[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gpsd-users] Fw: ✘Grandpa got a new hammer

From: nick
Subject: Re: [gpsd-users] Fw: ✘Grandpa got a new hammer
Date: Sat, 8 Sep 2018 13:56:13 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.9.1


I was scrapping some old hard disks yesterday and thought about your HDD platter experiments. Here's some comments.

That looks like an active patch antenna in your picture.

The amplifiers in those antennas usually have around 25dB gain and a 2dB noise figure.

With that much gain the overall noise figure of the system is determined by the noise figure of the amplifier.

So the 4 way splitter does not measurably degrade the overall SNR.

Adding an extra amplifier between an active antenna and the splitter cannot improve the overall SNR.

Adding an HDD platter ground plane will raise the antenna 'gain' (directivity) which will increase the SNR, but degrade the axial ratio which will reduce the SNR.

The platter I have here is 95mm in diameter with a 25mm hole in the middle.

provides useful information for estimating the performance of patch antennas over square ground planes, but not round ones with a hole in the middle.

My guess is that you might get 1 or 2 dB of overall improvement in the SNR.

But even this might be significant in terms of number of satellites tracked and DOP.

I recently replaced my roof mounted patch antenna with about 150mm x 150mm ground plane with an AMEC ANT-21 antenna which appears to have a ground plane about 90mm diameter. Although the C/No numbers have not gone up much, there is a big difference in the number of satellites tracked and the availability of a 3D fix for time synchronisation purposes.

I suspect that an amplified axial mode helix might give the best results in static applications since you don't have to trade off directivity against axial ratio. I plan to build one some day.

This is also interesting


On 07/09/18 23:14, Gary E. Miller wrote:
Yo All!

And everything looks like a nail.

TL;DR: don't fear the GNSS antenna splitter.

For years I have been comparing GPS side-by-side, each with its
own antenna.  This led to complaints that I was a likely measuring
differences in the antennas instead of differences between GPS.

The classic solution to the problem is to use one antenna, feed that
antenna into a power splitter, each output of the power splitter feeding
a single GPS.

To keep the antenna DC power from the GPS's from fighting, insert a DC
block on the line to all but one GPS.

A two way splitter cuts the signal power in half.  A four way splitter
cuts that into a fourth.  So some people think they need to add an
amplifier to regain that lost power.

Not required, maybe even harmful.  I have tried a few cheap chinese
GHz LNA, with bad results.

The theory is that the GPS cares about the Signal To Noise Ratio (SNR)
of the signal from the antenna, not the absolute power.  The splitter
should add very little to the noise and thus the SNR should be little

An amplifier can increase the power, but also adds to the SNR.

In practice, just a splitter has seemed to work for me, but I lacked
hard numbers.  With the new gpsprof polar plots I have some hard

With the new gpsprof polar plot statistics I can finally measure the
difference a splitter makes.  After running alternating tests, with and
without, a 4-way splitter, the results are very close.  The differences
between with/without are smaller than the short term jitter in the SNRs.
Averaging by eyeball there is maybe a loss of one in SNR.

By placing an HDD platter under the antenna I get back more than the
splitter lost.

See attached.  Notice the 50 loads and dc blocks on the 3 unused
splitter outputs.

Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
     "If you can’t measure it, you can’t improve it." - Lord Kelvin

reply via email to

[Prev in Thread] Current Thread [Next in Thread]