[time-nuts] GPS Antenna Feed Line Decision

Bob kb8tq kb8tq at n1k.org
Mon Sep 4 21:26:10 UTC 2017


Hi

A lot depends on just which outfit you buy your attenuators from. There 
certainly *are* outfits out there that supply you just over 20 db RL when
the spec is 20. They also don’t charge very much for their attenuators ….

Bob

> On Sep 4, 2017, at 4:53 PM, jimlux <jimlux at earthlink.net> wrote:
> 
> On 9/4/17 1:18 PM, Wes wrote:
>> 
>> 
>> If these are COTS attenuators, their own return loss is unlikely to be
>> 40 dB.  In fact grabbing an old HP catalog off my bookshelf (I'm dating
>> myself) I see a typical type N attenuator specified as 1.2 VSWR (~21 dB
>> RL).  I went on a quick "shopping" trip looking for an L-band, type N
>> bias tee.  I'll spare you the links, but typically they are also rated
>> at 1.2 VSWR.
>> 
> 
> Be careful, that's the "catalog spec" which means "what we can inexpensively measure"..
> 
> It's like SMA connectors, which are specified at 1.05:1 or 1.1:1 and <0.3dB loss.
> https://www.amphenolrf.com/connectors/sma.html
> 
> 
> In reality, they are a LOT better, it's just that measuring that in a production environment is tough.
> I'd not want to set up a manufacturing test set that measured loss with an uncertainty of 0.01 dB.
> 
> I'll point folks to:
> Jesch's paper in 1976
> 
> http://ieeexplore.ieee.org/document/6312234/
> 
> 
> 
> and then Maury
> https://www.maurymw.com/pdf/datasheets/5A-021.pdf
> 
> Someone at Maury did a paper which I can't find right now where they measured a bunch of SMA connectors over hundreds if not thousands of mate/demate cycles.
> 
> 
> RF cafe has a nice summary
> http://www.rfcafe.com/references/articles/Joe-Cahak/rf-connectors-cables-joe-cahak-6-2014.htm
> 
> 
> 
> 
> 
> 
>> Perhaps NIST, with an unlimited supply of tax money, splurged and
>> manufactured bias tees with >40 dB RL.  Maybe they did the same with the
>> attenuators.  We'll never know because they didn't provide an equipment
>> list or a measurement procedure.  They said nothing about the cable
>> either, other than they started with RG-58 and replaced it with "better"
>> cable.
> 
> But you could probably send an email to the author and ask.  NIST, like JPL, is one of those places where people work forever.  Tom Otoshi, who wrote a report on N connectors in 1963 cited by Maury, above, still works at JPL (I think.. I confess I haven't seen him recently, he might have retired, but he was certainly around in the last 10 years), and given the span of years, that N connector work was probably when he was a just out of school engineer.
> 
> 
> 
> 
> 
> 
> 
>> 
>> A few words about cable, since that is what this discussion is all
>> about.  Cable, regardless of type and manufacturer, has its own RL, also
>> known in that business as Structural Return Loss (SRL) See:
>> https://www.belden.com/docs/upload/hdcarltp.pdf and
>> http://www.keysight.com/upload/cmc_upload/All/E206COMPTEST_METHOD.pdf.
>> 
>> At least the authors admit, "Thus far we have seen little difference in
>> the data."
>> 
>> Wes
>> 
>> 
>> On 9/3/2017 3:02 PM, Bill Byrom wrote:
>>> For precision timing measurements, I would think that there would be
>>> concern about the double reflections of a badly mismatched low loss
>>> transmission line (such as using 75 ohm line in a 50 ohm environment).
>>> The re-reflected signal will act similar to  multipath (as a delayed
>>> aggressor) on all satellite signals equally. The impedance mismatch
>>> delayed reflection aggressor could aggravate timing errors due to
>>> changes in temperature or stress in the cable. Whether this is important
>>> for you depends on how time-nutty you want to get.
>>> 
>>> See these papers:
>>> 
>>> Effects of Antenna Cables on GPS Timing Receivers:
>>> http://tf.boulder.nist.gov/general/pdf/1384.pdf
>>> 
>>> Absolute Calibration of a Geodetic Time Transfer System:
>>> http://xenon.colorado.edu/paperIrevise2.pdf
>>> --
>>> Bill Byrom N5BB
>>> 
>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.




More information about the Time-nuts_lists.febo.com mailing list