|Dec 16, 2009, 10:54 PM|
Build your own TX antenna for FPV
Building your own antennas for FPV use
This is a basic guide on how to build and calibrate your own antennas for use in FPV. For anyone who is handy with a soldering iron and a calculator, making these should be a breeze. I will not be going into extreme detail as there are hundreds of papers on the designs of these antennas. However I will explain their construction, and most importantly, how to connect them and how to test them without specialized equipment. I will be covering 3 basic antenna types:
The basic half wave dipole
Half wave whip
I will discuss the limitations, design considerations, and best applications for each of these antennas.
I also have a tutorial on how to build your own Yagi (or Moxon rectangle) antenna here: Home brew Yagi page
For those who wish to build a patch antenna, Mr. RC Cam has an excellent how to on his website titled The GP Patch antenna. I will refrain from reposting it here.
|Dec 16, 2009, 10:55 PM|
Why bother building your own?
Why bother building my own antenna?
There are a few reasons for doing this:
1. You can build an antenna to suit your application better
2. You can build a better matched antenna for better efficiency
3. General satisfaction
4. Better fit to the airplane
5. Lighter and more durable antennas are possible.
6. Less servo jitter
7. Faster GPS lock (if using GPS system)
8. Longer range
9. Cooler VTX unit
|Dec 16, 2009, 10:56 PM|
A few definitions first
A short description of terms and definitions:
VSWR (Voltage standing wave ratio) also known as SWR, it compares the impedance of the transmitter to the impedance of the antenna. Perfect match is a 1:1 ratio. Power output begins to drop considerably at 2:1 and greater. Note, this does not indicate resonance! If you attach a resistor with the same impedance to your transmitter and measure SWR it will come out 1:1, yet a resistor is NOT an antenna!
Coaxial cable A single conductor with a surrounding shield and a jacket. Typically high loss and thus is best used in short lengths. Some common types are RG-58 and RG-316.
Gain Indicates the directivity of the antenna. The higher the gain, the more directional the antenna.
Impedance Resistance of the antenna or load expressed in Ohms.
Propagation Transmission of a radio wave.
Polarization Orientation of a radio wave. Polarized antennas communicate best when the polarizations are in line (ie vertical to vertical).
Multipath interference interference generated by your own transmitter due to signal reflection being received at different phase angles
|Dec 16, 2009, 10:57 PM|
About Gain and Power
About Gain, power and polarization
Dbi (gain) – This is perhaps the most commonly misunderstood concept. This indicates the transmitters wave compression compared to the idealized isotropic node antenna (which cannot exist). An isotropic antenna would have a spherical pattern. Higher gains have more directivity (compression), where lower gains are less directional. HIGHER GAIN DOES NOT MEAN HIGHER OUTPUT POWER!!! It simply means the antenna is more directional. Lower gains will have better general coverage where higher gains are more suitable for long range. However in the airplane’s case, it is constantly changing orientation and thus high gains can in fact be detrimental especially on the TX.
For example – Take the basic 2.15 dbi dipole. The radiation pattern looks like a doughnut with the antenna running through the center. Now let’s increase the gain to 3dbi (such as a full wavelength dipole). The doughnut gets flatter, but now has a larger diameter. Now let’s go way beyond reality and go for 100 dbi. It now looks like a flat washer (I’m ignoring the lobes concept here) with a very large diameter. That’s great for range, but what happens when you need to turn? What about when you climb? Your pattern now completely missed your receiver! Thus gain isn’t everything.
On the ground, this is very different. Your ground station is static and unmoving. Thus you can point it in the general direction of your airplane and pick up signals with little difference regarding the airplane’s orientation. However extremely high gains are very direction and in most cases should be avoided except for when making a distance run or when you have a tracking system that can keep your antenna locked on target.
Basically gain is the directivity of the antenna. The higher the gain, the more need there is to aim it.
Power is simply the power output of your transmitter. Some people make the assumption that twice the power yields twice the range. This is incorrect! Transmit power falls off at a factor of the distance squared. Thus doubling your power increases your range by about 40%. Twice the power means twice the signal strength received for a given range. In order to double your range you must increase TX power by a factor of 4 or use a more directional antenna.
About polarization - This comes from Thomas Scherrer. I couldn't describe it better.
Radio Technical stuff
Any wireless radio system contain of a transmitter side antenna and a receiver side antenna.
Both sides must have same polarization to perform most optimal, the most common polarizations are horizontal, vertical, circular left or right.
If a horizontal is “talking” to a vertical, the link loss will have an added extra loss of 26dB
If a circular left is “talking” to a circular right, the link loss will have an added extra loss of 26dB
If a circular left is “talking” to a vertical or horizontal, the link loss will have an added extra loss of 3dB
This is why it is smart to combine a horizontal or vertical often mounted in a plane, with a circular receiver on the ground,
then the 26dB drop can most likely be avoided.
|Dec 16, 2009, 10:59 PM|
Building the antenna - Calculating your quarter wavelength
Building the antenna - Calculating your quarter wavelength
The first step in building any antenna is to calculate your effective wavelength. Many of us know the old formula: 2.98 * frequency in MHz = Wavelength in meters. This is the wavelength of a signal in free space, however we are interested in the wavelength in our transmitters. Thus we must include a “velocity factor”, which is usually around 95% of the actual calculated quarter wavelength in free space.
Everything can be determined based on a quarter wave antenna. I highly suggest using the exact frequency you plan to transmit. Why? Varying from the frequency causes antenna mismatch resulting in reduced VSWR! This is why the store bought antennas aren’t always the best match. They are made for a small range of frequencies.
Below are the formulas to calculate your active quarter wave assuming a 95% velocity factor:
Length in Inches (for 22AWG wire) = 2808/frequency in MHz
Or for you metric folks:
Length in Centimeters = 7125/ frequency in MHz
Or for 12-14 AWG:
Length in inches = 2750/f in MHz
Or on Centimeters = 6985/f in MHz
Alternately, you could go to any of these sites:
Thus for my 910 MHz system, my quarter wave is 3.086 inches or 7.84 centimeters.
Note: In all of the antennas designed this number will be the length of the ACTIVE part of the antenna. The length inside the coaxial line or inside the coupler is not considered active. It is active once the conductor leaves the shielding.
|Dec 16, 2009, 11:07 PM|
The good ol' half wave dipole
Half wave dipole antenna:
This is by far the most commonly referenced real antenna. It has excellent efficiency and has a simple doughnut shaped radiation pattern. The gain of this is approximately 2.15 dbi. So it has fairly wide coverage. This is best suited for long range and altitudes. For range and general flying, it should be mounted vertically. However for high altitudes directly above your ground station, this is often better mounted horizontally. Change the antenna to suit your flying.
Coaxial fed dipole
This is the easiest to make and the one which I get the best results. You need RG-174 50 ohm coaxial cable and an SMA fitting for RG-174. You can also use RG316, which is more tolerant of heat.
Strip 3/4" of the jacket off of one end of the coaxial cable. Then strip 1/4" from the center element in the cable. Solder the center element to the pin. Then insert the cable with the pin and fold the shielding over the outer end of the SMA plug. Solder the shield to the outer part of the plug. (Note: you can also buy an extension cable and cut it instead of doing this step)
On the other end of the cable, strip 1/2" off of the cable and then strip 1/4" off of the center conductor. Pull the shielding around the center conductor and bend the center conductor 90 degrees. Now solder these to a proto board (you can get proto boards at Radio shack) so they run in opposite directions. Now solder on your elements (I used 12 AWG solid wire). One element is connected to the shield, the other is connected to the center conductor. On principal, the white is element is always connected to the center conductor and the black is always connected to tht shield. Each leg should be exactly your 1/4 wave measurement from the center of the antenna. The antenna should therefore be 1/2 wave
If you plan to optimize your antenna, you can leave the elements an inch longer or so and trim using the RSSI method described later.
Direct connected dipole
The half wave dipole is constructed from two separate quarter wave antennas oriented 180 degrees apart. The angle doesn’t need to be perfect. So don’t worry too much about it.
To build this I’m using 22 AWG solid wire, an SMA plug, and a ball point pen. The insulation on the wire doesn’t matter. This wire is available online or at your local Radio Shack store. The SMA plug is also available at radio shack or www.dpcav.com . You can use an antenna tube or any narrow diameter plastic tube as a substitute. Eve a coffee stir works. DO NOT USE CARBON FIBER!
Cut off 2 pieces of wire about 2” longer than your calculated quarter length (you will trim this later). Solder one wire to the needle of the SMA plug, solder the other to the outside of the SMA plug. Wrap the wire just above the needle in electrical tape until the wrapped diameter fits tightly into the SMA plug. Insert the needle into the plug until the entire narrow portion protrudes from the base. Bend each wire 90 degrees so they are flat and in line with one another. Cut each wire to exactly the quarter wave measurement from the bend. If you plan on tuning this antenna as described later, cut it to ½ to ¾” longer than that measurement.
Remove the ink cartage from your ball point pen and evacuate the ink. Melt or slice a small section about ½” long down from one end. Now slide the tube over the active antenna wire (the one connected to the needle) until the wire is completely encapsulated. The wire should protrude out of the bottom of the tub where it goes into the SMA plug. Slide another tube over the other end of the antenna and then secure them in a straight line.
You may also build this antenna on a circuit board and feed it directly with coaxial cable. In this case simply strip the coaxial cable and feed one end to each wire. The wires should be less than ¾” apart and the same lengths as your quarter wave calculation. Slide your tube over each end and secure.
EDIT: Alternately, you could build Mr RC Cam's dipole using a length of coaxial cable and stripping it back. It is truly an excellent design. Total length would be exactly twice your quarter wave length.
|Dec 16, 2009, 11:09 PM|
Half Wave whip
Perhaps the simplest antenna, but the least efficient of those described herein. It typically needs a base reflector to operate at peak efficiency. However, there is not one available in the airplane so we’ll do without. The radiation pattern and gain are similar to the dipole. You can hack the junk antenna that came with your TX for this antenna.
You will need a small tube (such as a ball point pen), an SMA plug, and a piece of 22 AWG solid wire. Cut a length of wire 2” longer than TWICE your quarter wave calculation. Solder this wire to the needle of your SMA plug. Place the needle inside the SMA plug so the narrow portion completely protrudes from the plastic in the base. Trim your antenna at exactly twice your quarter wavelength calculation from the top of the SMA plug barrel (where the wire exits the connector). Place your tube over the top of the wire down inside the SMA plug and secure.
Alternately, you can modify the junk andtenna that comes with the TX as described in the Modifying the junk antenna thread.
|Dec 16, 2009, 11:10 PM|
The inverted Vee
I have been getting raves about this antenna on people's airplanes and many success stories. It is my highest recommendation that you try it.
This is perhaps the best general purpose antenna. Suitable for medium range and altitude, this antenna has lower isotropic gain (2.15-1.5) however has a very uniform radiation pattern. It allows for best signal when doing stunts such as loops and rolls without loss of signal. Polarization is also less of a problem. Higher gains are achieved at higher angles (closer to flat like the dipole). Lower angles up to 90 degrees increase the radiation patterns resemblance to a sphere. Im going with 120 degrees as it has the best characteristics for my purposes.
This is simply the dipole antenna with a lesser bend in it. There is however a degradation factor for the angle. Refer to the above construction of the dipole, then use the below values to trim your antennas to the proper lengths. See the diagram for how to feed this. I no longer use a direct feed, but an offset feed. You can see this on the tail of my P-38 airplane in the pictures. Note how the coaxial cable comes out at an angle.
So the antenna quarter wave lengths are calculated as:
Quarter length * degradation = Proper antenna length.
For 180-165 degrees 0%
You can also use this calculator: http://www.cdxa.org/antenna_cal_diople.html
Empirically, the proper element length (already compensating for the degradation factor) comes out to: Length in inches = 2715/f in MHz - or - Length in mm = 69200/f in MHz
Note that the elements are separate. One is connected to the driven pin, the other is connected to ground. This is the same antenna as the Dipole but just bent at an angle.
Mount this with the open side of the Vee facing the nose of the airplane when flying distance as this will give a stronger signal for the return trip. Mount it with the Vee facing sideways for stunt flying As this will help keep the antenna in polarization better.
This is a video of my Vee antenna with an 8dbi patch on the ground. 2 miles with plenty of strength! This is why you build a homebrew antenna!
|Dec 16, 2009, 11:11 PM|
Obtaining higher gain odd multiple antennas
Obtaining higher gain odd multiple antennas
The 3/2 wave antenna:
These are very similar to the ½ wave antennas mentioned above such as the inverted Vee and the ½ wave dipole, but they are three times as long. This greatly increases the gain but not necessarily the efficiency of the antenna. Similarly, you could use a 5/2 or a 7/2 and the input impedance would be very similar and you would achieve higher gain.
Less sensitive to frequency ranges these are easier to tune and have a larger bandwidth of operation
Higher gain You can achieve longer transmission distances.
Slightly more directional
Larger and more cumbersome
For the 3/2 wave, instead of measuring your quarter wavelength, you instead use a ¾ wave length. The formulas are as follows:
¾ wave length in inches = 8424/frequency in MHz
¾ wavelength in centimeters = 21375/frequency in MHz
Constructing these is exactly the same as the ½ wave dipole and inverted Vee except your antenna length is 3, 5, or 7 times as long. Im going to assume that the majority of people will not go to 5/2 or 7/2 wave due to size limitations and directivity problems, so Ill post the formula for only 3/2. If you want 5/2 or 7/2, you can figure it out as it is easy.
|Dec 16, 2009, 11:14 PM|
Tuning your antenna to the exact frequency
Note that this step is generally uneccessary if you used a caliper and were very precise on your measurements. However it is great for assurance that all is working properly.
Now that you have built your antenna, it is time to tune it in to your frequency. There are multiple instruments out there that can help you with this, however I’m going to assume that most people don’t have them and don’t want to fork out the money for them. So these tests will be done with only the TX and RX and a multimeter.
RSSI Method (most accurate, but more difficult)
How to test:
This test is best done outside. You are going to need to locate the RSSI (received signal strength indicator) pin on your RX. This will tell you the strength of the received signal. It is also known as AGC. It is often pin #8 on the receiver module (counting from the bottom side moving towards the antenna output). it has an output varying between about 5V for full signal to .1V for lost signal.
Separate the TX and RX by the farthest distance possible. 100 yards (or meters) is plenty. Even 100 feet should be ok. Connect power up to your TX with the camera , GPS and all equipment (you don’t need to power up the airplane) connected. Connect a known good antenna to the TX. Make sure both antennas are in the same orientation. If you have a directional antenna on your RX, aim the antenna away from the TX but keep it in the same polarization (ie both vertical or both horizontal). Make sure you have a clear video signal. If not, either bring the antennas closer together or aim your RX more toward your TX.
Hook up you multimeter so that it is as far away as possible from the receiver. When taking readings, you want to be as far away as possible from your multimeter. You’ll want to stand in the exact same place each time you take a reading and do not disturb anything on the receiving side. So it is best to set up your recieving station and then not touch it until the test is complete. It is also a good idea to have a pair of binoculars to take the reading even farther away from the RX to get the most accurate measurement.
Note your RSSI voltage with your control antennas. Write this number down. This will be your control reference number. Ideally this should be about 1.5-2 volts or so, but up to 3 is acceptable. If it is too high your antennas are too close and are seeing too high a signal strength, thus you will not be getting the most accurate readings.
Now replace the antenna with your antenna that you just made keeping it in the same orientation as the receiving antenna. The ends should be between ½” and ¾” longer than your calculated quarter wave. Again record your RSSI voltage as well as the antenna length. Try to be as accurate as possible. Then go out and trim 1/16” (2mm) (or less if you want more accuracy) off of either end of the TX antenna. Again record your RSSI voltage. Repeat this process until your antenna is ¾” shorter than your calculated quarter wavelength or until you notice a significant drop in RSSI voltage. Remember to stand in the same place as far away as possible from the RX and not between the TX and RX.
Graph the RSSI voltage VS antenna length via a spreadsheet program. The peak point on the voltage is your best antenna length. Somewhere in this process, your peak RSSI voltage should be very close to if not greater than the RSSI voltage you recorded on the control antenna if you used a 2dbi control antenna. Rebuild your antenna trimming the ends to this exact length. You can verify that it has peak strength only if you didn't disturb your test set up. If you disturbed your test set up in this process, the RSSI readings will be different.
Current measurement (easier)
Another way to tune your antenna is to watch the current in your VTx transmitter. When the current is at a minimum, the antenna is the correct length. When tuning this way, be sure the antenna is as far away from anything metallic as possible. I recommend putting it on an empty wooden table in the middle of the room. Below are the currents I measured for a 910MHz Vee on a 500mW transmitter:
Note that the calculations on the first page of this thread state that 2.98" is optimal length for the elements.
3.55" - 278mA
3.41" - 274mA
3.36" - 265mA
3.27" - 258mA
3.22" - 251mA
3.165" - 241mA
3.12" - 233mA
3.075" - 230mA
3.045" - 225mA
3.008" - 219mA
2.97" - 214mA (Calculated resonant length)
2.95" - 211mA
2,92" - 213mA
2.89" - 213mA
2.85" - 214mA
2.80" - 219mA
Note that the numbers stabilize, then start climbing. When the numbers begin to climb stop trimming and your antenna is ready.
You now have a tuned antenna! Go fly.
|Dec 16, 2009, 11:27 PM|
Feel free to post questions here or just send me a PM. I unfortunately do not have an SWR meter to read the SWR of my antennas, however I was finding my RSSI voltage with my tuned antenna a bit higher than with the dipole and rubber duck antennas I had bought. The junk antenna that came with the unit had considerably poorer performance.
Also note that the surroundings make a larger impact on the RSSI voltage than the antenna length when testing in short range so be sure to keep anything that generates or absorbs a signal such as your body or a cell phone, as far away as possible and in the same location for each reading to keep from skewing the data.
I will add more pictures as well as some field testing (or should I say air testing) in the next few days.
Good luck with this for all who try.
Coaxial cable 1/4 wave Myths:
I have heard in numerous places that you need to avoid lengths of coaxial cable equal to an odd multiple of your quarter wave. This is a myth that started back in the early days of CB radio. So long as you use 50 ohm coaxial line you will be ok at any length that will fit on your plane. I would recommend keeping it short only to reduce spurious emissions and keep it light weight. Any type of coxial line will work, so choose whatever suits you best. For short lengths RG-58 is just as good as RG316.
Yes I did neglect to mention antenna polarization. Remember that your transmit and receive antennas should be in the same plane (ie vertical recieves vertical, horizontal receives horizontal). This is why you need to mount your antenna upright for range and both antennas horizontally for altitude. Also, the inverted Vee is less polarized due to the angle. I mount mine upright with both ends trailing forwards at about 30 degrees from vertical. Why forward? Signal strength is greater at the smaller angle. I need video reception to come back. If I lose video feed on the way out and have my antenna pointing backward, I will possibly not have it for the return trip.
Ever wonder why you lose the video feed in a sharp bank? This is mainly due to polarization. If the antennas are 90 degrees to eachother, then reception could be as low as 2% strength! Thus, I recommend the inverted Vee antenna for flying aerobatics. It is not as polarized being in 2 separate planes. This can also explain why a patch antenna is particularly useful at a sharp angle to the ground (It is polarized as well).
Why whole wavelengths won't work:
The reason is impedance matching. A full wave dipole has an impedance of about 300 Ohms. A half wave is 70 or so. Our transmitters are mostly 50 Ohm impedance. This means the best SWR we can acheive with a full wave antenna is 6:1. The half wave dipole will top out around 1.3:1 or so.
How are the elements connected?
On principal, I connect my center conductor of my coaxial cable to my white element and connect the shield to my black element. These are separate and in no way touch.
Can I use a different material for a mounting plate?
Yes. Wood or plastic work well. Some people use nothing at all and mount it right to their airplane for support.
|Dec 17, 2009, 07:33 PM|
|Dec 17, 2009, 09:54 PM|
United States, WA, White Salmon
Joined Nov 2005
Man you must be crazy, to go to all the effort put into this awesome thread!! Thanks, IBcrazy! Very helpful and informative.
|Category||Thread||Thread Starter||Forum||Replies||Last Post|
|Discussion||trex 600e build your own fuse.||catbalu||Large Electric Helis||10||Oct 27, 2009 01:10 AM|
|Sold||Build your own brushless motor sale pending||Douglas Cheatwo||Aircraft - Electric - Power Systems (FS/W)||1||Oct 07, 2009 12:26 PM|
|New Product||Build your own two channel servo tester for less than $15!||W9GFO||Product Announcements||0||Sep 29, 2009 08:19 PM|
|Question||Build Your Own Parkzone Mustang Question||Blue Jacket||Electric Warbirds||8||Sep 13, 2009 09:49 PM|