[time-nuts] Non-impedance matched antenna cables

phil fortime at bellsouth.net
Sat Jun 14 02:43:07 EDT 2008

----- Original Message ----- 
From: "Bruce Griffiths" <bruce.griffiths at xtra.co.nz>
To: "Discussion of precise time and frequency measurement" 
<time-nuts at febo.com>
Sent: Saturday, June 14, 2008 2:16 AM
Subject: Re: [time-nuts] Non-impedance matched antenna cables

> phil wrote:
>> I know you guys like to drill holes through hairs, but ....
>> As I follow the discussions on merits of 50/75 ohm cable, cable length
>> changing with ambient temperature, tuning cable to a fraction of 
>> wavelength,
>> power supply noise etc.
>> Does not the "weakest" link determine it's best accuracy?  If so, the
>> receiver/electronics and internal programming seems to be the weakest 
>> link
>> and all these small nano/pico second variables discussed seems moot, at
>> least with unit in question.
>> If your receiver for example computes your location with a rather large
>> error in lat/lon or altitude, that error I would think would be greater 
>> than
>> the sum of all the "small" factors/errors being discussed. I have found, 
>> at
>> least with the Thunderbolt receivers I have used, they are rather sloppy 
>> in
>> it's location fix but even worse in it's altitude fix.
> All GPS receivers have larger height errors than latitude and longitude
> errors.
>> As I understand it, each foot of distance is a little over a nanosecond 
>> in
>> delay so would not position/altitude accuracy be the biggest variable, 
>> not
>> to mention the proper calculation and offset of antenna cable
>> attenuation/length.
>> I would be curious how the older Thunderbolt units compares to a newer
>> technology receiver/timebase in the "real world".
>> Just a thought
> Read the posted paper.
> The error in determining the position using modulation on the GPS
> carrier can be much larger than the cable delay variations.
> The effect will also depend on the correlator type used.
> Also for those that have better receivers like the M12+T, MI2M T etc,
> every last nanosecond of variation matters.
> The more that is known about cable instabilities etc, the better chance
> one has of actually realising the potential performance of a receiver.
> There is no point in using impedance transformers until the actual
> impedance of the receiver and antenna are known.
> Bruce

I can appreciate the implication of all these small variables and the 
wanting to better "every last nanosecond of variation"

My comment was, at least as I see it, only with a more expensive receiver 
and better timebase will these "smaller" variables be significantly 
relevant. Since the discussion revolved around the Thunderbolt receiver, I 
would think it's internal errors are far greater than the expansion of the 
antenna cable or cable impedance.


More information about the time-nuts mailing list