Let's Make Robots!

How to make an 2.4GHz antenna

 

Hello LMR!  I am assembling a wireless video transmitter that calls for an antenna.  As an amateur radio operator (licensed at age 11) I ought to  know how to build an antenna, but unfortunately I do not.  Google is kind of helpful -- I found a few suggestions about turning a coax cable into an antenna, but there's a lot to sort through.  Also, a 50ohm lead is called for in the data sheet, and I don't know if I can substitute a resistor (the data sheet calls for a 50ohm= 75mil@1mm thick trace -- which seems like it would have 0 ohms if I made it out of copper).  Basically, I am confused, and wonder if any folks here at LMR have built something like this before and could tell me what the data sheet is asking me to do.

Thank you for your help!

-John

Files:
http://letsmakerobots.com/files/AWM630TX.png
http://letsmakerobots.com/files/AWM634RX.png
http://letsmakerobots.com/files/AWM634RX_no_Rssi.pdf

http://letsmakerobots.com/files/AWM630TX.pdf

Search links:
http://www.rc-cam.com/forum/index.php?/topic/833-airwave-2-4ghz-modules/page__st__40__p__8464&#entry8464__s__0985846b6e20ec4df7a98a45c440eb1d
http://www.rc-cam.com/forum/index.php?showtopic=993
http://wireless.gumph.org/articles/homemadeomni.html
http://flakey.info/antenna/waveguide/

The Airwave modules were from a Trimersion prototype, so I also found this example of how Trimersion built an antenna: pic at http://www.rcgroups.com/forums/attachment.php?s=fafdf6d42d090a17dbdba07be38d4b02&attachmentid=2977244 from thread http://www.rcgroups.com/forums/showthread.php?p=14015873
The little gray wire coming off the big metal module is the TX antenna.

 

AttachmentSize
AWM630TX.png94.77 KB
AWM634RX.png111.35 KB
AWM634RX_no_Rssi.pdf78.87 KB
AWM630TX.pdf85.47 KB

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Suppliers:

$9+S/H (~$15) Camera: http://myworld.ebay.com/charmlhgogo
This is a neat little device -- it is a camera with an adapter plug that takes about 9V and converts it.  I disected it when a fellow tinker came over and discovered that in the middle of the camera cord, there is a small power regulator that takes the 7V-12V and converts it to 5V then to 3.3V (or was it 3V?)... the unregulated 9V and regulated 3.3V are then both sent to the camera module. 
Why would I dissect it?  Because I want to cut out as much weight as I can before I strap it to a flying platform, and was wondering what if anything the lumpy cable was doing.

$20 Wireless: As mentioned, it's an Airwave TX/RX pair ($20- for both).  This one is surplus from a fellow in Florida, USA, as discussed in the following link: http://forums.adafruit.com/viewtopic.php?f=8&t=12513

$45 Display: The goal was for a daylight-viewable camera link -- so I thought, why not use a binocular display?  Kopin display is the type used in prototype Trimersions (320x240).  If you buy a Trimersion, you get a 640x480 Kopin display and the Airwave A/V wireless modules, plus some nifty gear, for about $70-.  I haven't built the binocular display power supply/connectors, so Icurrently use a portable DVD player for the video (I have a 3-year old Toshiba portable DVD player that has a cable for AV input, and which I've used to test the camera).  The video link could also be an NTSC/PAL video input to a computer, which would record the video.  There's likely a bunch

Summary: I have a ~$35 USD wireless camera, which I intend to use in a remote-control plane/quadcopter.

Thank you rik!  Your answer clears up exactly what I was confused about.  Now that I can interpret the data sheet I should be able to understand the other material I found on the RC forums or other sources.

I am not used to units of mils, but my calculator tells me that 75mils is about 1.9mm.  I should be able to etch 1.9mm traces to a precision of +/- 0.5mm.  Avoiding sharp corners in the PCB layout is probably also important for something like an antenna trace -- in an example I found (a cordless phone), I noticed that when the antenna trace bent, it was given a smooth contour.  I'll try to post a blog entry once I get it working since wireless video could easily have a robotics application.  Thanks!  -John

You're right about the bends and angles. Read antonb's latest walk through on PCB design. I seem to remember that EagleCAD has options to make sure corners are round, rather than angled.

The "mil" is on of the most confusing units I encountered in my new hobby. I had never heard of the thing. Who would want to break down an inch into one-thousands!?!? Maybe 1/1024, but never had I expected 1/1000.

How much is that in nano furlongs?

I've no idea who first broke inches into a thousand pieces, but I'm one of the folk that's used to using it. All my measuring devices (calipers, dial indicators, etc...) are in thousandths (well, except my one 0-25mm caliper, which I rarely use).

It does make some sense. 1/10 = 0.1, .1/10 = 0.01, .01/10 = 0.001 = thousandth, .001/10 = 0.0001 = ten-thousandth (or "tenth" in machinist-speak). The shop I first started learning machining at had us making punches and dies down to "half-a-tenth", or 0.00005 inch.

It's really entertaining when I say to the wife that something is "five hundred thousandths". That's 0.500 inch, or half an inch. She wants to know why the hell I don't just say "half an inch". The point of it is that if I say "half an inch" I'll expect the item to be measured with a tape measure, and the tolerance is the width of the paint line on the tape (which is highly arbitrary). If I say "500 thou" you know right away that the expected tolerance is 0.001, and I expect you to use something like a dial caliper to do the measuring.

For your question, a furlong is 7920 inches. So a nano furlong is 0.00000792 inch. 0.001 / 0.00000792 = 126.262626. So there are approximately 126 nano furlongs in a thousandth of an inch. I think. I spent all morning at the DMV so I'm feeling rather stupid today. Someone check my math ;)

I got that after the first three paragraphs. Sorry to hear about your stay at the mythical DMV. Even we in Europe heard of it....

Antenna and their leads are rated for many properties. Impedance is the first to know about. Confusingly enough, it is measured in the same unit as Resistance (Ohm), but it is not the same thing.

Coax cable has the impedance often written on the (outer) insulation. 50 Ohm is a pretty widely used value.factory made antenna have  similar rating. Maybe not written on the product, but easily derived from its purpose. Most wifi equipment for example call for 50 Ohm along the entire chain [tx/rx, cable, antenna].

The width of those traces on PCBs are relevant because they are "transporting" some very high frequency signals. Note that the distance between them is also deliberately designed.That should minimize interference and stuff like that. Probably impedance as well, since that is a function of how greatly the two traces influence each other.

The distance in this picture

AWM634RX.png

is the thickness of the PSB (1 milli-meter), the traces themselves are much thinner (1.6 milli-inch) and narrower (75 milli-inch). Also the dielectrical property of the PCB material is demanded here: the mentioned "FR4" is (apparantly) a quality requirement.

Finally, Spark Fun (among others) sells cables and connectors for these frequencies.

The modules look a lot like the ones Wavecom uses in their 2.4Ghz Tx/Rx units. They are easy to modify and increase the power to 60mw or more. Wi-Fi antennas are plentiful and cheap, and would seem to be a simple solution.

73 KE6GHS

Thank you for the tip -- I have two wifi antennas around -- either of which would make a good TX/RX antenna.  I might be able to figure out a smaller lightweight antenna with some of the RC forums' tips.  With the 50ohm lead on the PCB - I think I'm misreading the significance of "50 ohms".  The antenna may be the 50ohm part, and the lead may have the recommended width for some other reason (e.g. a wide lead could be less likely to cause interference when a 2400MHz signal is applied).  I dissected a 2.4GHz cordless phone and noticed wide traces on those antennas as well.  In the end, nothing is as educational as experimentation.  Once I get something that transmits data, I can work on the range.

Thanks Gary!

-N2TGW