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.

How do you make a PCB antenna?

Status
Not open for further replies.

blueroomelectronics

Well-Known Member
I'd like to use a 315MHz RF module and a 1/4wave antenna (I think) should be about 226.5 mm long. The PCB will be shorter than this so is it ok to bend, fold or spiral a PCB antenna?
 
Disclaimer: I don't have much experiance with RF.

Bending it would make it inductive.

You could go for a loop antenna and put the correct capacitance in parallel with it so you have a resonant circuit.

The trouble is if it's short, then it'll be slightly capacitive, you could put an inductor in series to see if it helps. At this frequency you can probably make good enough small inductors and capacitors by drawing them on the PCB.
 
The short answer is yes it is ok, but of course the devil is in the details. Assuming you are not transmitting a lot of power (like more than 1 watt), then we don't need to worry too much about overheating the transmitter due to a bad impedance match, you can bend things around a bit and get away with it. Bear in mind that even if your circuit board was long enough for the full quarter wavelength, it still would need a large ground plane to meet all the performance expectations of a typical quarter wave. This just means that a full quarter wave needs more than 226mm to work well.

Anyways, it is fairly harmless to bend the top 1/3 of the antenna at a right angle and still expect decent radiation from it. Another thing that I have done with good success in the past is what we call a "meander line" where you zigzag the pcb trace in a rectangular pattern. This starts to fall apart when you shorten by about 60% so don't push it shorter than 40% of that 226 mm. A reasonable meander dimension might be horizontal lengths of about 20mm and vertical part about the same. I'll run that through a simulator and tell you how it looks. But first, tell me what ground plane you have available.

edit: after some quick analysis, I should adjust the meander size recommendation to more like 5 to 10 mm on edge because it looks like a lot more inductance is needed than 20 x 20 mm would provide.
 
Last edited:
BRE,

Are you sure about the length? I have an FM tx made by Velleman for the usual VHF band, it is on the pcb and isnt that long...

Seems Rr posted before I did...mebbe that's What Velleman did...
 
Last edited:
Well it's a new layout on a 100mm x 50mm doUble sided PCB. I've not started layout yet but it's for a thermostat and the RF is a premade 315MHz Sparkfun $4 module.
 
Cant Sparkfun assist?

Surely they have data on aerial design..or am I being too naive again? :(
 
Last edited:
A loop antenna might be a reasonable alternative. Nordic Semi has some app notes on this:
**broken link removed**
(click Application Notes on the right hand side)

You could also consider just buying a pre-made pcb antenna but that wouldn't be any fun, would it?

I would be able to simulate something in a pcb antenna for you to get you close, but would need to have some idea of how you will partition your pcb for electronics vs antenna space. Perhaps PM to collaborate.
 
I just happen to have a 434MHz transmitter board with ~lambda/12 antenna. It's coupled by a 39nH choke from the collector of the oscillator transistor. The other squiggly trace is the RFC, which also include a deformed half-loop of wire on the other side of the PCB (for tuning?). The PCB is ~29.4mm wide, antenna (very roughly) ~56mm.

I've also heard of people using 1/4 L wire antenna loosely wound into a helix/spiral shape getting good results.
 

Attachments

  • 434_ant.jpg
    434_ant.jpg
    204 KB · Views: 2,262
The other squiggly trace is the RFC
Are you sure?

Chokes are normally quite high impedance at the intended frequency, that squiggly trace will have an inductance of about 10nH at the most which is an impedance of just 27R.
 
Are you sure?
Yes. It's in the right spot for the RFC; i.e. C2 is to ground, R5 is to VCC (C5 to ground also).

I'll have a look for the recommended RFC inductance value for the 434MHz Tx... just a sec..

EDIT: unfortunately I can't find the design note I was thinking of, but here's a snippet from an elector article from 05/98 on the transmitter bit.
 

Attachments

  • 434tx.gif
    434tx.gif
    4.7 KB · Views: 1,462
Last edited:
Is C3 needed?

For UHF it can be omitted, the internal CE capacitance of the transistor is normally large enough.
 
what this "E" emits? E-waves? i find weird all these antenna shapes out there... sometimes i'm wondering if they are really operational or just decoration_al :D
 
what this "E" emits? E-waves? i find weird all these antenna shapes out there... sometimes i'm wondering if they are really operational or just decoration_al :D

Designing antennas is a very creative process. There are some antennas that are designed using what is called a Genetic process (the antennas may be called Evolved or Evolutionary antennas) and the appearance of these antennas is often very sculptural and mysterious. This process uses a computer simulator. In the process, you start with some structure, simulate its performance, then allow one part of it to grow or shrink in a random direction. Then you check its performance again. If it got better than before, you keep that change, if it did not, you back up and try another change. This is a pretty dumb way to design an antenna, but with a computer doing all the dumb guesses and calculations it is fairly easy to do. It is really just automated trial and error (of course I am oversimplifying the details). It is surprising how interesting the results can be sometimes.

Evolved antenna - Wikipedia, the free encyclopedia
 
Last edited:
Designing antennas is a very creative process. There are some antennas that are designed using what is called a Genetic process (the antennas may be called Evolved or Evolutionary antennas) and the appearance of these antennas is often very sculptural and mysterious. This process uses a computer simulator. In the process, you start with some structure, simulate its performance, then allow one part of it to grow or shrink in a random direction. Then you check its performance again. If it got better than before, you keep that change, if it did not, you back up and try another change. This is a pretty dumb way to design an antenna, but with a computer doing all the dumb guesses and calculations it is fairly easy to do. It is really just automated trial and error (of course I am oversimplifying the details). It is surprising how interesting the results can be sometimes.

Evolved antenna - Wikipedia, the free encyclopedia


wow... i knew about fractal antennas because i like to play with Wifi's and i thought it was over engineered toy...
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top