The Museum of HP Calculators

HP Forum Archive 13

[ Return to Index | Top of Index ]

HP41 Battery question
Message #1 Posted by Harry (Germany) on 26 July 2003, 2:19 p.m.

Does the original HP41 rechargable battery pack have 5 cells? I just put 4 N-size rechargables into the battery-holder and the calculator didn't power up ?!?


Re: HP41 Battery question
Message #2 Posted by David Smith on 26 July 2003, 2:23 p.m.,
in response to message #1 by Harry (Germany)

Nope, they have four cells...

Re: HP41 Battery question
Message #3 Posted by Vieira, Luiz C. (Brazil) on 26 July 2003, 2:54 p.m.,
in response to message #1 by Harry (Germany)


maybe it's a silly question, but... Have you tested each cell? Do all of them show the same voltage when measuring with a voltmeter/multimeter? Also, have they all been found with correct voltage, it's good idea to "load" them and check for current intensity. For testing purposes, a 15 ohms resistor should demand about 90 mA, as fully charged NiCad's offer about 1.2Vcc to 1.3 Vcc. If voltage is too low on any battery or current does not reach at least 80 mA in all of them, recharging each sell separately or replacing bad one(s) may be a solution.


Luiz (Brazil)

Re: HP41 Battery question
Message #4 Posted by Harry on 26 July 2003, 5:50 p.m.,
in response to message #3 by Vieira, Luiz C. (Brazil)

Yes I have tested them. And they are all good, as i expected, as they are new. They all have 1.30 to 1.32 Volts and go down to 1.1 at 3 (!) Amps. So thats not where the problem is. The Alkalines all have about 1V and still work fine. I cleaned all the contacts again. Even soldered metal plates to the cells now, because I thought they might not touch the contacts for some funny reason.
Now its getting wired: If I put 4 Alkalines in, it works. If I replace them one by one with rechargables and put the pack in the calculator after each one and turn it on and off again, then I get it working on all 4 rechargables. But if I take the 4 Alkalines out and put 4 rechargables in, it does not.
I have no explanation for this VERY WIRED behaviour.

Harry, who goes crazy now

Re: HP41 Battery question
Message #5 Posted by Vieira, Luiz C. (Brazil) on 26 July 2003, 7:48 p.m.,
in response to message #4 by Harry

Hi, Harry; (Harald, right?)

if it is like you say, find a place for my brain beside yours, too...

I once had a problem with one alkaline cell, but as you measured all of them, not a way! Actualy weird!

One last possibility: have you thought aobt physical measures? I mean, size? Try "stretchning" the battey compartment's springs a little bit to enhance the pressure.

Success. An let us know what happens.

Luiz (Brazil)

Re: HP41 Battery question
Message #6 Posted by Harry on 27 July 2003, 7:59 a.m.,
in response to message #5 by Vieira, Luiz C. (Brazil)

Yes, Harald is correct.
I tried it over and over again. Same behaviour as described. I guess i always need a set of "startup" alkalines to get it running on the rechargables.
Here is one idea i have: When I put the batterys in, the first thing it does is probably to charge the backup capacitor. Now what if it draws a too high current because of the very low resistance of these new rechargables. Do you think that could lock it up then? Maybe I avoid this high current by replacing the alkalines one by one with the rechargables. The alkalines have only 1 Volt per cell left, so the Voltage of the rechargables is higher now and the capcitor has to be charged.
I will get new Alkalines and try replacing those with the rechargables (all four at the same time) maybe that works.

Regards, Harry

Still strange behavior...
Message #7 Posted by Vieira, Luiz C. (Brazil) on 27 July 2003, 11:15 a.m.,
in response to message #6 by Harry

Hi, Rarry;

I'm following this situation with a particular interest and curiosity. A behavior like this is usualy related to a simple problem that was not supposed to happen. I'm thinking of faulty (dried) capacitor in mainboard, bad contact somewhere which somehow happens when batteries with different dimensions are inserted, and stuffs like this. Based on what you mention, when voltage raises in steps of about .3 volts, the calculator accepts rechargeable batteries. I also understand that if voltage drops of about 1 volt (when you return alkaline batteries back) then it's O.K. But if you change the set completely you'll force an extra 1.2 volt step and the system seems to hang at this point, right?

If in your case you replace a set of nearly exhausted batteries (1 volt each) for a new, fresh set, voltage will be increased in about 2 volts, from 4 Vcc (4 × 1V) to 6 Vcc (4 × 1.5V). I wonder of what's gonna happen!

Mostly I wonder about the cause, the reason. I replaced exhausted batteries for new, fresh ones in almost all of my calculators. Let's not forget that the HP41 was the first LCD calculator from HP and some units get stuck with a high pitch frequency sometimes. I once tracked (with the unit touching my ear) to "see" where it was comming from and I found that it was from the small coil in the power supply and not from the buzzer.

I think that there are situations where the power supply circuits may hang and others taht it will not hang.

I cannot go further from here. I am now waiting to "see" what's gonna happen if you use four new, fresh batteries at once and then you change for the rechargeable pack, say, a reduction in voltage. To be completely honest, I cannot even dare pointing a faulty condition, neither an operational characteristic as a cause for this "behavior".

Let us know as you have new information. Sorry not helping further.

Luiz (Brazil)

Re: Still strange behavior...
Message #8 Posted by Harry (Germany) on 27 July 2003, 2:31 p.m.,
in response to message #7 by Vieira, Luiz C. (Brazil)

Thanks for your interest in this strange problem.
I have found annother set of new alkalines. They are exatcly the same as the old ones, but fresh. I put them in and .... nothing, it does not work, same behavior as it shows with the rechargables.
Now I diged my mint HP41CV out of the desk, that I did not really wanna use, but this was driving me crazy. So I did it anyway. Surprisingly - or maybe not surprisingly anymore at this point - it works just fine on all 3 sets, the exhausted Alkalines, the new ones and on the rechargables.
Now please help me, what do I do if my calculator runs on exhausted batterys only?!?

Regards, Harry

Re: Still strange behavior...
Message #9 Posted by Vieira, Luiz C. (Brazil) on 27 July 2003, 2:52 p.m.,
in response to message #8 by Harry (Germany)

Hi, Harry;

I was driving myself crazy. But now we can be sure that at least all battery sets are working fine. (Hey, Valentin, a "supposingly" useful hint for working calculators kept by collectors: testing each other!).

Have you checked if it is a fullnut or a halfnut?

This is an original MoHPC image.

I remember that the fullnuts have an IC in their mainboard that's the heart of a power supply. I read about this and never called my own attention for it... 'till now.

I'm not sure about how does it work, but there is something that I did not ask you so far: do these four 1volt alkaline batteries cause a low battery condition, say, is the BAT annunciator on? If it is not on, chances are that something is wrong with your calculator's power supply. I mean, if there is a voltage sensing component, like a voltage divider, and this(these) component(s) is(are) faulty, then the low battery condition is not sensed in fact. Maybe the 4V from the exhausted battery set is taken as normal voltage, and raising it slowly does not bother the power supply. Let's also think that the same faulty reference informs that anything above 5V is excessive voltage and the circuit does not allow the power supply to turn itself on to protect calculator's circuits. So, connecting anything higher than or equal to 5V instantaneously will not allow the calculator to switch to ON, but changing batteries one by one will not trigger sensing/protective circuits.

I'd first check schematics for this calculator (there are some different types of mainboard for the HP41; I've seen four myself) or inspect the mainboard itself in order to try to "see" any electronic components connected to the power supply IC that would be working as voltage sensing and/or reference. I have no schematics in hands for now, but I'll try to find some further info. Maybe Tony Duell or David Smith or others (I remembered their names because I read some of their posts a few minutes ago) will add other important information.

Success! As I have anything new I'm posting again.

Luiz (Brazil)

Edited: 27 July 2003, 2:56 p.m.

Re: Still strange behavior...
Message #10 Posted by Harry (Germany) on 27 July 2003, 5:47 p.m.,
in response to message #9 by Vieira, Luiz C. (Brazil)

Well, first, it is a fullnut (s/n 2127S11343). The BAT indicator is not on after turning it on. But when I run a program it comes on after a few seconds. BTW, thats when I stoped calculating primnumbers. I wanted to put the rechargables in and go on. I didn't know it would get so complicated....
When I bought that calculator about 1.5 years ago it had a strange behaviour too. After putting batteries in for the first time, it had a very dim display that was hardly readable. It got better the longer it was turned on. After like a month with the batteries in it that problem was gone and never appeared again. I have taken the batteries out of it an put back in, sometimes after a month of not using it, and it started with "Memory Lost" but a bright display. I used the same set of batteries all that time. The one that is exhausted now.

Regards, Harry

PS: I did not open it yet, I will probably do that next weekend.

Re: Still strange behavior...
Message #11 Posted by Axel Poqué on 27 July 2003, 3:00 p.m.,
in response to message #8 by Harry (Germany)

From what you're describing it seems like the voltage regulator circuit of your HP is broken. Electronic circuits need certain voltage levels to operate properly (e.g. 5 V for TTL circuits). Typically the battery/charger voltage of an electronic device is higher than the actual required operating voltage to allow for voltage drops and optimal use of batteries. If the regulator doesn't work, the supplied voltages are too high and can cause strange behaviour or even damage the device.

Axel Poqué

Re: Still strange behavior...
Message #12 Posted by Harry (Germany) on 27 July 2003, 5:51 p.m.,
in response to message #11 by Axel Poqué

Isn't the HP41 cmos, wich would not have problems with the higher voltage?
The powersupply is definatly the first thing I will have a closer look at when I open the calculator

Regards, Harry

Re: Still strange behavior...
Message #13 Posted by Axel Poqué on 27 July 2003, 6:30 p.m.,
in response to message #12 by Harry (Germany)

Yes, that may be right, but some other part might have narrower voltage requirements (the display perhaps? or are there TTL support chips? TTL chips do have a much narrower margin - 4.75 to 5.25 V I think). Have you tested if the calc beeps? If this is so, only the display can't cope with the voltage levels.

It would be nice if we had some schematics/specifications, so we could test this theory...

Regards, Axel

P.S.: Viele Grüße aus Aachen :-)

I am quiet sure I know what it is now
Message #14 Posted by Harry (Germany) on 27 July 2003, 8:16 p.m.,
in response to message #13 by Axel Poqué

I powered it by an external powersupply now, and here is what happened:
If I don't restrict the maximum current, it draws a current of about 500mA when I connect it to the powersupply at a Voltage between 4 and 6.2V.
If the Voltage is below or above (I tested it up to 6.5V) there is no problem and it works just fine.
Once it is connected I can change the voltage and there is no problem between 4 an 6.2V anymore.
The problem also does not appear when the current is restricted to 50mA, wich makes sence as the Voltage drops when i connect it and then rises again.

All this tells me it is the internal powersupply that is faulty. Its probably the oscillator that does not start oscillating for some reason.

Any ideas what that reason could be?


Re: I am quiet sure I know what it is now
Message #15 Posted by Vieira, Luiz C. (Brazil) on 27 July 2003, 9:16 p.m.,
in response to message #14 by Harry (Germany)

Hi, Harry;

an off-topic, relaxing comment: I see that not only we are "night workers" but also completely "workaholics", if that's the way it's spelled... In Brazil it's 22:10 PM, I'm delving into my post-graduation text and as I need a brief pit-stop, I read the posts. Wow!

I read Axel's posts and I would add that this selectivity "sounds me" like semicondutor trick, not exactly passive-component behavior. A few months ago I had an HP41C in hands with hig-current drain only when it was "ON". The LCD driver was the cause, and David Smith mentioned the fact it is a bipolar technology device. As it has capacitive load (LCD segments), current drain for this particular IC is not so much relevant. Anyway, depending on its' internal configuration/arrangement, chances are that depending on voltages supplied to it, a diffeent set of transistors may be switched to ON or OFF. If one of them is bad, high current drain may happen as well for an specific voltage range. I'd not be surprised if you have normal operation after simply replacing the LCD driver (the one with an 'RAYC' in its ID code). The problem is that a donator must be found, no commercial replacements so far.

Best regards and success.

Luiz (Brazil)

Edited: 27 July 2003, 11:03 p.m.

I am quite sure I *don't* know, but...
Message #16 Posted by glynn on 28 July 2003, 12:40 a.m.,
in response to message #15 by Vieira, Luiz C. (Brazil)

While not discounting the idea that it may well be an LCD driver, INCLUDING of course the saga Harry recounted about the dim display when he first got the calc, I am perhaps one of those conservative types who like to check spark plugs before going off to buy a new engine. Unless Harry happens to have a couple of spare known good LCD drivers around, he will also want to look at the most common failure modes of circuits before moving on to the more esoteric.

An oscillator circuit can be designed different ways, of course, and I don't have any knowledge of the HP41. But let us ask if the circuit has any external components to "tune" the frequency of the oscillator. Are any of them not at rated value? Especially capacitors (a good case could be made for replacement of these in lieu of trying to measure them; alternately, temporarily paralleling one in with another sometimes helps figure out what is wrong). What frequency should be output? Is there no output when the calc is acting "hung"? Is the calc in fact inactive, or is the lack of a display leading us to think "hung" when in fact the cpu is active and the output is simply not visible to us?

Visual inspection of contacts, pads, traces can be of value, if only to assure that some corrosion or partially-made contact is not acting like an added resistor in some timing-dependent function. Of course, measurement with voltmeter or even better, having an oscilloscope handy, is useful IF you have ready access to someone who can tell you what voltages you SHOULD be seeing or what waveforms you SHOULD be catching.

I merely suggest that while cannibalizing one machine to test another is very Tempting, leave your good machine good for now and instead gather as much information on this badly-behaving one as you can. I'd be willing to bet it *IS* either failure of a passive component like a capacitor, or a contact/trace/conductive failure near the power-supply circuit. At least, that is my first "intuition".

Hi, Glynn!
Message #17 Posted by Vieira, Luiz C. (Brazil) on 28 July 2003, 2:26 a.m.,
in response to message #16 by glynn


You know, I read your post and I think any valid suggestion, analisys, question, reasoning like yours MUST be taken into account.

One of the most "interesting" fact here, if I can classify it so, was the 500mA (½A) current consumption, a considerable high current for the HP41, when the calculator was "hung". And it happens when the calculator is connected to a power supply delivering a very specific voltage range. Out of this range, up or down, there is no more high current.

I "connected" this fact to semiconductors, not to passive components or faulty cooper trails. But you're right, a pre-check for not so complicated causes must be made. A broken cooper trail may disconect two pins and may also cause strange behaviors. I wrote about this in one of my first posts in this thread:

A behavior like this is usualy related to a simple problem that was not supposed to happen.
It seems I forgot my own words...

Thank you for the enlightment, Glynn.

Best regards.

Luiz (Brazil)

Re: Hi, Glynn!
Message #18 Posted by Chan Tran on 28 July 2003, 12:39 p.m.,
in response to message #17 by Vieira, Luiz C. (Brazil)

I read in an article somewhere when the 41 was introduced that it has a voltage regulator that can provide a steady 5V when the battery voltage is from 4 to 6V. So is it possible that the regulator went bad?

Re: Hi, Glynn!
Message #19 Posted by Harry (Germany) on 28 July 2003, 2:44 p.m.,
in response to message #18 by Chan Tran

Thats exactly what I am thinking. I was talking about the oscillator in that voltage-regulator (or powersupply).


HP41C power supply
Message #20 Posted by Ellis Easley on 30 July 2003, 12:31 a.m.,
in response to message #19 by Harry (Germany)

You know, there is a 41C service manual on CD5 of the Museum CD-ROM/DVD.

HP41C power supply
Message #21 Posted by Ellis Easley on 30 July 2003, 5:38 a.m.,
in response to message #19 by Harry (Germany)

I've been looking at the 41C service manual on the Museum CD5 (schematic on page 36 and power supply Theory of Operation on page 10) and comparing it to the re-drawn 41C schematic that has been available for a while and I think the re-drawn schematic introduces some errors. For one thing, the 14 pin bipolar chip labeled "LCD-DRIVER" on the re-drawn schematic is actually the power supply chip (U2). One of its functions is to provide three voltages (nominally 1.1, 2.2 and 3.3V on pins 4,3 and 2) to the actual LCD driver chip which is part of the Display Assembly. The power supply chip also temperature compensates the three voltages, which is required for consistent display contrast over temperature. Another function of the power supply chip is to monitor the input voltage (the power supply chip gets its operating power on pin 14 from C1 and through CR1 from the battery) and if it falls below 4.2V, it pulls its pin 5 low, which is connected to CPU pin 6 (LLD) - the text says this pin's function is low battery voltage detection.

The Theory of Operation section doesn't go into the details of how the power supply chip generates the 6V Vcc operating power supply but it describes how the CPU turns the power converter on by driving a current into pin 13 (Vci) (from CPU pin 8 through R1) and how once the converter output voltage reaches 6V, the power supply chip drives its pin 6 (Vco) low to signal the CPU (on pin 7) that power is good, with a capacitor (C5) storing the low level (between power converter switching cycles, I think).

My guess is that the resistor (R3) and capacitor (C6) on pin 7 control the time constant of the power supply chip's power converter and that pin 10 (IND) is the converter switching output, driven low to store energy in the inductor (T1), and then when the current is turned off, the counter-EMF of T1 is rectified by CR2 and filtered by C3 to generate Vcc, which is sensed by the power supply chip on pin 9 to control the timing of the switching output on pin 10 to maintain Vcc at 6V.

The service manual Theory of Operation doesn't say anything about the resistor (R2) connected between power supply chip pin 12 and ground (although there might be something about it in the troubleshooting section). The signal is labled "ISET" which would generally mean "current set". Some oscillator chips use a constant current source to charge a capacitor to establish a time constant, so this pin might control that aspect of the power converter. Or the current could control the internal voltage references that determine the output voltage and/or the low battery voltage and/or the three LCD voltages. Or, given that it is named like a current input and the pin next to it (Vci) is described as a current input, the current flowing from pin 12 to ground might establish the current level into pin 13 that represents a logical "1".

There are a lot of differences in the component reference designators on the service manual schematic and the re-drawn schematic - this might be because the two schematics represent different revisions of the logic board. There are also differences in the pin numbers on the board-to-board connectors on the two schematics. It might be that the person who re-drew the schematic assigned his own pin numbers, if they aren't marked on the PCB's (which could also explain the different reference designators). Also, the re-drawn schematic doesn't have the two transistor circuit that controls the DATA signal from the Phase 2 clock. Since both signals are connected to the CPU, maybe the transistors are eliminated by a revision of the CPU.

[ Return to Index | Top of Index ]

Go back to the main exhibit hall