[time-nuts] Comparing PPS from 2 GPS units
davidwhess at gmail.com
Tue Dec 18 01:39:15 UTC 2012
I wish there was an source for helically wound shielded differential
transmission line like the type used in later analog oscilloscopes.
The only place I know where to find it is oscilloscope part mules.
Essentially it was transmission line with a ridiculously low velocity
factor. It is great for building instant digital delay lines up to
the low 10s of nanoseconds range in a small space.
On Mon, 17 Dec 2012 20:04:15 -0500, Bob Camp <lists at rtty.us> wrote:
>The nice thing about a spool of coax is that it's got a bit of thermal mass. It will average out a lot of minor temperature ups and downs.
>On Dec 17, 2012, at 4:34 PM, Hal Murray <hmurray at megapathdsl.net> wrote:
>> lists at rtty.us said:
>>> If you are trying to set up say a 1 us delay, you will get ~ 50 ps per
>>> degree C in your delay. That's a lot .....
>> A while ago, tvb at LeapSecond.com said:
>>> A long delay cable is fine too. If these are timing receivers you probably
>>> don't need more than 100 ns of delay, once you figure out which receiver is
>>> ahead of the other. The cable tempco is low enough not to worry about.
>> 100 ns is 50-100 feet. That's a reasonable length to work with. But I was
>> curious about the temperature coefficient. Google found this:
>> which says:
>> # Belden 8240 (solid) shows a temperature coefficient of around
>> -0.252ps/m/deg in a temperature range between -20 and 30 deg. The coefficient
>> becomes steeper beyond 30 deg.
>> # Belden 8219 (foam) shows a larger temperature coefficient of around -0.352
>> ps/m/deg than that of 8240 in the similar temperature range. The coefficient
>> becomes steeper beyond 30 deg, but less steeper than that of 8240.
>> # Fujikura RG58-A/U shows the smallest temperature coefficient of around
>> -0.152 ps/m/deg, but in a narrow temperature range between -10 and 20 deg.
>> The coefficient beyond 20 deg is much steeper than the others.
>> To pick round numbers, 30 meters and 3 C and 0.25 ps/m/C gives 25 ps.
More information about the time-nuts