Threaded Mode | Linear Mode

4ster · 10-12-2016, 01:33 PM

I've had a 41-CV forever. (At least it seems that way.) In the last year I've acquired two working CXs. I have found I use the CXs time functions more than expected, to the point where loosing the date and time in the clock is annoying.

I *think* I am resetting the clock when I pull the battery pack out to change batteries. At least when I notice the clock has been reset, I think back on the machines recent history and a battery change is in the timeline. The manual seems to hint that this *may* happen when the power is interrupted with a battery change, but it implies that if it does the memory would clear as well. I am not loosing continuous memory on a battery change. The CX I use the most holds memory for over 24 hours with the battery out of the case.

This is a rare enough that I am just becoming aware of the pattern. Is loosing time but not memory normal on CXs?

Sylvain Cote · 10-12-2016, 03:37 PM

The HP-82182A Time module and the HP-41CX contains a crystal (or RC circuit?) that keep time and need the batteries to keep the clock working.
Each time you remove the batteries or remove the time module for a C/CV, the clock stop and time is lost.
Sylvain

wojtek · 10-12-2016, 04:28 PM

(10-12-2016 01:33 PM)4ster Wrote: I've had a 41-CV forever. (At least it seems that way.) In the last year I've acquired two working CXs. I have found I use the CXs time functions more than expected, to the point where loosing the date and time in the clock is annoying.

If there is enough room inside the CX body to keep 2 lithium batteries (2 x CR1220 for example) you could try to improve it. Just add a micro switch activated when the main battery is pulled and connecting the backup lithium battery to keep the clock running. When main battery would be inserted the switch would cutt off the lithium battery. Just an idea, don't know if it is possible.

JurgenRo · 10-12-2016, 08:08 PM

(10-12-2016 03:37 PM)Sylvain Cote Wrote: The HP-82182A Time module and the HP-41CX contains a crystal (or RC circuit?) that keep time and need the batteries to keep the clock working.
Each time you remove the batteries or remove the time module for a C/CV, the clock stop and time is lost.
Sylvain

Hi Sylvain,
checked it a minute ago on my HP-41CX Halfnut, S/N > 2700. XEQed "TIME", removed batteries, inserted batteries, XEQed "TIME" again. On my machine, the clock did not stop. Is your machine a Halfnut or a Fullnut?
Juergen

Sylvain Cote · (This post was last modified: 10-12-2016 10:19 PM by Sylvain Cote.)

(10-12-2016 08:08 PM)JurgenRo Wrote:
(10-12-2016 03:37 PM)Sylvain Cote Wrote: The HP-82182A Time module and the HP-41CX contains a crystal (or RC circuit?) that keep time and need the batteries to keep the clock working.
Each time you remove the batteries or remove the time module for a C/CV, the clock stop and time is lost.
Sylvain
Hi Sylvain,
checked it a minute ago on my HP-41CX Halfnut, S/N > 2700. XEQed "TIME", removed batteries, inserted batteries, XEQed "TIME" again. On my machine, the clock did not stop. Is your machine a Halfnut or a Fullnut?
Juergen

Hello Juergen,

You were right to challenge my memory.
I took 8 x HP-41 from my collection, representing the different models, put a time module in each C/CV/CL and then I did the following test ...

20.0000 SETTIME // set time to 8 PM
TIME // to check if the time was correctly configured
OFF
remove battery holder, wait for timer (1, 2, 3, 4 & 5 minutes) expiration then put back the battery holder
TIME // check if the time was kept or lost

Test results:

Code:

Model          ROM      Serial#     Time kept  

-------------  -------  ----------  -----------

HP-41C  FN/TK  GFF      2046A03187   3  minutes 

HP-41C  FN     GFF      2101A05170   5+ minutes 

HP-41C  HN     HFF      2411S42991   2  minutes 

HP-41CV FN     HFF      2407A40051   3  minutes 

HP-41CV HN     HFF      2552S20514   3  minutes 

HP-41CX FN     NFL/DCC  2521S41245   4  minutes 

HP-41CX HN     NFL/DCC  2814S20780   3  minutes 

   41CL FN     V3                   10  seconds

Acronyms: FN=Fullnut / HN=Halfnut / TK=Tallkeys

Sylvain

hth · 10-12-2016, 11:16 PM

I cannot really say for sure why, but I lose clock sometimes. But I have not felt it is related to battery changes, it is more random.

Put it this way, I have never felt it to be all that rock solid. It can work for weeks without problems, but it happens from time to time. I have a feeling that it is more frequent these days as the calculator gets older, but that is just a hunch.

The clock is a bit like the HP-41 is by design, it takes active measures to ensure that the data is correct. If the system suspect something is wrong, clear it. It has a hardware bit as well as a warm start constant and if something seems fishy, you lose the clock. I was not so happy with this behavior in the past, but I think I kind of grown to accept it. The HP48 tries its recover memory and leaves the system half corrupt instead, I do not think that it was a step forward. On the other hand, while the HP48 clock is more solid, its memory system is not.

Garth Wilson · 10-13-2016, 04:49 AM

My cx Halfnut loses the time and date perhaps once a year, but not when removing the battery pack to change the batteries. It's clearly from a less-than-perfect connection reaching the piggy-back board it's on, and then the calc won't turn on either. I beat the side of the calc against the heel of my hand and that restores it, probably by causing some self-cleaning wiping motion of one contact surface against another. To recover more quickly when I don't always remember the right instruction words or their exact spelling, I wrote a 9-step program to prompt me for the time and date and set them, and I assigned it to a key. I use the cx's time capabilities every day. I wouldn't think of being without them.

4ster · 10-13-2016, 03:39 PM

Thanks for the information everyone.

(10-13-2016 04:49 AM)Garth Wilson Wrote: ---Snip--- To recover more quickly when I don't always remember the right instruction words or their exact spelling, I wrote a 9-step program to prompt me for the time and date and set them, and I assigned it to a key. ---Snip---

Thankfully, I don't have to re-set the clock enough to remember the commands as well. I like this idea.

Related question: I can't tease out of the manual where the accuracy factor is stored on the 41. If it is stored on the time module, then it is likely lost when the clock does a self reset, correct? If it is stored in continuous memory then RCLAF in a "clock restore" program would restore the accuracy factor in addition to SETIME, SETDATE and other display preferences.

hth · 10-13-2016, 04:33 PM

The accuracy factor is stored in the timer chip, there are RCLAF and SETAF to recall and set it. Once you get it calculated (after a CORRECT), you can easily save it elsewhere, and restore it in case of a clock loss.

Håkan

JurgenRo · 10-13-2016, 06:41 PM

(10-12-2016 10:05 PM)Sylvain Cote Wrote:
(10-12-2016 08:08 PM)JurgenRo Wrote: Hi Sylvain,
checked it a minute ago on my HP-41CX Halfnut, S/N > 2700. XEQed "TIME", removed batteries, inserted batteries, XEQed "TIME" again. On my machine, the clock did not stop. Is your machine a Halfnut or a Fullnut?
Juergen

Hello Juergen,

You were right to challenge my memory.
I took 8 x HP-41 from my collection, representing the different models, put a time module in each C/CV/CL and then I did the following test ...

20.0000 SETTIME // set time to 8 PM

TIME // to check if the time was correctly configured

OFF

remove battery holder, wait for timer (1, 2, 3, 4 & 5 minutes) expiration then put back the battery holder

TIME // check if the time was kept or lost

Test results:

Code:

Model ROM Serial# Time kept ------------- ------- ---------- ----------- HP-41C FN/TK GFF 2046A03187 3 minutes HP-41C FN GFF 2101A05170 5+ minutes HP-41C HN HFF 2411S42991 2 minutes HP-41CV FN HFF 2407A40051 3 minutes HP-41CV HN HFF 2552S20514 3 minutes HP-41CX FN NFL/DCC 2521S41245 4 minutes HP-41CX HN NFL/DCC 2814S20780 3 minutes 41CL FN V3 10 seconds
Acronyms: FN=Fullnut / HN=Halfnut / TK=Tallkeys

Sylvain

Hi Sylvain,
thank you for the comprehensive Analysis. And, yes, I agree! My HP-41CX does not hold time-information for a longer period than just a few minutes either (if at all). But it does hold the Information long enough for changing batteries without a hurry, though. Thank you again for the investigtion and sharing the results!
All the Best,
Jürgen

Garth Wilson · 10-13-2016, 06:48 PM

(10-13-2016 03:39 PM)4ster Wrote: Thanks for the information everyone.

(10-13-2016 04:49 AM)Garth Wilson Wrote: ---Snip--- To recover more quickly when I don't always remember the right instruction words or their exact spelling, I wrote a 9-step program to prompt me for the time and date and set them, and I assigned it to a key. ---Snip---

Thankfully, I don't have to re-set the clock enough to remember the commands as well. I like this idea.

Related question: I can't tease out of the manual where the accuracy factor is stored on the 41. If it is stored on the time module, then it is likely lost when the clock does a self reset, correct? If it is stored in continuous memory then RCLAF in a "clock restore" program would restore the accuracy factor in addition to SETIME, SETDATE and other display preferences.

Good point. I didn't even think of that, because I've had the good luck that the clock in my particular cx is extremely accurate without using the accuracy factor. When the time changes to and from daylight-savings time twice a year, I just add or subtract an hour and don't have to change the minutes.