The Museum of HP Calculators

HP Forum Archive 14

[ Return to Index | Top of Index ]

Spice "2 Disease" -- Any Treatment?
Message #1 Posted by Paul Brogger on 30 Dec 2004, 6:39 p.m.

I just got a solderless 31E working (replaced corroded flex conductors between battery compartment and power board with wire) and notice it has the "2 Disease". Every time I press the "2", the minus sign and commas light up.

I remember a discussion of this condition (by Norm, I think) but don't recall a solution ever having been arrived at. Is there one? Should I do anything to prevent further deterioration?

Just curious ...

Re: Spice "2 Disease" -- Any Treatment?
Message #2 Posted by Vieira, Luiz C. (Brazil) on 30 Dec 2004, 7:38 p.m.,
in response to message #1 by Paul Brogger

Hi Paul, guys;

I and Norm had discussed the "2 disease" for some time (in one of our last post about this I think I assumed an unflexible position at that time, and I still believe Norm resents the episode; I apologized by e-mail but seldom saw his posts here after that... I wish it never happened. In fact, I thought about removing these words before posting, but there is no harm, keeping), and we both disagreed in one particular point: I still think it's a firm/soft/hard characteristic and Norm believes it is a malfunction. Norm tells all Spices come without the "2 disease" in essence (he did not use this very expression, I'm just posting my own expressions), meaning there is nothing wrong with the design, and after some specific cause, the "2 disease" appears. I myself believe that it is an intrinsic, harmless "characteristic", that have probably been seen by the HP designers and (for not being harmfull) corrected later.

Whom of us is right, just time (or an HP designer) will tell. All I believe is that a malfunction (ESD?) that could permanently destroy a "blank", cutting switch (MOS transistor) responsible exactly for turnning of a "2" and a sequence of commas/dots is unlikely to be, but I accept the possibility of. Also, the "2 disease" was found in some of all Spices, being them of either soldered or solderless type. I think this is the biggest reason Norm believes it is a malfunction. I think a serial # track was started in order to figure out a possible period in time when the "2 disease" appeared, but I am not quite sure if anyone got thru it.

I had some Spices (thanks to someone I'm supposed not to mention, but I can thank him anyway :-) and I actually gifted a few, trade two or three and kept five with me: one HP31E, one HP34C (my dearest of them), one HP37E, one HP38E and two HP38C (one of them is not working; I used parts of it to repair the others). At least two of them have the "2 disease", and I take it as a funny characteristic: "these babes have personality", I'd say.

So far, I have not been told about a "cure". I guess taht we could "isolate" the "infected" IC by "transplating" parts of identical units (solderless type, preferably) having one the "2 disease" and the other without "infection" till the infected one becomes healed. Unfortunately I have no such two identical units in this condition.


Luiz (Brazil)

Edited: 2 Jan 2005, 1:04 a.m. after one or more responses were posted

Re: Spice "2 Disease" -- Any Treatment?
Message #3 Posted by Randy on 30 Dec 2004, 8:08 p.m.,
in response to message #2 by Vieira, Luiz C. (Brazil)

Most all of the spices I've ever fooled with exhibit the "2" ghosting problem. I long ago concluded it was a firmware/hardware issue, not a failure of any kind. Too many units have the identical behavior to think it is anything other than a unintended feature.

Re: Spice "2 Disease" -- O.K., so I have too much time on my hands . . .
Message #4 Posted by Paul Brogger on 31 Dec 2004, 11:45 a.m.,
in response to message #1 by Paul Brogger

One possibility: I'm running Energizer "e2" batteries rather than NiCds -- sometime today I'll fetch my 34c and swap in the rechargeables and see if the voltage difference changes anything.

Meanwhile, however I'm messing with a 31e at an advertised 3 volts, and . . .

I see two distinct forms of "ghosting". The minus sign flashes, and commas appear, whenever there is a "2" in any position in the display.

It appears that the brightness of the minus sign's flash is directly related to how many 2's are being displayed. That's weird, and it doesn't seem to appear with any other digits, in any combination. But whenever 2's are in the displayed value, the minus sign flashes, and the more 2's in the display, the brighter the flash.

The ghosting appears identically with whatever keypress results in a particular display.

The flashing commas are really weird.

I'm experimenting after setting <f> <FIX> <9>, entering something like "0.020020002" and pressing <+> or <ENTER> repeatedly. With this example, the ghosting display shows briefly "-0.020,0200,02" . Notice the positions of the decimal point and ghosting commas relative to the 2's.

A few more examples with 2's and 0's:

0.2 0 0 2 0 0 0 2 0 -0.2 0 0,2 0 0 0,2 0

0.0 0 2 0 0 0 2 0 0 -0.0 0 2 0,0 0 2 0 0

0.0 0 0 2 0 0 0 0 2 -0.0 0 0 2 0,0 0 0 2

So it seems the ghosting commas appear, relative to the 2nd & subsequent 2's, in postitions analogous to that of the decimal point relative to the first 2.


Then it gets weirder yet.

A few more examples, this time with 2's and other mixed, zero and non-zero digits:

   0.3 0 2 0 0 2 0 0 0
  -0.3 0,2 0 0,2 0 0 0

0.0 1 1 1 2 3 4 5 2 -0.0 1 1,1 2 3 4,5 2

1.0 2 3 4 2 5 6 7 2 -1.0 2,3 4 2,5 6 7 2,

So it seems the ghosting commas actually appear, relative to 2's following any non-zero digit, in postitions analogous to that of the decimal point relative to the first non-zero digit.


I'm sure there's a (perhaps not-so-straightforward) physical explanation for this behavior, but it is a different case!

But now, I've been staring at my calculator display a bit too long -- time to get back to life . . .

[ Return to Index | Top of Index ]

Go back to the main exhibit hall