Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

antenna options

Status
Not open for further replies.

justin_t

New Member
Hi

Im making a small wireless FM receiver based on RF solutions pair of recievers/transmitters at 433 MHz.

Until now ive just been using a 17.5 cn length of cable as an antenna with good results but for production im not sure if it is a good idea to continue with this method.

How easy/difficult is it to encorporate the antenna into the PCB tracks of the PCB. I can find maybe a square inch to play with on the board. Is there alot of calculations/fine tuning required or sould i just be able to make a track say 17.5 cm long that may double back on its self to fit on the board (the board is approx 9 cm long.

Any thoughts?

Justin
 
Hi

Im making a small wireless FM receiver based on RF solutions pair of recievers/transmitters at 433 MHz.

Until now ive just been using a 17.5 cn length of cable as an antenna with good results but for production im not sure if it is a good idea to continue with this method.

How easy/difficult is it to encorporate the antenna into the PCB tracks of the PCB. I can find maybe a square inch to play with on the board. Is there alot of calculations/fine tuning required or sould i just be able to make a track say 17.5 cm long that may double back on its self to fit on the board (the board is approx 9 cm long.

Any thoughts?

Justin

half wave di-poles are popular.perhaps you can have fractions of that too.
 
Hi

Im making a small wireless FM receiver based on RF solutions pair of recievers/transmitters at 433 MHz.

Until now ive just been using a 17.5 cn length of cable as an antenna with good results but for production im not sure if it is a good idea to continue with this method.

How easy/difficult is it to encorporate the antenna into the PCB tracks of the PCB. I can find maybe a square inch to play with on the board. Is there alot of calculations/fine tuning required or sould i just be able to make a track say 17.5 cm long that may double back on its self to fit on the board (the board is approx 9 cm long.

Any thoughts?

Justin

If you don't have a good understanding of antennas you are probably best sticking with someone else's design. There are lots of app notes out there, check out chip maker's websites, for example
 
Last edited:
Antenna design is a magical art that is beyond me. I've used these 433Mhz chip antennas in the past. I got them from Mouser.
 

Attachments

  • fire-xmit-01.jpg
    fire-xmit-01.jpg
    196.4 KB · Views: 704
Here's a PCB-antenna as it used with a MICROCHIP transceiver. The dimensions are lambda/4.

The feedline does not add to antenna length.

Maintain a top trace keep out area of 0.21X0.2 inches at the feedpoint.

Details shown in attachment.

Boncuk
 

Attachments

  • PCB-ANTENNA.jpg
    PCB-ANTENNA.jpg
    16 KB · Views: 851
Mark, Which piece of the board is the antenna? The grey component at the bottom of the picture?

The chip is at the bottom, and it's actually a light blue-ish. It's beside the transmitter and SAW filter. It's a Yageo part. Checking on Mouser and Digikey, they have the parts listed, but they aren't stocked. The only other alternative chip antenna is a Linx part from Digikey. Part #: ANT-433-SP_
 
To get any reasonable amount of efficiency from an antenna (power radiated vs power consumed as resistive losses), the length of the antenna needs to be a half-wavelength long.
At 433MHz, λ/2=3e8/(2*433e6)=0.23m or 23cm or 9".
Such an antenna is usually configured as a center-fed dipole, placed either horizontally or vertically.

Various tricks are used to make a dipole physically smaller, like bending the sides into a V, or shortening them and then adding inductance (loading coils) to maintain resonance, or making half a half-wave dipole (¼λ monopole) operated against a conductive ground-plane (vertical ¼λ antenna)

The sad reality is that anything you do to modify the antenna so that its "aperture" becomes less than ½λ greatly compromises its efficiency.

There are lots of examples of "compromised" antennas being useful. An AM Broadcast receiver uses a tiny loop-stick, but the transmitter uses a 100m full size ¼λ vertical with hundreds of wires in the earth as a counterpoise and transmits kW of power. A mobile-mounted ham transceiver uses a loaded vertical (and only radiates ~3W out of 100W fed to the antenna), but is able to communicate hundreds of miles.

In both of these examples, the trick is POWER! You can afford to lose efficiency if you can make it up with brute force. However, if your transmitter is in the mW range, and your receiver is not the best (which it isn't in the inexpensive data communications devices), then if you try to use short compromised antennas for both the transmitter and the receiver, you deserve what you will get, i.e. very short range. :(

btw- the transmitting antenna must have the same orientation (vertical, horizontal) as the receiving antenna; otherwise there is an additional 20db cross-polarization loss.
 
Last edited:
To get any reasonable amount of efficiency from an antenna (power radiated vs power consumed as resistive losses), the length of the antenna needs to be a half-wavelength long.
At 433MHz, λ/2=3e8/(2*433e6)=0.23m or 23cm or 9".
Such an antenna is usually configured as a center-fed dipole, placed either horizontally or vertically.

Various tricks are used to make a dipole physically smaller, like bending the sides into a V, or shortening them and then adding inductance (loading coils) to maintain resonance, or making half a half-wave dipole (¼λ monopole) operated against a conductive ground-plane (vertical ¼λ antenna)

The sad reality is that anything you do to modify the antenna so that its "aperture" becomes less than ½λ greatly compromises its efficiency..

This is not quite true. Theoretically there is no problem making a small antenna with arbitrary gain, what happens however is that the Q of the antenna starts to increase astronomically. Practically this means that you can shrink the size of the antenna and initially you only sacrifice bandwidth, but once you have shrunk things by something like a factor of 10 these limits really start to bite. There are plenty of folk out there selling quite efficient 433MHz antennas that are only 12mm or so in size.

There are lots of examples of "compromised" antennas being useful. An AM Broadcast receiver uses a tiny loop-stick, but the transmitter uses a 100m full size ¼λ vertical with hundreds of wires in the earth as a counterpoise and transmits kW of power. A mobile-mounted ham transceiver uses a loaded vertical (and only radiates ~3W out of 100W fed to the antenna), but is able to communicate hundreds of miles.

The AM situation is not directly comparable to the 433MHz one as you are not normally receiver noise limited, so lousy antenna gain is not so much of a problem, but yes simple ferite rods are not greatly efficient.

I think that there would be lots of hams out there that would take issue with you on 3% efficiency for a loaded vertical. Well matched and plated with a good conductor I would expect something over 90%.
 
This is not quite true. Theoretically there is no problem making a small antenna with arbitrary gain, what happens however is that the Q of the antenna starts to increase astronomically. Practically this means that you can shrink the size of the antenna and initially you only sacrifice bandwidth, but once you have shrunk things by something like a factor of 10 these limits really start to bite. There are plenty of folk out there selling quite efficient 433MHz antennas that are only 12mm or so in size.

Keeping the antenna resonant so that it presents an impedance conducive to acting as a proper load for the transmitter (or receiver input) is possible with a shortened antenna. That is where the lack of bandwidth of a resonant, but shortened antenna comes into play. However, once you shorten the antenna, the efficiency goes to hell.

The only way of making a "directive" antenna is by increasing its "aperture", i.e. making it physically larger. All directive arrays consist of spaced elements, where each of the elements, and the spacing between them is on the order of ¼ to ½λ; think yagis, colinear arrays, t.v. antennas, phased arrays, etc.

Get me the dimensions your 12mm shortened antenna, and I will put it in my antenna-modeling software and will compare the field patterns to a resonant dipole for the same frequency. I have done this hundreds of times, and I can predict what the results will be. I'll be happy to post them here.

The AM situation is not directly comparable to the 433MHz one as you are not normally receiver noise limited, so lousy antenna gain is not so much of a problem, but yes simple ferite rods are not greatly efficient.

I think that there would be lots of hams out there that would take issue with you on 3% efficiency for a loaded vertical. Well matched and plated with a good conductor I would expect something over 90%.

A top-loaded big-wheel capacitance-hat 75m mobile antenna with silver-plated bug-catcher is like 5% efficient, i.e. it radiates 5W for 100W fed to it, the other 95W goes into heating the antenna and the earth under the car. I know, cause I have been a judge in ham radio antenna efficiency contests. A simpler base-loaded whip is like 3% efficient. btw- the guy with the 5% efficient antenna won the contest over the guy whose antenna was only 4% efficient :D

I have used a hiQ resonant 6' diameter loop with a bandwidth so narrow that it wouldn't pass a 3kHz wide SSB voice signal on 40m, and best I was able to measure (field strength) suggested it was about 7% efficient compared to a 66ft dipole.

There isn't any such thing as an efficient shortened antenna.
 
Last edited:
The only way of making a "directive" antenna is by increasing its "aperture", i.e. making it physically larger. All directive arrays consist of spaced elements, where each of the elements, and the spacing between them is on the order of ¼ to ½λ; think yagis, colinear arrays, t.v. antennas, phased arrays, etc.

Get me the dimensions your 12mm shortened antenna, and I will put it in my antenna-modeling software and will compare the field patterns to a resonant dipole for the same frequency. I have done this hundreds of times, and I can predict what the results will be. I'll be happy to post them here.

All I was saying is that you can trade size for bandwidth down to roughly lambda/10 when practical problems take over. For example check out this datasheet.

You are correct that most directive arrays consist of elements spaced on the order of half a wavelength and practically it is hard to do much better, but that is not what we are discussing here.

A top-loaded big-wheel capacitance-hat 75m mobile antenna with silver-plated bug-catcher is like 5% efficient, i.e. it radiates 5W for 100W fed to it, the other 95W goes into heating the antenna and the earth under the car. I know, cause I have been a judge in ham radio antenna efficiency contests. A simpler base-loaded whip is like 3% efficient. btw- the guy with the 5% efficient antenna won the contest over the guy whose antenna was only 4% efficient :D

I have used a hiQ resonant 6' diameter loop with a bandwidth so narrow that it wouldn't pass a 3kHz wide SSB voice signal on 40m, and best I was able to measure (field strength) suggested it was about 7% efficient compared to a 66ft dipole.

OK, at 75m I agree efficiency is lousy. To go even further look at the kHz antennas that the navy operate for submarine communication. Almost all the energy goes into heating seagulls. I still maintain that it is possible to efficiently match antennas that are around lambda/10 to lambda/20.

There isn't any such thing as an efficient shortened antenna.

There isn't any such thing as a broadband efficient shortened antenna.

If you are prepared to sacrifice bandwidth, there is scope to scale down by a factor of somewhere up to 10.

Quick look for some measurements, all I could find was this loop, a ham design with side lambda/20 and a claimed 68% efficiency.

The fundamentals haven't changed much over the years, you've probably read it, but for those who haven't my favourite reference is:

Fundamental limitations in antennas
Hansen, R.C.;
Proceedings of the IEEE
Volume 69, Issue 2, Feb. 1981 Page(s):170 - 182
 
If i have been using the 17.5 cm of cable with good results, is it viable to coil or loop the wire inside the enclosure i.e. does it make a difference if the length is still 17.5 cm but the cable isnt in a straight line...?
 
If i have been using the 17.5 cm of cable with good results, is it viable to coil or loop the wire inside the enclosure i.e. does it make a difference if the length is still 17.5 cm but the cable isnt in a straight line...?

Coiling the wire will certainly change the resonant frequency of the antenna.

Looping the wire around the equipment enclosure will affect the antenna characeristics in an unperdictable manner.

The bottom line: Try it, if it gives fully acceptable results, job done.

JimB
 
If i have been using the 17.5 cm of cable with good results, is it viable to coil or loop the wire inside the enclosure i.e. does it make a difference if the length is still 17.5 cm but the cable isnt in a straight line...?

Did you check the paper in my first post - it gave an example of a 433MHz helical antenna
 
If i have been using the 17.5 cm of cable with good results, is it viable to coil or loop the wire inside the enclosure i.e. does it make a difference if the length is still 17.5 cm but the cable isnt in a straight line...?

Coiling a piece of wire that was formerly an antenna makes it into an inductor; any resemblance it has to an antenna at that point is purely coincidental.
 
Last edited:
Did you check the paper in my first post - it gave an example of a 433MHz helical antenna

That link was very useful to me. There's not much of a question here, but I'd like to outline what I'm going to try to do and if anyone here has any suggestions, I'd be happy to hear them.

I bought some MSP430-RF2500 development modules and have been fairly impressed with them. I'd like to make a multipoint network of devices in my house that work off of a single remote. The devices will be simple on off switch's or controllers for things like my fireplace (which require temp control), or dataloggers that report to the remote, like weather station...

I'm planning on getting some CC1101 transceivers with a matching 900Mhz Johanson Balun. I'll be playing with antenna designs.

Like I've indicated, I know very little about this stuff. As far as I can figure the best step for me would be to make up a bunch of designs and simply test them all out and tweak them until I get a range/coverage that I'm looking for. Unfortunately to test them, I really need to make up a whole new board for each try. I can't just simply make a bunch of PCB antenna's and plug them each in to try out, since they will have to be integrated into the board in final design that's how they have to be tested. This isn't that big of a problem. I won't have too much difficulty desoldering/soldering the main chips for multiple test boards.

Anyway, that's my plan. Any suggestions would be helpful, but I'm going to go through what I've learned so far from places like that article and the TI reference designs and simply experiment.
 
The simplest antenna is a 1/4 wavelength whip (3.28 inches long) with a solid ground plane behind it.
https://www.electro-tech-online.com/custompdfs/2009/05/antenna-1.pdf
Might provide some good reading. If not heavy.

Good article. It says basically what I tried to say. The gold standard is a ½λ diplole, or ¼λ whip over a large ground. Any shortend antenna is 10 to 20db worse. If you use shortened antennas at both transmitter and receiver, the loss is additive (not just another 3 db).
 
That was the same article from Post #3 in this thread. The one I referred to in my post.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top