HP 10bII+ self test


Hi all,

I found this self test for the old HP 10bII.


it does not appear to work for the new HP 10bII+ - all i get is a "620_ " display.


That factory test is no longer there. It is more a factory test like the 20/30b. You get a menu which you can select different types of tests/information using + or - to move up or down, and then INPUT enters the test.

Also, we no longer publish the factory test information in the manuals. There have been problems in the past of users doing things incorrectly in there and deciding the calculator is "broken" when in fact it is just fine. Thus the decision was made to not publish the information.

I guess the idea is that "if it isn't in the manual, nobody will know about it. . ." ;-)


Edited: 15 May 2011, 9:00 p.m. after one or more responses were posted


Indeed, your comment will achieve exactly the opposite effect!

Outside of those who already know how to do this from "inside information", who's going to be the first one here to figure it out?



It would save us time if one of you would just tell us but ... I guess it wouldn't be a challenge then. :-)


It isn't a secret or anything. Just press ON-PMT (from memory - might be remembering wrong) and then you are in a menu. From there, it is pretty easy to figure out - especially if you've seen the 20/30b menu before.

There is one easter egg in the calculator, and it is in the test system menu. :-)


Edited: 15 May 2011, 9:02 p.m.


LOL! this is going to be fun, i guess it won't be very long before someone in this forum going to prise the easter egg and the self-test sequence.

now, pressing ON PMT together yields " 1 5crEEn "

hmmm.... let's see...


That is "1 Screen" within 7 segment limitations. Press + and - to move up and down the menu. INPUT activates the test/setting.


Edited: 15 May 2011, 10:48 p.m.


An Easter Egg? I had no idea, sounds like fun!

When viewing the "6 Copyright" message, what's the significance of 1752 that displays after the "Copr HP 2010"? I know it's the year Ben Franklin did his famous kite experiment. Does this mean that the 10bii+ is useful as a lightning rod too? :)

Edited: 16 May 2011, 12:15 a.m.


Size in bytes of the internal calc structure.



Actually pressing the "=" key does it not the "Input" key from what I'm seeing on mine.


Ah, well that is correct then. I made that about a year ago so its been a while. :-)



On my 20b the battery test gives a 2.7. Is that good or bad? I've never had a low battery warning, but most times when I turn the calculator on it's completely reset.


I'd say try to notice when the low battery warning comes on then check what the voltage says at that time so you'll know where HP set the lower threshold.


FWIW, in wp23s I've set the threshold to 2.4V. If I understand the HP 20b SDK sources correctly, HP sets the low battery annunciator below 2.4V and shuts down the calculator below 2.1V.


I'd set the low voltage threshold somewhat higher then 2.4 volts to give the user more warning time, especially if the firmware doesn't check the shutdown voltage pretty much constantly.

The 20b/30 draws a lot of current when running a program (assuming a reasonable clock speed) and becasue of this the internal resistance on the CR2032 cells gets up to a very high value near their end of life. When this happens the voltage drops precipitously when running a program. A measured 2.4 volts when idle might be well under 2.0 volts as soon as a program starts

IMHO, two CR2032 cells in parallel is an extremely poor choice for a power source on a programmable calculator that runs at high speed. While the 10bii+ and 20b were never intended to be programmable the 12C+ (certainly) and 30b (probably, Cyrille has pointed out here that it was supposed to just have keystroke macros) were.


Thanks for the responses. Between the constant resets and the keyboard problems I think it's time to use it for target practice. Maybe I'll put up a YouTube video of the results.


Here's my HP 10bII+ (serial no. CNA 04201622) self test results:

1 Screen -lights up all segments, icons, display

2 Keyboard - at the end of the test, yields "Good"

3 Checksum - yields 244.244 (anyone got a different value?)

4 battery - 3.500

5 version - 7-13-2010 (any different version?)

6 copyright - HP 2010 then 1,752

7 test sys - pressing = turns test sys off ?

8 Slow clk (clock?) SC -28,500.00 - what does this mean?

where's and what's the Easter Egg then, huh? :-)

actually, I am beginning to fall in love with this little algebraic HP ;-)


8 Slow clk (clock?) SC -28,500.00 - what does this mean?

I have an idea. The chip is equipped with two independent oscillators with frequencies of 32.768 KHz and 2 MHz respectively. Since none of them is quartz controlled the precision of the oscillators is questionable at best. In the 20b/30b a quartz can be added to the 32KHz oscillator but in the 10bii+ there are no pads on the board.

The "slow clock" is the 32KHz oscillator, driving the RTC, the LCD, the watchdog timer, and the PLL for the higher operating speeds (up to around 40 MHz). The 2MHz oscillator is an alternate clock source for the CPU. Since the two are independent, they can be compared. There is a special register in the chip counting the clock pulses of the 2MHz oscillator with respect to 16 pulses of the slow clock. The interpretation of this value depends very much on the confidence in the precision of the two oscillators. I assume that HP believes the 2MHz clock to be more precise than the slow clock and tries to guess the actual slow clock speed from there.


I believe this has been talked about before when looking at the 20/30b menu.

Basically, it allows some adjustment of the clock speed if the serial communications are behaving unreliably. This was needed on a few early models that has some issues and needed some "fine tuning" to get talking properly.

I don't think it has been needed/used in quite a while.

You can give it a new value by giving it a number and pressing INPUT (I think you must load the value into the INPUT register before going into the menu unless I am remembering incorrectly).



So the value seen here is not a measured but a "believed to be correct" value settable by a user. Are you using the DBGU pins in production machines?


3 - No other checksum. Basically the numbers before and after should match (on is the calculated number, the other is the number put in the flash to compare against). We've never seen a non-match. . .

5 - Only version. Hopefully forever since that means there have not been any bugs reported! :-)

I am pretty sure I mentioned it at HHC last year, but maybe not. If nobody gets it by then I will consider mentioning it again. ;-)

Glad to hear you like it! I think its a fine unit, but of course I am biased. I did essentially the whole thing (taking 20/30b code as a base), so this qualifies as my first calc.



While I'm sure that this is not the easter egg....

When starting the LCD test, if the first key you press is [=] then when the test ends it will keep looping by itself after the 11th key-press. If the first key you press is not [=] the test will terminate. Also, there appears to be a tiny bug in this test... if you end the auto-loop (by pressing any key) when the up and down shift arrows are displayed they will remain on the screen when the test ends.

I'm wondering if this behavior is indicative that the easter egg is in the LCD test routine since the code is clearly paying attention to what keys are pressed and when.


Nah, not a bug (re:indicators). Just a "I didn't bother to fix it when I found it" as it doesn't really cause any problems and isn't meant to be used outside the factory.



Not a bug

because you didn't bother to fix it as it wasn't going to get stepped on by regular users.

This can be a useful definition when testifying before congress, but I don't think it's going to work with the crowd reading this forum. :)



A minor display issue on something that isn't used (officially) by anyone except the factory, corrects itself next time you do anything, and doesn't cause any sort of issue with calculations? Sounds like not a bug to me. . .




I was just kidding, you did a great job on the 10bii+. I'm surprised how much I like the 10bii+, I never thought I'd like any calculator that wasn't RPN nor programmable.

It's hard to imagine any calculator with this many functions being totally bug free, but I've tried really hard and haven't found any.



Yeah, I knew it was friendly ribbing. :-)

Thanks for the kind words. If Cyrille hadn't done a great job on the 20/30b and the underlying math library though, I'm sure it would have been a disaster.



I have to agree with Katie. I am surprised at how much I like this calc. Of course if you were to give us RPN, a wide enter key, and programability in the same basic form factor I would be in heaven. Here's hoping for a 10bIII or a 10bII++ in the not too distant future :-).


I am beginning to fall in love with this little algebraic HP ;-)

Hi hpnut. How do you like the keyboard? Can you descibe it?




I am OK with the keyboard. It does not have the mushy feeling of the HP 20b. It does not feel like the HP 12C keyboard which IMHO is better.

i have many HP models in my collection and most of them feel quite different to press.


The keyboard has two mylar sheets on top of the circuit board. Each sheet has its own dome. When you press the key, it does not have to 'bridge' a connection on the circuit board or anything like that to make an electrical connection, but just make a single point of connection between the two sides. This is mechanically more resistant to small alignment issues and the like.

While it does have more moving parts, it seems to be more resistant to missed key presses and things. The 10bII was produced by two different sources the past few years, and this design proved to be very reliable.

The 12c, 35s, 17bII+, 30b, and graphing calcs all use a metal dome underneath the keys. The feel on them will be more crisp then the mylar as the metal has more snap to it, but the keyboard on the 10bII+ is definitely not bad in any sense of the word.




yesterday, I just got the original HP 10B (made in Singapore, circa 1992 from the serial number).

the keypad is excellent, I would rate it even better than the new HP 10bII+.

I guess HP should "reverse engineer" and look back at the old production materials and methods :-)


hpnut in Malaysia

(HP calculator user since acquiring HP 10C in 1980 in the United Kingdom as a student)


Katie, that's a lot of insight. In wp34s, I reduce the maximum speed to 10 MHz when the voltage drops to 2.4V. I hope that helps. Power measurement is continuous while the system is doing work. I hope that helps, too. I tried to write the flash at 2.4V which worked but the power source was a lab supply. I don't know what happens when aging batteries are used instead.

Edited: 16 May 2011, 1:42 p.m.

Possibly Related Threads…
Thread Author Replies Views Last Post
  10bii Financial Calculator is on launch sale - 0.99$ only John 4 2,008 11-10-2013, 08:02 PM
Last Post: BShoring
  HP-42S ESD self-test Yriarte 2 1,566 10-24-2013, 09:08 AM
Last Post: Yriarte
  So I took a test with my Prime today ... kris223 12 3,512 09-25-2013, 07:19 PM
Last Post: kris223
  Best way to test HP-71B RAM modules (the 32K/96K ones etc.)? Philippe Lasnier 4 2,417 09-25-2013, 09:54 AM
Last Post: Philippe Lasnier
  Concern about Voyager keyboard test Matt Agajanian 2 1,443 08-30-2013, 07:56 PM
Last Post: Matt Agajanian
  10BII LCD Repair K Gobeski 0 971 06-19-2013, 08:58 AM
Last Post: K Gobeski
  any open source HP 10BII emulators? John 15 4,754 06-12-2013, 09:58 AM
Last Post: Kimberly Thompson
  Voyager Self Test Mike (Stgt) 5 1,987 05-16-2013, 03:28 PM
Last Post: Mike (Stgt)
  HP 41 Advantage module test RyanM 6 2,121 02-27-2013, 02:30 AM
Last Post: Angel Martin
  Physics Test Howard Owen 37 8,526 01-24-2013, 03:33 AM
Last Post: Csaba Tizedes (Hungary)

Forum Jump: