Page 1 of 8

DM42- clock losing time

Posted: Fri Oct 26, 2018 6:44 pm
by cappy
I hadn't used my DM42 for a few months. When I pulled it out of the drawer, turned it on and compared the time on the DM-42 display,
to my comptuer's time (which is the reference I used when I first set the time on the DM42), the time on the DM42 was off by several minutes.
I don't remember if it was fast or slow now, as this happened about a month ago and I forgot to post a message.

Is the clock time shown updated in software, or is there a HW clock?
I could see a software clock loosing several minutes over a few months, but not a HW-based clock.
In the time span of a year, the clock would probably off by an hour or more, which is very bad.

Re: DM42- clock losing time

Posted: Sat Oct 27, 2018 3:48 am
by Walter
IIRC, there were several complaints about the precision of this "Swiss" clock many months ago. They were answered that this is no quartz-controlled clock. So you experience what your grandparents did. That won't help but may make you feel better.

Re: DM42- clock losing time

Posted: Sat Oct 27, 2018 4:43 am
by cappy
Thanks. It doesn’t make me feel better, but at least it’s confirmation it’s not just my DM-42 experiencing this.
It’s a shame they didn’t add a crystal. The additional cost would have been worth it.

Re: DM42- clock losing time

Posted: Sun Oct 28, 2018 2:00 pm
by Rmollov
cappy wrote:
Sat Oct 27, 2018 4:43 am
Thanks. It doesn’t make me feel better, but at least it’s confirmation it’s not just my DM-42 experiencing this.
It’s a shame they didn’t add a crystal. The additional cost would have been worth it.
What about colour screen, GPS, wifi and stereo speakers?

Re: DM42- clock losing time

Posted: Sun Oct 28, 2018 7:25 pm
by toml_12953
Rmollov wrote:
Sun Oct 28, 2018 2:00 pm
cappy wrote:
Sat Oct 27, 2018 4:43 am
Thanks. It doesn’t make me feel better, but at least it’s confirmation it’s not just my DM-42 experiencing this.
It’s a shame they didn’t add a crystal. The additional cost would have been worth it.
What about colour screen, GPS, wifi and stereo speakers?
Well why not? I know you're being facetious but here goes: Color would need too many changes in the firmware to support. How many people would use GPS? Now Wi-Fi could be very useful! For stereo, again too many changes in the firmware. Since time functions are included in the DM42, a crystal would be very useful to make them accurate. Maybe there's a way to add one ourselves as there was in the HP-45. That calculator had undocumented timer and stopwatch functions but no crystal so the accuracy was terrible. Someone came up with a way to add a crystal and then the 45 became truly useful at timing events. The incremental cost of adding a crystal is negligible while the benefit is large. The cost/benefit ratio approaches zero.

Re: DM42- clock losing time

Posted: Mon Oct 29, 2018 4:49 am
by Walter
toml_12953 wrote:
Sun Oct 28, 2018 7:25 pm
Since time functions are included in the DM42, a crystal would be very useful to make them accurate. Maybe there's a way to add one ourselves as there was in the HP-45. That calculator had undocumented timer and stopwatch functions but no crystal so the accuracy was terrible. Someone came up with a way to add a crystal and then the 45 became truly useful at timing events. The incremental cost of adding a crystal is negligible while the benefit is large. The cost/benefit ratio approaches zero.
I concur. Does anybody know the pins where a quartz could be added to make precise timing possible?

Re: DM42- clock losing time

Posted: Mon Oct 29, 2018 12:13 pm
by Nobody
Is it possible to realize a software-based time correction on DM42?

HP-41CX has a very good RTC. My 41CX loses only 0.2 sec a day with its default setting. Furthermore, it has functions to correct its time according to its actual accumulated time errors. As a result, the time error of my HP-41CX is less than 1.0 sec a year.
https://i.imgur.com/dXP4tCt.jpg

My HP 48GX is a late model (made in 1993) and its RTC is not so good. It loses 1.6 – 2.0 sec a day. Unfortunately, the model has no clock correction function as HP-41CX has. I found a very useful application (HP CLK) and I installed to my 48GX and run the application on background. So, now my HP 48GX does not lose or gain more than 1.0 sec for more than 6 months.
Ref. http://www.hpmuseum.org/cgi-sys/cgiwrap ... read=52748

Re: DM42- clock losing time

Posted: Mon Oct 29, 2018 12:50 pm
by dlachieze
Walter wrote:
Mon Oct 29, 2018 4:49 am
I concur. Does anybody know the pins where a quartz could be added to make precise timing possible?
As far as I can see on my DM42 there is already a 32.768KHz quartz (or oscillator) loaded (Q1). So I think that improving the accuracy may be done in software using some of the techniques described in this application note.

Re: DM42- clock losing time

Posted: Wed Jan 09, 2019 5:32 pm
by cappy
Walter wrote:
Mon Oct 29, 2018 4:49 am
toml_12953 wrote:
Sun Oct 28, 2018 7:25 pm
Since time functions are included in the DM42, a crystal would be very useful to make them accurate. Maybe there's a way to add one ourselves as there was in the HP-45. That calculator had undocumented timer and stopwatch functions but no crystal so the accuracy was terrible. Someone came up with a way to add a crystal and then the 45 became truly useful at timing events. The incremental cost of adding a crystal is negligible while the benefit is large. The cost/benefit ratio approaches zero.
I concur. Does anybody know the pins where a quartz could be added to make precise timing possible?
The micrcontroller or CPU in the DM42 already has avery precise clock. It should be more than enough for a RTC that is far more accurate
than what we have now.

Re: DM42- clock losing time

Posted: Thu May 09, 2019 4:39 am
by CNCJerry
When I bought my DM42, I could have sworn there was a note about an RTC or accurate clock. I hadn't checked mine until I had to change the firmware to support the printer I was given. I then noticed the clock was off by several minutes. This sort of surprised me. I'm a time-nut so it's not like I would expect it to keep i sync with my Cesium beam or Rb, etc. But still, this seems a little more off.

Jerry