Domestic time pulse transmission
I'm wondering if the group wisdom could help me work out a good way to send
a ~1pps pulse from a mechanical clock to a time monitoring system? This is
for a pendulum clock project where the clock is electronically controlled
and I want to measure its timekeeping remotely to avoid having to build the
time reference and logging R-Pi into the case. I do not want to have to run
a cable from the room where the clock will be to the place where the logging
is done, so the options are radio or using the mains wiring (the clock will
be mains powered with battery backup). Obviousy there are options such as
the cheap 433MHz transceivers that are available which I am investigating,
but does anyone here have any experience of techniques that work (or, just
as importantly, don't work!) please?
John.
On 2023-08-04 05:47 AM, john.haine--- via time-nuts wrote:
measure its timekeeping remotely
How "remotely"? To the next room, the next building, the next continent,
or the next planet? The answer will make a huge difference in terms of
what's practical.
Steve Hendrix
Hi, John.
A co-worker of mine measured the 1pps from a grandfather clock by placing a Raspberry Pi nearby and using a microphone to “listen” for the sound the pendulum makes when it emits the “pulse”. He tweaked the code until it was spot on to the pendulum swing and then could route the pulse from there using TCP/IP from that Raspberry Pi to anywhere.
-Mark
On Aug 4, 2023, at 9:54 AM, john.haine--- via time-nuts time-nuts@lists.febo.com wrote:
I'm wondering if the group wisdom could help me work out a good way to send
a ~1pps pulse from a mechanical clock to a time monitoring system? This is
for a pendulum clock project where the clock is electronically controlled
and I want to measure its timekeeping remotely to avoid having to build the
time reference and logging R-Pi into the case. I do not want to have to run
a cable from the room where the clock will be to the place where the logging
is done, so the options are radio or using the mains wiring (the clock will
be mains powered with battery backup). Obviousy there are options such as
the cheap 433MHz transceivers that are available which I am investigating,
but does anyone here have any experience of techniques that work (or, just
as importantly, don't work!) please?
John.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
John,
Depending on the details, there is a few different options.
One is naturally to steer something that measures and which returns
measures as time-stamps over say WiFi.
A more direct approach can work in some cases, where you beam an RF
frequence (which naturally needs to be a frequency allowable for the
purpose in your country) onto the pendulum and then measure on a
receiver either reflection or transmission. This is directly comparable
to using a laserbeam in similar fashion, and that is an option too.
Probably this approach will not be possible, because of how other things
like doors, humans and house anminals will disturb things, so you need
to revert back to the first approach, steer something so you can use
that to measure.
You end up having to put some logic inside that. Be it an ESP32,
Beaglebone or RPi. I would probably go with the RPi due to my higher
familiarity with it. If you can put a GPS receiver in there, you can use
it's PPS as timing reference and it will be far better and fairly
straight-forward than anything you do. You will probably not have to go
all the way to TICC for time-stamping, but if you feel extra time-nutty,
that is where you will be heading.
I have considered doing something simliar.
Cheers,
Magnus
On 2023-08-04 11:47, john.haine--- via time-nuts wrote:
I'm wondering if the group wisdom could help me work out a good way to send
a ~1pps pulse from a mechanical clock to a time monitoring system? This is
for a pendulum clock project where the clock is electronically controlled
and I want to measure its timekeeping remotely to avoid having to build the
time reference and logging R-Pi into the case. I do not want to have to run
a cable from the room where the clock will be to the place where the logging
is done, so the options are radio or using the mains wiring (the clock will
be mains powered with battery backup). Obviousy there are options such as
the cheap 433MHz transceivers that are available which I am investigating,
but does anyone here have any experience of techniques that work (or, just
as importantly, don't work!) please?
John.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
On 8/4/23 2:47 AM, john.haine--- via time-nuts wrote:
I'm wondering if the group wisdom could help me work out a good way to send
a ~1pps pulse from a mechanical clock to a time monitoring system? This is
for a pendulum clock project where the clock is electronically controlled
and I want to measure its timekeeping remotely to avoid having to build the
time reference and logging R-Pi into the case. I do not want to have to run
a cable from the room where the clock will be to the place where the logging
is done, so the options are radio or using the mains wiring (the clock will
be mains powered with battery backup). Obviousy there are options such as
the cheap 433MHz transceivers that are available which I am investigating,
but does anyone here have any experience of techniques that work (or, just
as importantly, don't work!) please?
what kind of timing uncertainty is acceptable? milliseconds?
microseconds? femtoseconds? (the latter will be difficult)
Do you mind having "missing ticks"?
The 433 MHz remote pushbutton/doorbell works, there will be a sort of
fixed offset between "contact closure" on one end and the other, and
sporadic interference/jamming from other devices in the band will be an
issue, hence the missing/delayed tick question.
Think garage door opener.
Many thanks for all the comments. To be clearer about the requirements,
this would be just for transmission around an averaged sized detached house
and maybe through 2 or 3 blockwork walls. I guess my question was more if
anyone has any experience and what order of accuracy was possible. If
course interference etc may also lead to missing pulses and so the other
thread on "ADEV with gaps" might be relevant! Anyway I have ordered a Tx/Rx
pair and will see how they perform, I do not have high expectations!
John.
I have a Bulle electrically powered pendulum clock. It has a contact
mounted to pulse a coil at a certain point of the pendulum's swing. I've
obtained a good pulse by monitoring the current in the coil with a current
probe, though voltage methods or even a pickup coil near the solenoid would
work.
On Fri, Aug 4, 2023 at 9:12 PM Lux, Jim via time-nuts <
time-nuts@lists.febo.com> wrote:
On 8/4/23 2:47 AM, john.haine--- via time-nuts wrote:
I'm wondering if the group wisdom could help me work out a good way to
send
a ~1pps pulse from a mechanical clock to a time monitoring system? This
is
for a pendulum clock project where the clock is electronically controlled
and I want to measure its timekeeping remotely to avoid having to build
the
time reference and logging R-Pi into the case. I do not want to have to
run
a cable from the room where the clock will be to the place where the
logging
is done, so the options are radio or using the mains wiring (the clock
will
be mains powered with battery backup). Obviousy there are options such
as
the cheap 433MHz transceivers that are available which I am
investigating,
but does anyone here have any experience of techniques that work (or,
just
as importantly, don't work!) please?
what kind of timing uncertainty is acceptable? milliseconds?
microseconds? femtoseconds? (the latter will be difficult)
Do you mind having "missing ticks"?
The 433 MHz remote pushbutton/doorbell works, there will be a sort of
fixed offset between "contact closure" on one end and the other, and
sporadic interference/jamming from other devices in the band will be an
issue, hence the missing/delayed tick question.
Think garage door opener.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com