Antenna question

Here's a Google search with several videos on how to increase the range of your system.

The first one attaches a coax and puts the antenna outside the garage. I have also heard of adding reflectors behind the wire antenna you have, just another piece of wire of the proper length and distance behind your existing antenna.

Mikek

You are making the assumption that the manufacturer has actually optimised the design rather than just using something that works well enough.

If you want to experiment non-destructively you could try folding the end of the wire back on itself and taping it tightly in place to get a shorter length. The "ground plane" will almost certainly be all the wiring coming out of the box.

John

and also 1/2 wave is also a viable antenna. but with a different terminal impedance.

What problem are you trying to solve?

For a receiver a long wire is a good antenna.

Dan

I'm just trying to understand antennas. I'm working on a project where I suspect the RF signal won't make it to the receiver, and I would like to understand what things to try to improve the range.

Great question! A 17 cm antenna length is recommended for my 434 MHz opener, which is about 57 mm longer than the theoretical quarter wave length. Perhaps the manufacturer manually adjusts the length in order to minimize the SWR?

Thank you,

A monopole antenna can be any length if you don't care about the pattern. If it's longer than 1/4 wavelength, then add some capacitance to tune it down in frequency. If it's shorter than 1/4 wavelength, then add some inductance (loading coil) to raise the tuning frequency. There's nothing sacred about 1/4 wavelength except that it's one of the convenient antenna lengths that produces a 50 ohm antenna.

Probably a meandering trace antenna. Something like this: Some detail on how it works, and other antennas that your might find inside the key fob should you find the courage to remove one screw and look inside.

No. There was a time during the early 19th century, when it was thought that radio worked best when the endpoint antennas were identical, but that was because the only frequency dependent device in the old spark type transmitters was the antenna. They had to be tuned to about the same frequency for spark gap to work and two identical antennas made that fairly easy. Once resonant antenna were discovered, where antennas of different lengths and construction could all be tuned to any specified frequency, the identical antenna concept was dropped.

Because 1/4 wavelength is NOT ideal in any way. It's just a convenient way to get a 50 ohm antenna (or 35 ohms over a ground plane). If the designer was using a high impedance output device and antenna with a matching high impedance would be closer to ideal.

No. A counterpoise, such as a ground plane, is useful, sometimes necessary, but doesn't really have a dramatic effect on the length of an antenna. The capacitance between the monopole and the counterpoise ground lowers the resonant frequency. To compensate, the antenna will be somewhat shorter than 1/4 wavelength.

Marketing, fashion, convenience, cost, cosmetics, ergonomics, or whatever the designer thought might work best.

Hint: If you calculate the gain of an antenna for different monopole wavelengths, you'll find that the gain is fairly constant down to about 0.1 wavelengths. I threw together a study on this a while ago to answer a similar question: For example, 1/4 wavelength monopole, above an infinite ground plane, shows about 5dBi gain. The gain remains fairly constant down to 0.050 wavelength. The wavelength is in the filename: What does change is the pattern with shorter antenna length and for antenna lengths longer than 5/8 wavelength is the antenna impedance, and the bandwidth (when including the matching network). Since your remote control uses only one frequency, bandwidth is probably not a problem. Since matching your key fob and antenna at one frequency is not a major feat of engineering, antenna impedance is not a major issue. Since the gain is fairly flat down to very short antenna lengths, that's also not likely to be a problem.

To quote "Mike", what problem are you trying to solve?

Make the antenna as large/long and as close to the receiver as possible. The idea is that you want to capture as much signal as possible and lose as little as possible. Forget impedance matching.

Build a model of your end to end system losses. Start at one end with the transmit power and at the other end with the receiver sensitivity. If digital, use the rx signal level at some BER (bit error rate) or PER (packet error rate). In between, include the transmit and receive antenna gains, coax cable losses, connector losses, path loss, and any other impairments such as Fresnel zone diffraction, foliage attenuation, atmospheric attenuation, wall material penetration, and anything else that might get in the way. The path loss through the air is determined by the Friis Equation.

There are plenty of online calculators to make the math painless ranging from really simple: to insanely complexicated: Google for "RF link calculator" for more online calculators.

The hard part is what to use for "fade margin" or "system operating margin". For digital, it depends on the performance of the demodulator. For wi-fi and such, I use 20dB fade margin. For your unspecified system, probably something different.

Good luck.

On Tue, 03 Oct 2017 18:22:32 -0700, Jeff Liebermann wrote: (blah-blah-blah)

Maybe an example might help. Here's something I scribbled comparing several possible Ubiquiti radios for a 2.4GHz wireless link: The numbers will probably not match your keyfob and receiver example, but the method and calculations used will probably be very similar.

Please note that when calculating such RF system losses that the numbers produces will always be BETTER than reality. When built, there will always be additional losses cause by factors that were not properly considered or numbers pulled out of a hat. If your SOM (system operating margin) is marginal, it will likely fail. Therefore use pessimistic numbers, such as lower than typical transmit power, worse than typical receiver sensitivity, realistic antenna gains, etc and you'll have something that has a chance of working when done.

Good luck.

FWIW my fob uses a SAW based CPCA transmitter to keep things small. The price you pay for such a small transmitter is limited range.

Thank you,

First, you have invested considerable time and effort trying to help me with your posts, and I really do appreciate that. But honestly, it's all way over my head.

The problem relates to my other thread about the OSH project. The device in question is this one:

And a video giving at least a hint about the transmitter pc board layout (at 5:30) and antenna is here:

In that video, it's shown that the transmitter antenna doesn't work very well, and it doesn't seem to matter much how far it's extended. Also, while the spec says 100 meters, all anyone on Youtube can get is about 100 ft., and that's line of sight.

In my use case, the signal has to go out through windows, travel along the outside of a brick building, then come back into the lobby to the receiver. A picure of that is in the alt,binaries.schematics.electronic newsgoup.

My guess is this isn't going to work, and I just want to see if there's anything relatively simple I can do to extend the range and make it more likely to work.

The receiver has no antenna, and the throughhole provided for adding one is right at the corner of the board. So I suspect there's no meaningful groundplane there. But for a first attempt, I thought I would use a wire that's 1/4 wavelenght long. I'm not sure that's the right place to start, but I don't know of a better option.

Then the transmitter is more complicated because it has a telescoping antenna built in, plus, it appears, additional copper plane (not ground) on the board as seen in the video. The problem is that it seems the antenna is the same for 433 and 315, so perhaps not right for either, but I suspect right for 433, whereas I'm using 315.

So the only thing I know to try is to replace the existing transmitter antenna with a wire which is intially longer than 1/4, and see what if anything happens to the range as I cut off pieces of it.

I think experimenting with antenna length is about all I can do. I might be able to change the transmitter amplifier circuit, but my guess is that's unlikely. The only other thing, if I can find a measuring point, is to fine tune whatever adjustments may be on the receiver to make sure the frequency is an exact match - on the theory that at this price, calibration may not have been precise at the factory.

Thanks again for trying to help.

Isn't it just a ~10% correction for the lower speed of light in a good conductor like copper as opposed to free space?

Indeed. It is only on transmit where matching really matters. (unless you are pushing for the absolute maximum sensitivity)

Interesting question arises how do they get a suitable quarter wave antenna for 318MHz ~ wavelength 1m into a key fob that is perhaps no more than 5cm on its longest dimension.

Fractal design, coil ferrite loaded or mismatched on a wing and a prayer. (or some other cunning trick)

The radiation pattern of a long wire is along the wire, not particularly useful here - on the contrary.

I'm interested in the input impedance of the garage door receiver. Any one have info on that? Is it really near 50 ohms? Mikek

Inductive loop with loading capacitors. Q is high, bandwidth is small, efficiency is small, range is small. Hey, what were you expecting? :-)

FYI, fractals are for compact wideband antennas.

Tim

Not likely and it doesn't matter either. The receiver is probably a dead-simple super-regenerative job.

Jeroen Belleman