[time-nuts] RTC DS3231: how well can it be regulated?

[Jim Harman]

The DS3231 trim register sensitivity is speced at 0.1 ppm per count typ at
25 C. The trim range is +/- 127 counts so the fastest you can slew the pps
pulse is about 12.7 ppm or about +/- 1.1 sec per day if my calculations are
correct.

The chip reads the temperature sensor and adjusts the oscillator frequency
in the same size steps so it is effectively a digital TCXO. The resolution
of the temperature sensor readout is 0.25 C  so stabilizing the chip
temperature using that sensor would probably help the overall stability,
especially at extreme ambient temperatures.

I have thought about making a GPSDO using a DS3231 as a digitally
controlled oscillator, with the control based on the time difference
between the GPS and DS3231 pps times, but that is still on the "to-do" list.



On Mon, Dec 30, 2019 at 6:09 AM MLewis <mlewis000 at rogers.com> wrote:

> I was watching the PPS LED on my M8T board vs. the LED I put on the RTC
> DS3231's SQW output enabled for 1 Hz for a PPS. As expected they're not
> aligned, and the drift of that misalignment shows they're obviously not
> at the same frequency.
>
> It got me wondering. Between the DS3231's Aging Trim, insulating and
> perhaps a heater for some temperature control, does anyone know if the
> DS3231's PPS could be synced and stabilized into a reasonable PPS source
> through an opto-isolator?
>
> Might a fibre-optic line be useful as a transmission cable, gapping two
> LEDs functioning as an opto-isolator, thereby providing both isolation
> and high-speed coupling for the PPS to the device that wants to read the
> PPS?
>
> I don't know if the onboard temperature sensor is precise/stable enough
> to be useful for maintaining a stable temperature with a heater. It's
> accuracy is listed as +/- 3 C, but that doesn't speak to the stability
> of the temperature it reports.
>
> The SQW output can be set to 1 Hz, 1.025 kHz, 4.096 kHz or 8.192 kHz.
>
> It also has a 32 kHz output pin. I don't know if that's actually 32 kHz
> or 2^15 = 32,768 Hz.
> Ah, there it is. "outputs a 32.768kHz square-wave signal".
>
> Has anyone done this, or does it seem like it may be worthwhile?
>
> Thanks,
>
> Michael
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>


-- 

--Jim Harman