[ntp:questions] Re: reasonable price for gps time source (Garmin GPS 17-HVS?)

Moe Trin ibuprofin at painkiller.example.tld
Fri Dec 16 02:29:51 UTC 2005


On Thu, 15 Dec 2005, in the Usenet newsgroup comp.protocols.time.ntp, in article
<-Oednb0hQ5FArDzenZ2dnUVZ_vidnZ2d at megapath.net>, Hal Murray wrote:

>Mine is indoors.  It mostly works but fades out occasionally.
>
>It works a lot better up high.  I haven't moved it around much
>to find the best location.
>
>Looks like about 5% of the samples are bad.  I think rain/clouds
>interfere but I don't have any data to back that up.

Assuming you are limiting the lower angle of sight to ten degrees, and
an absolute bitch of a storm (150mm/hr with cloud tops at 20 KM), the
atmosphere costs you 0.68 decibels, and the torrential rain adds
roughly 0.53 decibels. (Source: CCIR XIIIth Plenary Assembly, Geneva,
1974, Vol. V, Report 233-3.) That extra attenuation is the same as roughly
1.7 feet of 1/4 inch coax, or 4.4 feet of the 1/2 inch coax. (Source:
MIL-HDBK-216, 1962, revised 1965.)   That's another way of saying
'virtually no effect'.   Plain old fashioned clouds have even less
moisture, are not likely to be anywhere near as extensive or thick, and
are therefore even less of a factor. If you can get to a public library,
see the December 12, 2005 issue of "Aviation Week and Space Technology"
and the cover stories (pages 44 to 47), which discusses space based
precipitation radar images of the recent hurricanes in the Gulf.
Several images show (not at the same time) rain laden towers to 53,000
feet, and rainfall near the eye in excess of two inches/hour.

Your problem is going to be the wet roof, and the crud on the roof before
it gets washed away by that cloudburst.  I don't have figures for the
expected attenuation, but it's way more significant than any atmospherics.
A flat roof is going to be _much_ worse, due to accumulations of standing
water.

        Old guy




More information about the questions mailing list