Small Antenna Close to Ground

[mattisback]

Hi All,

I have a system which is essentially just a one-way 2.4GHz digitally encoded radio. The transmitter is quite small (PCB 1x2cm) and the receiver is a lot larger. On the receiver is an ordinary +5dBi rubber ducky antenna. The transmitter uses an antenova Impexa (A6150) chipscale antenna.

I went out to do a range test recently and found that the range was very dependent on the distance of the transmitter from the ground. The receiving antenna was left fixed at 1 meter off the ground while the transmitting antenna was moved about. When the transmitter was close to the ground (~5cm off the ground), the maximum range (with a reliable connection) was about 10m, however when keeping the transmitter 1m off the ground I could easily manage a range of 100m+ (ran out of space to test) with a reliable connection.

I replaced the chipscale with a wire whip which increased the range while the transmitter was close to the ground to about 40m . I'd strongly prefer to have a chipscale though as it could be completely encapsulated.

I'm at a bit of a loss as to what is causing this effect. It seems like the antenna is being de-tuned by the ground, but I'm not sure that's the only thing going on. Does anyone have any idea how I could get this system to work at greater ranges with a chipscale antenna while the transmitter is close to the ground? Do you think it's just detuning or could there be something else going on here that I don't understand? I chose the Impexa because it has a wide bandwith and should be resistant to detuning but that doesn't seem to be helping, my ground plane is quite small though.. Does anyone have any other information on designing very small RF equipment?

Thanks guys.

 

[VEBill]

I found the chip scale antenna specification on this page. It appears to be a wee little dielectrically loaded spiral. They claim performance levels that are reasonable. Looking at their reference PCB design, I'd be tempted to replace the A6150 with a 50 ohm load just to see what happens - perhaps I'd be impressed.

Your reported range ratio of 100m/10m is 10:1. As you probably already know, due to inverse square law this ratio hints at 100:1 power ratio or a 20+ dB impairment. As you suspect, there's something else going on here.

Perhaps your transmitter is unintentionally radiating from points other than just the A6150 antenna, and so the resultant power is being radiated in directions other than you expect. This might impact the directionality, or it might make the resultant signal more sensitive to ground influence than just the A6150 antenna would be. You could confirm or refute this by plotting relative power in the various directions to make sure that the data plots match the OEM data for the A6150. If they match, good. If not, then you have an unintentional array.

Perhaps when you raised the transmitter off the ground, if you were holding it, your body acted as a parasitic element and improved the range. This is the same effect seen with car key fobs held in various positions to improve the range.

There are a host of possible improvements to increase the range:
Use a high gain directional antenna on the receiver.
Ensure the received LNA has a very low noise figure.
Use modulation coding or spread spectrum to provide coding gain.
Many more.

 

[mikekilroy]

hmmmm... guess if i had more energy this evening I could do the math for you, but perhaps you need to google "line of sight transmission distance formula" and see that height above ground directly determines line of site distance. IIRC from basic ham radio, once your antenna is a wavelength or so above earth, earth will have little effect: what is a wavelength at your frequency? pretty small. so it is most likely your distance is related to line of site rather than most anything else. Example from google search - search eq 6 and see if it doesn't explain your issue with distance:

http://www.wulfsberg.com/PDFs/150-040017.B.pdf

Have you any hams in your company? Might be worthwhile for you to talk to them......

73, AC8V

 

[VEBill]

The OP's experiment is operating over very small distances; the OP mentioned 100 meter (330 feet) range scale.

The usual line of sight relationships for height versus distance to radio horizon are intended to be valid on much larger scales where the curvature of the Earth enters into it.

 

[IRstuff]

Your antenna has a specific propagation pattern, so that might be one issue.

Additionally, there are some ground effects associated with emissions that close to the ground. Specifically, if your emission is horizontally polarized, there may be some suppression of transmission:

http://en.wikipedia.org/wiki/Antenna_(radio)#Effect_of_ground

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[mattisback]

Thanks for the replies! All the information has been very useful.

VE1BLL, yep that's the antenna. Why would you replace the antenna with a 50 Ohm load? I'm not quite sure what info you'd get from doing that. I could try it and post up some pictures of the VSWR with a 50 ohm resistor and the antenna.
I had a bit of a look at the radiation pattern yesterday. I don't have an anechoic chamber to test in so the results I took had about 2dB of noise on them. I found it extremely difficult to measure accurately, the radiation pattern of my antenna. Basically I just attempted to rotate the transmitter with the receiver fixed. I took measurements every 30 degrees in a full circle. Unfortunately my results were not very useful as the angle sample rate was quite course and the error on the measured signal strength quite noisy. The results I attained seemed fairly omni-directional however the radiation pattern of the antenna in that orientation (X-Z plane in the datasheet) was not omni-directional.

I don't think there were any spots with huge amounts of gain. It seems like the efficiency of the antenna is what is being affected rather than its directivity.
For the case with the transmitter a meter off the ground, we secured it to a wooden pole so that no-one had to hold it and affect its performance.

Mikekilroy, I had a look into the line of sight transmission distance formula and similar and read the pdf you linked to. I think that might be more related to long distance communications. In the pdf you linked equation 6 has more to do with the restrictions that the curvature of the earth impose on LOS communications. I'm talking much shorter distances that that. I'm using 2.4GHz as my transmission frequency and about 1mW transmission power.
I wish I did have a ham to talk to, I am however the sole engineer working for my (small) company. I've been having a few issues with RF lately, I think I really need to find someone with more experience than me to talk to about it all but I don't really know anyone. Anyway, that's why I'm here on the forums .

IRstuff, I don't think the radiation pattern of the antenna is the problem here. As I mentioned earlier in this post, I did find the horizontal radiation pattern to be very roughly omni-directional by experiment.
I read that section of the wikipedia article you linked on the effect of the ground on antennas. This does sound like it could be part of the problem. I also remember reading a paper recently that specified that the maximum Q achievable for electrically small antennas was quite a bit worse for horizontally polarized antennas than for vertically polarized antennas. This could be resulting in increased detuning of my antenna if it is horizontally polarized. So how would I know if my antenna is vertically or horizontally polarized?

Thanks guys.

 

[VEBill]

The idea about the 50 ohm load is to see how much signal is leaking from elsewhere. If those other areas of radiation are not well down then they'll combining to make a unique radiation pattern.

 

[mattisback]

Ahh, I understand. That's a great idea. Will give it a go today.

 

[zappedagain]

Did you try your radiation pattern with the transmitter at different heights?

 

[Higgler]

Rf signal bouncing (actually reradiating) off the ground will cancel or help your direct signal. It's a significant problem that has led to many inventions in signal diversity (both antenna placement, switching (home wifi- 3 antennas on the box) and signal modification (OFDM)

OFDM=

http://en.wikipedia.org/wiki/Orthogonal_frequency-division_multiplexing

= The primary advantage of OFDM over single-carrier schemes is its ability to cope with severe channel conditions (for example, attenuation of high frequencies in a long copper wire, narrowband interference and frequency-selective fading due to multipath)) so that if a "bad bounce" from offending objects hurts your transmission, then an alternate antenna/signal can be chosen to re-connect the signal link.

Dielectric of ground is 3-13 based on a report I read awhile ago, changes based on water content. Cement is worse for bounce than asphalt due to higher dielectric value.

Ground bounce antenna ranges must position the test antenna a very specific height above the ground for the ranges to work. Then that antenna is moved up and down to optimize the results.

Transmit and Receive antenna height, fix one, adjust the other to peak your signal.