Ground plane for M4 ladybird

[Astudillo]

Hey Guys,

I started designing a groundplane for my M4 ladybird build. I read in the wiki that you can use a 1mm pcb or just a piece of plain copper plate. Now with weight being important on such a small build, i was wondering if there is a minimum thickness of the copper groundplane?
I also read in the taoglas pdf that the size of the groundplane matters as well. Does anyone have any experience with different sizes groundplanes for their antenna?

Here are the 2 designs i made weighting approx. 3.79 and 2.5 grams.


The round one is a bit heavier, but it looks like it will shield the antenna a bit better than the square one.
What do you guys think?

Cheers,

Ray

[JussiH]

No, thickness does not matter. Basically any material with a conductive surface or inner conductive plane would work. Basically a CD would work well and not weigh much. Maybe even a piece of thin CF would work too.

I made one from 0.25mm alu and it worked but was pretty flimsy. As for area, basically the more, the better. For a 10mm antenna, I would shoot for at least 30 mm diameter.

Distance from the FC itself is probably a factor too. So the higher you can lift the antenna up off the FC surface, the better.

[kinderkram]

Ray, I stole your image for a Wiki entry.
http://autoquad.org/wiki/wiki/m4-microc ... PS-Antenna

[Astudillo]

No problem,

If you need another render, just let me know!

[kinderkram]

Thx!
I might come back on your offer. You renders look great.

[Kisssys]

Everything you read will tell you that the larger ground plane will make the antenna work better. Usually everyone stops reading right their. If you read further you will find the very small patch antenna's have a very narrow bandwidth of 5 mhz. If you look at the table attached you will see that adding a ground plane can shift the frequency of the antenna so far up that even though it works better it's on the wrong frequency and actually performs worse. I think we gain more from RF shielding than from improving the antenna performance.

[Max]

Nice resource Steve. So if I read that chart correctly, we want at least a 30 mm2 GP, and ideally 40 mm2, but for a small patch not larger than, say, 45 mm2

Also is the definition of "ground plane" here involve that it be connected electrically to the antenna ground? Or is any chunk of conductive shielding under the antenna considered a GP?

Thanks,
-Max

[Kisssys]

Max,

Yes, looks like anything above 30mm is fine with diminishing returns the larger you make it. The larger the patch the less help the ground plane has on improving the signal

The chart is a little hard to read but if you look at the frequency you see that a 30mm sq gp can raise the frequency 30 mhz from its original frequency. If you buy a 10mm patch on 1575 ( the gps freq ) it may actually be on 1605 if it sitting on a ground plane. The ground plane for the 10mm patch may in fact do as much harm as good. When you have a 10mm patch with a 5 mhz bandwidth it doesn't take much to throw it too far off frequency. I'm going to experiment with tuning the patches by adding a little metal to the patch. That's all theory.

The testing I have done has shown this to be true also. I get a stronger signal without the gp than with a gp. Testing antenna's like this can lead you astray very quickly so I take all the data I get with some skepticism. My testing consisted of a signal generator setting on the top of my shelf. The variety of gps antenna's were about 5 feet below hooked to a spektrum analyzer viewing the signal strength. With the 10mm patches I consistently got a weaker signal when I placed a ground plane below the antenna. I had one 25mm patch that the signal went up and the larger 35mm patches showed little difference. I still believe that getting the ant away from any noise sources is just as important.

[Max]

[Kisssys]

Normally the gps antenna shield touches the ground plane for a ground. It acts as both a RF shield and ground plane even if it is not hooked directly just not as well.

There is always the question as to whether to run a ground wire to the shield and to the FC board. It may help but it may make things much worse. The ground is strictly a dc ground, as far as RF is concerned it may just be a good antenna and radiate the noise from the FC board. RF grounds have to be very short. I've tried it every way you can think, without extremely expensive equipment it trial and error.

To make things even more complicated is if you enclose the antenna in some sort of plastic you can lower the frequency.

I've just come to the conclusion that to make a 10 mm antenna work you need a specific installation to tune it to which is not reasonable. The answer is use a 25 mm ant or larger and you should be fine with or without a ground plane. I also think their is value in a remote GPS and ant combination. You can insure the antenna is on frequency, feed it close with a low noise regulator and you can mount it away from the FC board's noise.