755 02-16-2013, 09:59 PM
Last Post: Brad Barton
  HP 65/25 shout-out in interview with Paul Lutus Egan Ford 3 621 12-21-2012, 11:09 PM
Last Post: Mike Morrow
  New HP-25 Emulator for iPad, iPhone, and iPod Namir 5 806 05-22-2012, 06:22 AM
Last Post: Mike (Stgt)

Forum Jump: