Maker Pro
Maker Pro

Simplest antenna for 5.8GHz 200mW transmitter to allow 100M range

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
I have Boscam TX5832 transmitter with above specs that I want to install in a mini quadrcopter. I only need a range of about 100-150 meters, so hopefully I won't need an elaborate antenna. Weight and size is of great importance here as the lift and available space is VERY limited,

On one of the RC forums a 12.3mm piece of wire was suggested. I have NO RF knowledge(and little electronics really, especially analogue), so I have no idea whether this will work and how it connects. Do I just solder it to the antenna pin? Does it not need to be connected to ground somehow as well? The transmitter has no documentation with it(China special but popular for First Person View in quadcopters). Some of the web pages warn that running it without an antenna will fry the board, so I want to get this right. I have some 5.8GHz clover leaf antenna, but they are huge for this installation.

I found this: length in meters =300 / frequency in MHz
Divide the result by 4 to get a quarter wave antenna.

In my case: 300/5800 = 0.051724138 meter = 51mm for full wave
0.051724138 / 4 = 0.012931034 = 12.9mm for quarter wave
That seems to make the suggested length of 12.3mm wrong.
How critical are these lengths? It looks like very critical. 5.8GHz is really the middle freq of 8 channels, so not dead accurate.
Does the thickness of the wire matter? Will a piece of bell wire do the job?

In the calculation, what makes the 300 the magic number?
Found the answer to the last question: It is the speed of light in m/s divided by 1000000 so the frequency can be expressed in MHz(rounded a little).
To be pedantic the equation is 299 792 458 / 5800000000 = 0.051688355 m
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
300 is the number of millions of kilometres light can travel in a second.

You only need to be accurate to about 0.1mm at this frequency. Preferably a little better.

The actual wire length will depend on the speed of propagation in the conductor. For a vacuum, that has a speed factor of 1, for coax, it can be as low as 0.5 (the signal travels only half the speed of light)

Yes, thickness will matter a little, but not as much as length.

(I can almost hear Davenn warming up his keyboard...)
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
Thanks Steve.
There will be no coax between the connector pin on the board and the 'piece of wire' - it will be soldered directly to the connection point. So i presume I can still work with a factor of 1. Maybe the slightly shorter (12.3mm) was adjusted because of lower propagation speed somewhere along the line.

The other question was about completing the circuit as if I solder the piece of wire directly to the connection point, I have no shield to solder to the GND. Will this not fry the board? I looked at how clover leafs are made and both ends of the wire is connected, one to the antenna pin and the other to gnd(as I understand it).
Correction: The shield of the cable to the clover leaf is connected to ground.

Found this calculator to approximate range: Range calculator
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The speed referred to is the speed in the conductor. In the case of the antenna, it's the speed in whatever the antenna is made from.

There are some antennas made using coax.

It is possible the propagation speed in a single piece of wire is less than C (even a single piece of wire has distributed inductance and capacitance).

Some antennas are shorts at DC, others are open. So in one case the ground will be connected to the same piece of wire as the centre pin, in other cases it will be insulated from it.

At low powers it is unlikely that you'll damage anything by connecting a poorly matched antenna.
 

pebe

Sep 3, 2013
83
Joined
Sep 3, 2013
Messages
83
The 12.3mm given by the makers is right. Signals travel slower in a piece of wire than in air because the wire has inductance. The ratio between the two is the 'velocity factor' and the figure of 0.93 they have used is about right.

Use a piece of wire and connect one end to the to the aerial pin (no connection to the other end). The wire should ideally be vertical, so you can either use stiff wire from the top of the quadcopter pointing skywards, or a thinner flexible wire dangling from the bottom.
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
There is an end effect so the wire is effectively a little longer than its physical dimensions. A thin wire will have litle end effect, a fat one maybe 5% at much lower frequencies.

Metal clobber near the antenna will have an effect on its effective length.
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
Thanks people for your replies. Much appreciated. Now I'm off to do some micro surgery on the quad. I will post back as to results.

Note: I might be stuck with the antenna being horizontal because of space constraints for running coax. I'm still waiting for the receiver to arrive.
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
The 10mW transmitter HERE is supplied with a 38mm antenna, which does not fit any of the calculations as it is too short to be a full wave antenna and much longer than a quarter or half wave one.

So it all still seems like black magic. I will try with a 12.3mm wire and see how it goes.
Maybe if I say abracadabra when I switch it on :D
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Five eigths of a wavelength is a popular size.
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
OK thanks. Still black magic.
I might try a half wave at 24.6mm as I am instinctively more comfortable with binary multiples.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The various antenna lengths will result in a different characteristic impedance.

google "antenna characteristic impedance" to find lots of discussion about the black magic (otherwise known as complex mathematics) behind these fancy numbers.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I would think that a quarter wave antenna would be best. Longer antennas will have some directional characteristics which would be undesirable.

Make an antenna a little long and trim it a mm at at time to find the maximum output. You will end up with a wire which is short so replace with the the correct length.
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
Thanks Steve, Duke:
It is a whole new project on its own to learn about RF and antenna. I will embark on some serious reading once I have this quad thing done(even though I won't understand/explain everything that made it work until after the reading).
You guys have been a great help.
 

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
Receiver arrived, so I tried it all.:)
Results: 12.3mm antenna: lots of interference. Vertical lines on screen. Sometimes signal lost.
12.9mm antenna: much better with no drop outs,but still some interference(but I am sitting next to 2 wifi routers.:)

Problem is I used a polulu step up module(should have been step down but ordered wrong one) that goes to linear step down mode when input is higher than 3.3V(required output). This get super hot as the transmitter draws 330ma.:(

Running at 3.3V the transmitter got really hot. It seems that some versions of the tx5823 support 5V input, so I thought I would give it a try with a LiPo (4.2 - 3.5V), sitting ready to yank the battery off when it got hot....and surprise, it runs a lot cooler. I don't quite understand this:confused: , but I am happy as it means I dont need the stepdown module, saving a whole 1 gram.:)

Now I am waiting for the LiPo to charge and then I will go and see how it performs away from WiFi routers and see what sort of distance I can get.

I might play a bit more with different length antennas.

What was that about low cost aircon on the other thread:D It still heats up to 79C after a bit. Trouble is, I have applied thermal paste between the metal shield and the chip that gets hot, but it is far away and the paste is not effective. Most of the heat is detectable on the PCB side, which also has vias, so a heatsink is not an option without mica insulalion and that(heatsink) will add significant weight. Any suggestions on how to make it run cooler.
 
Last edited:

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
really meant to get involved with this thread earlier as I have done much video and voice work on 5.8GHz with my amateur radio gear
A crashed computer twice in 2 weeks has had me tied up

do you really need an omni-directional antenna or can something with some directivity do ?

if you can go with a directive antenna, there are many choices one of the easiest being a patch antenna.
Here is my 5.8 GHz quad patch that I have annotated with dimensions

attachment.php


cheers
Dave
 

Attachments

  • 5700MHz Quad_Patch Dimensions.jpg
    5700MHz Quad_Patch Dimensions.jpg
    73.5 KB · Views: 13,394

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
I dropped out of the thread and happened to reread the thread title

200mW .... gosh, even into a coax dipole 200mW will get you at least 1km with a video signal. many kms with NBFM voice

that patch antenna in my previous post and 200mW would give me ~ 100km at 5.8G

to have a noisy signal over 100m you must have a problem or 2

firstly ... is it a clear path ( line of sight) between TX and RX
at 5.8G a LoS is pretty essential
secondly ... are you REALLY sure you are producing 200mW ?

if you really must have omni-directionality, then a coax dipole would be the easiest

preferably using Teflon coax, strip off 2.6 cm (26mm) of the sheath and then fold back the exposed braid down over the existing sheath and flow some solder over it so it doesn't unravel.
the centre conductor doesn't need to be stripped out of its Teflon dielectric insulation

presto ... 1 dipole ... no signif gain, but that's the price you pay for a basic omni antenna

omni gain requires a more complex co-linear antenna, or you go with directive gain as with my patch or a Yagi or horn antennas

cheers
Dave
 
Last edited:

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
@davenn:
Thanks for all that info.

Unfortunately it cannot be directional as it is a video transmitter on a micro quadcopter(drone), so orientation is changing all the time. Can I use a patch antenna on the receiver? OK found that I can use a patch antenna on my receiver. Now have to dig out my etching gear - I am packing to move to South Africa, so will have to find the relevant box LOL. Current receiver antenna came with it and is a rubber ducky. Also ordered 2 clover leaf antenna from China as these seem fiddly to make - depending on size I might be able to fit one to the chopper.

The tests I have been doing is with transmitter and receiver about 600mm apart. I have not done any distance tests yet. The interference might be the hoojup of the camera to the transmitter or the receiver to screen. I will dig some more.

The transmitter draws about 370ma at 3.3V (1.22W) and gets extremely hot very fast at this(rated) voltage. I pulled the plug at 79C. At 2.9V it draws about 130ma and runs at a sustainable 45-50C without heatsinks (which is what I want to save weight).

Assuming the same efficiency(I don't know if I can), it means that if 370ma gives 200mW then 130ma should be roughly 65mW which should still do 200m without a problem as some of the RC guys are getting this range with 25mw transmitters. On desk tests the picture quality doesn't deteriorate at the lower voltage.
I use a piece of bell wire, 12.9mm long, directly soldered onto the pad on the transmitter module as an antenna - it gave the best results so far.

Now i Have to find a LiPo 1S (3.5 to 4.2V) to 2.9V stepdown in a micro package. I will look at the adjustable Polulu(HERE) converter, but it must not generate too much heat as it will be installed in a very tight space with limited ventilation. I hope it is not too bad as it has to dissipate about 100mW i figure((3.7- 2.9) * 0.13.
 
Last edited:

shumifan50

Jan 16, 2014
579
Joined
Jan 16, 2014
Messages
579
Just noticed you have had a crashed computer. Sorry to hear that. However. install Ubuntu Linux, and unless you have a hardware failure, this will not happen to you again. Lots of Windows programs run happily in the windows emulator(Wine) on Linux.
 
Top