|Dec 09, 2010, 08:56 PM|
The Skew-Planar Wheel circular polarized omni antenna
These antennas are now available as DIY kits and fully assembled units. Kits can be ordered from: Video Aerial Systems and fully assembled units are available from: Video Aerial Systems and Ready Made RC.
With all of the attention circular polarization is getting, it's about time for a circular polarized omni directional antenna. So far there have only been two in the FPV world: Old Man Mike's CPOD and the Skew planar wheel first built by Nigel (Devonboy).
I had tried to develop a turnsile antenna with a little help from blupp which we called the "bluppnik" based on the attic antenna. While quite robust, we found it to be not useful as a TX antenna.
However, with the availability of circularly polarized patches in the 2.4 and 5.8 GHz band, it only made sense to pursue one that could be built for these bands.
There are other tutorials on this antenna. I found this one to be quite good: http://www.igs.net/~graham/SkewPlanarAntenna/
Warning: This antenna is very delicate and isn't very easy to build for a first timer. I would suggest building an inverted Vee first if you are a first time antenna builder. The reason it is so difficult is joining 4 wire ends to a single point on the coaxial line is rather difficult. I have a few ideas on how to overcome this, but showing how to do it will not be easy.
Lawmate RX problem: - The skew planar wheel does work with Lawmatre 1.2GHz recievers. This problem was unique to my particular setup which I fixed easily.
Demo video: This shows the advantages of a circularly polarized system. The performance difference is staggering.
EDIT: Geometry video by Terry74:
|Dec 09, 2010, 09:01 PM|
You need two items:
- Some sort of stiff wire conductor (2 lengths of it just over 2 wavelengths each)
- A length of coaxial cable
You will want to use a good stiff wire. Copper is a bit too soft. I tried music wire, but I had to weld it (not solder but WELD) to get it to stick together. .030-.045 mig welder wire is great and you can probably get it from your local machine/weld shop for free
OldManMike used gold plated "memory wire" from Hobby Lobby.
I used .030" welding wire for the wheel and RG316 coaxial cable for the feed.
|Dec 09, 2010, 09:18 PM|
Calculations - Your element dimensions
The antenna is made from two lengths of wire whose lengths are:
Wire length in inches = 23675/f in MHz
Or for you metric folks
Wire length in mm = 601345/f in MHz
910MHz = 26"/660mm
1280MHz = 18.49"/469.8mm
2.4GHz = 9.66"/245.5mm
5.8GHz = 4.08/103mm
Now you need to calculate your quarter wavelength which is simply 1/8 of your wire length.
Quarter wavelength in inches = 2960/f in MHz
Quarter wavelength in mm = 75168/f in MHz
910MHz = 3.25"/84mm
1280MHz = 2.31"/60mm
2.4GHz = 1.2"/31mm
5.8GHz = .51"/13mm
You may notice this is a little longer than the true electrical quarter wavelength such as that is used in the BiQuad tutorial. Admittedly, I don't know why this is the case, but it appears to work best 1-2% longer than the electrical wavelength, so we'll go with it.
|Dec 09, 2010, 09:26 PM|
Building the antenna - Making the S-curve
Now it's time to start building. You will start out by making an "S" curve.
Measure back from each end of the wire one quarter wavelength (as calculated above) and bend each end in opposite directions 90 degrees. You now have an elongated "S" with tips equal to your calculated quarter wavelength.
Next, we complete the "S". Measure back from each bend TWICE your quarter wavelength and bend 90 degrees in the same direction as your previous bend on that end. Do this to both ends to form an "S"
Repeat this on both lengths of wire. They should be identical.
|Dec 09, 2010, 09:34 PM|
Building the antenna - Making the pinwheel blades
Now we need to bend the radius in our "S". I did this by taking my pliers and making small bends all the way around the 1/2 wavelength side until the end met with the middle of the antenna. The tip of the wire should meet the center at a ~105 degree angle. It will not (and should not) intersect at 90 degrees.
Do this to both wires. They should again be identical and should look like a figure "8".
Now take your figure "8" and twist one side 90 degrees along the central axis of the element. One radius should be flat and the other should be upright. Do this to both elements. They should be identical.
|Dec 09, 2010, 09:50 PM|
Building the antenna - Soldering it all together
Now this is the hard part. Soldering this without overheating the coaxial cable is difficult.
First, mark the centers of the pinwheel blades. Now place them over top of each other 90 degrees apart. Each of the blades should set at a 45 degree angle from the "X". Solder the point where the flat lengths of wire cross. No radius should be flat at this point. If one is, you can always rebend after soldering.
You should notice that all of the tips come together at the center at 45 degree angles with the plane of the "X" and 90 degrees apart from each other. Now is a good time to tin the tips if you haven't done so already.
Now you must connect your coaxial cable. While it is probably easier (and certainly better) to do this from underneath with a really short piece of coaxial line I opted to feed mine from the side. The reason I side fed mine is because the antenna is quite large on 1280MHz and I want to lay it flat on top of my wing. Bottom feeding will be more efficient and easier. Side feeding should only be done with 900MHz-1.3GHz.
Strip about 1/2" or so off of the end of your coaxial line and pull back the shield. Wrap the shield up into a straight tail, then wrap the shield around the center of the "X" so the cable rests against the side of the "X". Solder the shield to the center of the "X" being careful not to let the elements move.
Now strip the insulation off of the center conductor leaving 3-4mm of insulation on to prevent shorting out to the shield. All 4 of the tips must be soldered to this. It isn't easy. I did this by turning up the conductor and then held the elements to it by spring tension and soldered them all in place.
That's it! You now have a circularly polarized omni antenna!
|Dec 09, 2010, 09:52 PM|
I will try to answer questions here:
Q: Will you build this for me?
A: Yes. These are part of my "new system" of FPV antennas. As much as I'd like people here to build it themselves, I understand that it is not easy and intimidating.
Q: What is the radiation pattern like?
A: Similar to a 1/2 wave dipole. About 2 dbi or so gain.
Q: Is this right hand or left hand polarized?
A: It depends on which way you bent the pinwheel. It is the opposite direction of your hand. If you curl your fingers in the direction of the bend coming off the center pin with your right hand, the antenna is left hand polarized. Both Old Man Mike's and mine are LHCP.
Q: Why left hand polarized?
A: It's just how it came out. I didn't choose before designing it.
Q: What if I polarized it in the wrong direction?
A: Twist the pinwheel 90 degrees across each other.
Q: Why should I build this antenna instead of all the others?
A: Performance. Circular polarization on both ends (RX and TX) makes a huge dfference in video quality.
|Dec 09, 2010, 10:34 PM|
I just finished up a 2 element quad design for 7.150Mhz that, due to tower height, had to have part of the elements folded back in toward the center. I found that this significantly raised the frequency of resonance, though I don't know why either. I ended up between 7 and 10% longer to get it back where I wanted it. I have yet to come across an explanation for it, but if I do, I will share it.
Of course, we are wavelengths apart in our center frequencies, and particular antenna designs in this case, but I still find it interesting that we both have run into the same issue.
I wonder if there is some sort of loading effect in both cases?
I didn't see a reduction in bandwidth as one would expect with a loaded antenna, but I only got to sweep it a couple of times before I ran out of daylight....
Nice work, BTW!
|Dec 10, 2010, 09:39 AM|
Joined Aug 2009
Good idea to start a seperate thread dedicated to CP Alex.
Another good link regarding the Skew-Planar CP antenna is here:
Both OMM and myself have had some correspondence with the authors of the links above, so please don't everyone with an interest start emailing them separately.
The Skew-Planar or PinWheel as OMM has called it, is very much an antenna in it's infancy and the more people who build and test them, the better and the greater will be our understanding of it.
Of all the CP antennas that might be suitable for the VTx, the S-P has a great omni-directional radiation pattern and near perfect match to 50 ohm coaxial cable.
To take full advantage of the Circular Polarity on the VTx, a same polarity CP antenna is required on the VRx.
If you have a decent CP patch, you're half way there (just make sure you know it's handing and make the VTx the same hand i.e. RH or LH).
Alternatively, a helical is not too difficult to build and even more gain can be obtained from a helical.
CP explained with pictures:
Your frequency numbers are off - from the links the calculations are +4% over normal frequency calculations and I quote:
N.B. Testing and analysis has shown that the elements should be
slightly more than one wavelength long ( WL x 1.0443 ).
The factors to use are: 31329 / F (centimeters) or 12334 / F (inches) long.
This antenna is broad in frequency response. In practice, element lengths are not supercritical.
|Dec 10, 2010, 10:27 AM|
I think you should see how I built it. My numbers are double those numbers. Why? I'm using 2 wires instead of 4 Your factors are for each element. My wire encompasses 2 elements to make this easier to build.
|Dec 10, 2010, 10:56 AM|
Joined Aug 2009
OK - get your idea now.
Only thing I can see is that the elements will be slightly off centred with your method.
I wonder whether this will have a detrimental effect to the CP pattern or not?
From my previous attempts at CP, the slightest deviation or miscalculation and CP is affected.
Both OMM and my builds were with 4 individual elements.
Nontheless, it's good to experiment
|Dec 14, 2010, 01:51 PM|
It's 20g copper wire, so the antenna is very soft. I made it for testing purpose only.
Then, I came with the idea of a radome or maybe even make a demi-sphere mold and encase the antenna is self expanding foam (insulation foam).
|Category||Thread||Thread Starter||Forum||Replies||Last Post|
|Mini-HowTo||DIY Ground Plane omni directional RX antenna||IBCrazy||FPV Talk||60||Nov 19, 2012 06:11 AM|
|Discussion||900 mhz circular polarized antenna?||g00bd0g||FPV Talk||9||Aug 21, 2012 09:54 AM|
|Discussion||Left/Right circular polarized antenna ?||SpookiePower||FPV Talk||44||Jun 04, 2011 06:41 PM|
|Discussion||18dbi Circular Polarized Patch ?||Coyote64||FPV Talk||3||Nov 10, 2010 04:59 PM|
|Question||Circular polarized antenna question||Sov4life||FPV Talk||8||Mar 05, 2010 05:00 PM|