On Wed, Oct 19, 2016 at 6:15 AM, Dave Martindale dave.martindale@gmail.com
wrote:

--

Chris Albertson
Redondo Beach, California

Even if the micro controller had to directly control the LEDs on the clock
your suggestion would be not only simpler bot several orders of magnitude
more accurate.    I'd guess that a first attempt to build a 32KHz GPSDO
would not be perfect but the PPS signal is effectively dead-in perfect in
this use case.

The disadvantage of using there GPS directly to increment the clock is you
loose hold over.  The clock stops when GPS goes away.

A GPSDO has a "flywheel effect" and can continue to run on it's own if the
GPS is interrupted.  When the GPS signal returns it can slowly recover.
But do you care?

One other unleaded idea, can you control the frequency of a 32K crystal by
controlling its temperature?  If you can then that might be easier then
using a diode tuner.  It is easy t build a heater, epoxy a resistor to the
crystal.

On Wed, Oct 19, 2016 at 6:15 AM, Dave Martindale dave.martindale@gmail.com
wrote:

--

Chris Albertson
Redondo Beach, California

Hi Lee:

32768 can easily be divided down to drive the clock.  So why not bypass the 32768 and drive the clock directly from the
1 PPS you now have.
http://www.prc68.com/I/QuartzClk.shtml

--
Have Fun,

Brooke Clarke
http://www.PRC68.com
http://www.end2partygovernment.com/2012Issues.html
The lesser of evils is still evil.

-------- Original Message --------

Hi Brooke,

Right, but that trick only works with analog stepper motor clocks. OP has a "big digital clock" with 8-bit cpu and 32 kHz xtal. He didn't mention the make/model of digital clock but in my experience very few commodity clocks actually accept a 1PPS input. These clocks use 32 kHz:

1. to drive the MCU which computes day / date / hh:mm:ss, or manages alarms
2. to maintain timekeeping
3. to multiplex digits of the LED / LCD display (e.g., at 128 to 1024 Hz)
4. to create the short bipolar stepper motor pulse (e.g., 1/32 kHz * 512 = 1/64 s = 15.6 ms).
5. to create the sound for the alarm/buzzer (some PWM based on 32 kHz)

The problem is that all these functions are usually integrated into one chip or even raw die/epoxy as in COB (Chip On Board). When hacking these sort of clocks it is often impossible to separate 32 kHz frequency features from the 1 Hz timing feature.

So when your goal is to improve timekeeping accuracy in these self-contained digital clocks it's usually easier and less invasive to make the clock use your precise 32 kHz signal instead of its own cheap xtal. You almost always have access to the xtal, but rarely access inside the MCU.

Note that you don't even need to unsolder the xtal -- you can "jam" the existing signal with an external 32 kHz sine or square wave applied to the XI pin (xtal in) of the MCU. Your external GPSDO/32kHz signal will "pull" the cheap xtal for free. Best yet, if your external signal goes away the clock keeps running using its own xtal without skipping a beat, like getting hold-over for free.

For a "no solder" or "no wires" solution, I have also tried to acoustically discipline a tuning fork xtal with an GPS-based 32 kHz signal and ultrasonic transducer. Poor results. I think I needed better coupling between the transducer and the xtal tuning fork. But in theory it should work. Plus it would keep small mammals and insects away from your clock.

/tvb

----- Original Message -----
From: "Brooke Clarke" brooke@pacific.net
To: "Lee - N2LEE" lee@n2lee.com; "Discussion of precise time and frequency measurement" time-nuts@febo.com
Sent: Wednesday, October 19, 2016 12:18 PM
Subject: Re: [time-nuts] 1PPS to 32.768 khz

Hi Lee:

32768 can easily be divided down to drive the clock.  So why not bypass the 32768 and drive the clock directly from the
1 PPS you now have.
http://www.prc68.com/I/QuartzClk.shtml

--
Have Fun,

Brooke Clarke
http://www.PRC68.com
http://www.end2partygovernment.com/2012Issues.html
The lesser of evils is still evil.

-------- Original Message --------

Hi

As has been already mentioned, a lot depends on what you have. The drop dead cheapest way to do it:

Start with an MCU with an internal oscillator. There are lots to pick between. Which sort really does not matter.
For example, I’ll use one that starts at 4 MHz.

Divide the 4 MHz down to 32,768, or as close as you can get. The common clocks aren’t going to divide
straight to 32768 so you will need to do two divides. One will be a bit fast, the other a bit slow. You flip between the
two in a fixed pattern to get the result to average out correctly.

Next take the same 4 MHz and run a counter /timer off of it. Capture the pps edge with the timer. It will drift
a bit since the clock in the MCU is not perfect. Based on the drift, modify your dividers to correct the outcome.
If you are “good” to 1 part in 32,768 in each second, that’s close enough for a wall clock. You will have no net
error long term if you do it right.

That’s all a sub $1 solution….

Bob

I think the way you drive the clock with a 1PPS signal is to trash the
existing controller.  Then get some Arduino-like board and connect each LED
to one of the board's output pins and the PPS to one of the boards
interrupt pins.  You replace the entire controller.  They don't cost.
Cheaper than building a GPSDO

On Wed, Oct 19, 2016 at 1:51 PM, Tom Van Baak tvb@leapsecond.com wrote:

--

Chris Albertson
Redondo Beach, California

I was thinking 32.768kHz VCXO and phase detector to make a simple
analog PLL.  I found a datasheet for a suitable VCXO and assuming a
total error of 20ppm, it would only need to be divided by 2 to prevent
locking to the wrong frequency making an analog PLL pretty simple.
Safer to divide by 4 or 16 of course.

But I wonder if a microcontroller using a 32.768kHz external clock
could be the phase detector itself.  Strobe the microcontroller ADCS
from the 1 PPS to sample the sine 32.768kHz clock, simmer, and serve.
Or just count cycles to make a FLL.

The most annoying thing about using a varactor for tuning is that they
are not amendable to low drive voltages.

On Wed, 19 Oct 2016 18:14:32 -0400, you wrote:

I think the suggestion to simply blast out 32768 pulses once per second at
a fast rate of (say) 40kHz is the simplest yet and VERY close to what the
OP asked for.  This makes the average frequency near perfect but of cours
with HORRIBLE phase noise but the OPonly cares about average frequency.

It would be VERY easy to program a small micro controller to output 32768
pulses on an output pin whenever a PPS is detected on an input pin.    I
could implement this in a bare 8-pin AVR chip in about 15 minutes.    The
chip would only need 5 volts power and ground.  The 32KHz signal would have
very poor short term characteristics but who cares?

I was about to write that is is theoretically impossible to create 32kHz
from 1Hz (PPS) and I stick by that.  (but notice in the above we don't
create 32kHz  we just output a fixed number of pulses that have some
imprecise timing then wait for the nest second "tick".

Back to creating 32K for 1Hz.  It is impossible, even in theory.  The best
you can do is output a signal you hope is very close to 32kHz, measure it
relative to the 1PPS reference then adjust it to make it closer to 32kHz
and keep on adjusting it every second.  It will never be perfect.

But the crude 8-pin AVR chip would be as good as at the 1PPS reference,
better on average then any GPSDO.

On Wed, Oct 19, 2016 at 8:07 PM, David davidwhess@gmail.com wrote:

--

Chris Albertson
Redondo Beach, California

Hi

One problem with a PLL and a 1 Hz input are the values of components
you get in the loop. The other issue is the cost of the VCXO that will get
you to 32,768 KHz. The PLL as described by the OP would need the 1 pps
divided by 2 with a lot of PLL chips. You now are locking 32768 to 0.5 Hz.

It gets complicated pretty fast.

Bob