Re: [time-nuts] inexpensive, black box, GPS or NTP based TTL time capture?
Attila Kinali wrote:
On Wed, 18 Oct 2017 08:47:58 -0600
Rob Seaman <seaman at noao.edu https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts> wrote:
I’m at the annual planetary sciences meeting (in Provo this year) and several
groups have expressed interest in duplicating our setup (details of FO
converters, Schmitt triggers, etc, omitted) in a “cheap black box” to quote
one fellow. Lots of people contribute productively to NEO observations,
including amateurs and small teams with little funding. Improving their
timekeeping would help keep rocks from falling on you and your family.
What is this black box supposed to do? Just provide a PPS? IRIG-B?
Or does it need to have time-stamping capabilities? If so, how many
channels?
What are your time precision/accuracy requirements?
What how cheap is "cheap"? What is the volume?
Thanks for the quick reply. I should have included the subject line in the message:
"inexpensive, black box, GPS or NTP based TTL time capture”
Telescope domes are filled with equipment, in particular the camera shutter, that can be instrumented to issue a pulse suitable for hardware time capture. We use Meinberg IRIG PCIe cards to trap these and read the timestamps using their library routines under Linux. “Black box” to the person I was talking to meant no IRIG in the dome, no Linux, and no PCIe slot, but rather a self-contained unit that syncs to GPS. When last he implemented such a feature at another telescope he didn’t even have time capture, but rather the device issued a trigger at a specified moment, so the timestamp and the shutter opening were inverted. He also seems to prefer the timestamps be issued on a serial connection, not via library.
Unlike laboratory instrumentation, a telescope environment needs to be both very automated and very forgiving. Money may also be constrained. However, telescopes are also often very flexible and I am willing - no, eager - to consider completely different arrangements of equipment.
So, reliable timestamping of a TTL input is the ulitmate goal. One channel would be sufficient, but multi-channel would not invalidate an option. PPS or IRIG-B (DCLS IEEE-1344) are not required, but might form an intermediate part of the solution. Reference could be GNSS or possibly NTP.
Precision varies, but milliseconds down to microseconds. Accuracy should match the precision, meaning UTC accurate to same. Extra credit for multiple timescale support.
Cheap is what I want to know. I see the previous thread on BB-black and could imagine a solution using the real time capabilities of that for a few hundred bucks, but these are not experimenters, per se. That’s why they want a black box. Volume is one to several, but could imagine a bulk order if savings were significant. Hundreds of dollars might be the price point.
Thanks!
Rob Seaman
University of Arizona
So, if I hear you right, you want a ttl input to trigger a timestamp,
accurate to ms or µs to a file, correct?
Bob
On Wed, Oct 18, 2017 at 1:33 PM, Rob Seaman seaman@noao.edu wrote:
Attila Kinali wrote:
On Wed, 18 Oct 2017 08:47:58 -0600
Rob Seaman <seaman at noao.edu <https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts>> wrote:
I’m at the annual planetary sciences meeting (in Provo this year) and
several
groups have expressed interest in duplicating our setup (details of FO
converters, Schmitt triggers, etc, omitted) in a “cheap black box” to
quote
one fellow. Lots of people contribute productively to NEO observations,
including amateurs and small teams with little funding. Improving their
timekeeping would help keep rocks from falling on you and your family.
What is this black box supposed to do? Just provide a PPS? IRIG-B?
Or does it need to have time-stamping capabilities? If so, how many
channels?
What are your time precision/accuracy requirements?
What how cheap is "cheap"? What is the volume?
Thanks for the quick reply. I should have included the subject line in the
message:
"inexpensive, black box, GPS or NTP based TTL time capture”
Telescope domes are filled with equipment, in particular the camera
shutter, that can be instrumented to issue a pulse suitable for hardware
time capture. We use Meinberg IRIG PCIe cards to trap these and read the
timestamps using their library routines under Linux. “Black box” to the
person I was talking to meant no IRIG in the dome, no Linux, and no PCIe
slot, but rather a self-contained unit that syncs to GPS. When last he
implemented such a feature at another telescope he didn’t even have time
capture, but rather the device issued a trigger at a specified moment, so
the timestamp and the shutter opening were inverted. He also seems to
prefer the timestamps be issued on a serial connection, not via library.
Unlike laboratory instrumentation, a telescope environment needs to be
both very automated and very forgiving. Money may also be constrained.
However, telescopes are also often very flexible and I am willing - no,
eager - to consider completely different arrangements of equipment.
So, reliable timestamping of a TTL input is the ulitmate goal. One channel
would be sufficient, but multi-channel would not invalidate an option. PPS
or IRIG-B (DCLS IEEE-1344) are not required, but might form an intermediate
part of the solution. Reference could be GNSS or possibly NTP.
Precision varies, but milliseconds down to microseconds. Accuracy should
match the precision, meaning UTC accurate to same. Extra credit for
multiple timescale support.
Cheap is what I want to know. I see the previous thread on BB-black and
could imagine a solution using the real time capabilities of that for a few
hundred bucks, but these are not experimenters, per se. That’s why they
want a black box. Volume is one to several, but could imagine a bulk order
if savings were significant. Hundreds of dollars might be the price point.
Thanks!
Rob Seaman
University of Arizona
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts
and follow the instructions there.
Yo Rob!
On Wed, 18 Oct 2017 11:33:28 -0600
Rob Seaman seaman@noao.edu wrote:
That’s why they want a black box. Volume is one to several, but could
imagine a bulk order if savings were significant. Hundreds of dollars
might be the price point.
For under a $100 you could get a Raspberry Pi, a GPS HAT, and connect
your input to a GPIO pin. Configure ntpd to log the real PPS and the
input as another 'PPS'.
RGDS
GARY
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
gem@rellim.com Tel:+1 541 382 8588
Veritas liberabit vos. -- Quid est veritas?
"If you can’t measure it, you can’t improve it." - Lord Kelvin
picPET -- Precision Event Timer http://www.leapsecond.com/pic/picpet.htm
On Wed, Oct 18, 2017 at 1:33 PM, Rob Seaman seaman@noao.edu wrote:
Attila Kinali wrote:
On Wed, 18 Oct 2017 08:47:58 -0600
Rob Seaman <seaman at noao.edu <https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts>> wrote:
I’m at the annual planetary sciences meeting (in Provo this year) and
several
groups have expressed interest in duplicating our setup (details of FO
converters, Schmitt triggers, etc, omitted) in a “cheap black box” to
quote
one fellow. Lots of people contribute productively to NEO observations,
including amateurs and small teams with little funding. Improving their
timekeeping would help keep rocks from falling on you and your family.
What is this black box supposed to do? Just provide a PPS? IRIG-B?
Or does it need to have time-stamping capabilities? If so, how many
channels?
What are your time precision/accuracy requirements?
What how cheap is "cheap"? What is the volume?
Thanks for the quick reply. I should have included the subject line in the
message:
"inexpensive, black box, GPS or NTP based TTL time capture”
Telescope domes are filled with equipment, in particular the camera
shutter, that can be instrumented to issue a pulse suitable for hardware
time capture. We use Meinberg IRIG PCIe cards to trap these and read the
timestamps using their library routines under Linux. “Black box” to the
person I was talking to meant no IRIG in the dome, no Linux, and no PCIe
slot, but rather a self-contained unit that syncs to GPS. When last he
implemented such a feature at another telescope he didn’t even have time
capture, but rather the device issued a trigger at a specified moment, so
the timestamp and the shutter opening were inverted. He also seems to
prefer the timestamps be issued on a serial connection, not via library.
Unlike laboratory instrumentation, a telescope environment needs to be
both very automated and very forgiving. Money may also be constrained.
However, telescopes are also often very flexible and I am willing - no,
eager - to consider completely different arrangements of equipment.
So, reliable timestamping of a TTL input is the ulitmate goal. One channel
would be sufficient, but multi-channel would not invalidate an option. PPS
or IRIG-B (DCLS IEEE-1344) are not required, but might form an intermediate
part of the solution. Reference could be GNSS or possibly NTP.
Precision varies, but milliseconds down to microseconds. Accuracy should
match the precision, meaning UTC accurate to same. Extra credit for
multiple timescale support.
Cheap is what I want to know. I see the previous thread on BB-black and
could imagine a solution using the real time capabilities of that for a few
hundred bucks, but these are not experimenters, per se. That’s why they
want a black box. Volume is one to several, but could imagine a bulk order
if savings were significant. Hundreds of dollars might be the price point.
Thanks!
Rob Seaman
University of Arizona
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts
and follow the instructions there.
picPET -- Precision Event Timer http://www.leapsecond.com/pic/picpet.htm
Thanks for the plug. Yes -- it does sub-microsecond relative time-stamping simply: one Schmitt-trigger TTL input pin and one RS232 output pin. But -- it's a $1 chip, not a black box. And the timestamps aren't UTC either. So I think it misses some requirements that Rob mentioned.
Another alternative is to use UBX-TIM-TM2 messages from ublox timing receivers.
Still, as much as I like ublox or picPET or John's high-res Arduino-based TICC or homebrew NTP / Pi project, I suspect they are not the turn-key commercial / industrial black box that Rob is looking for.
/tvb