[time-nuts] moon bounce for synchronization
eb4apl at gmail.com
Sun Jan 31 11:32:36 EST 2016
We were not measuring frequency but clock offset. The frequency offset
was estimated from the daily clock measurements.
If I remember correctly we got .1 uS accuracy.
We had several Rubidium, Cesium and later a H-maser frequency standards
and we kept continuous recordings (strip charts) of the phase differences.
The transmitter precompensation kept care of combined Doppler (Moon
motion and Earth rotation, which is the main component). We operated at
X band, the transmitting antenna had 85 ft or maybe less, I don't know
exactly and the transmitter probably had 20 Kw.
By contrast the receiving antenna was small, I found at a picture
one that looks the same, at least the dish and pedestal are almost
I forgot to say that this system was abandoned in favor of VLBI which
gave more accuracy and a lot more useful info. Both systems had the
drawback that the monitoring was not continuous and needed a lot of
resources at both ends. By contrast Loran-C and GPS used existing
infrastructure that was continuously available. BTW, I don't know how
Loran-C stations were synchronized, this came granted by the Navy.
We also used once a traveling clock, it was an HP cesium, the one with a
Patek-Philipe clock on the front. The crew who operated it told us funny
stories about the clock traveling on commercial jet seats and afraid
passengers asking about this bomb looking device and being told that it
was some kind of "atomic" thing.
On 31/01/2016 a las 2:21, jimlux wrote:
> On 1/30/16 4:27 PM, Hal Murray wrote:
>> eb4apl at gmail.com said:
>>> Back to the Control Room you contact the transmitting station (I
>>> think it
>>> was DSS12) by voice to insure that they have the station manned and
>>> transmitting, and began to operate the "thing". The transmission were
>>> specific for each receiving station, because all the complex
>>> processing was
>>> done at the transmission end: the transmitting equipment accounted
>>> for the
>>> instantaneous round trip distance between the transmitter and the
>>> via the moon and continuously adjusted the modulating code "early"
>>> in order
>>> to to be received on time. The equipment also introduced a one
>>> shift each second. The receiver had a correlator whose output went
>>> to an HP
>>> strip chart recorder which draw the correlator output in one
>>> channel and a
>>> PPS with a minute mark in the second.
>> What sort of frequency accuracy were you after?
>> Did somebody have to correct for the Doppler due to the rotation of
>> the Earth?
> I'm pretty sure they did. Ignacio can say for sure.
> We precompensate on transmit to get the signal to arrive at the
> spacecraft at its "best lock frequency". Or, actually slightly off,
> then we ramp through the BLF so the receiver can acquire the carrier.
> The receiver bandwidth might be as narrow as a few 10s of Hz, so you
> don't want to be too far off or ramp too fast.
> On S-band (2-2.3GHz), the earth rotation Doppler is about 2-3kHz.
> (0.5km/sec), depending on latitude and where the Moon is in the sky.
> If you're doing the Moon, you put its motion via a SPK file into the
> algorithm. In 1970, that would have been a bit more tedious<grin>.
> With the moon, it depends on what part of the moon you're aiming at, too.
> Last year (March 3rd), I was doing an experiment bouncing a DSN signal
> from Goldstone off the moon and receiving it at JPL. As I recall, the
> "spot" on the moon was about 800km in diameter. For what it's worth,
> you don't need a particularly good receiver when your transmitter is
> 20 kilowatts into a 34m antenna (DSS 24).
> Since our transmit and receive site were pretty close together, the
> Doppler of earth motion wasn't much, but there was Lunar Doppler, on
> top of that, there's libration.
> Chuck Counselman did a bunch of work with VLBI type techniques using
> various ground stations around the world to determine where things
> (like the lunar rover) were on the surface of the moon.
El software de antivirus Avast ha analizado este correo electrónico en busca de virus.
More information about the time-nuts