Build IBCrazy's 10dbic CP Double Quad antenna

[IBCrazy]

Well it’s finally designed and tested. My CP Double Quad is here. This is a different design that Old Man Mike’s CP BiQuad, but I feel that it is much easier to build. Before any of you ask, stacking these is possible for even higher gain. I will explain how to do this later once I have that tested.

OMM’s CP BiQuad can be found here: http://www.rcgroups.com/forums/showp...&postcount=406
To scale it to your frequency simply multiply ALL dimensions INCLUDING WIRE THICKNESS by the equation: 1280 * pictured dimension/your frequency

I give you this tutorial and all the support in the name of my Lord and savior, Jesus Christ. I wish all of you the best of success with your FPV ventures. Please enjoy.


View this video on Vimeo


-Alex Greve
Video Aerial Systems LLC
www.videoaerialsystems.com

[IBCrazy]

I designed this antenna by myself using hand calculations based on limited antenna modeling. I was never able to fully model it due to my inability to build a complex antenna on a modeling program. Thus all data to this point is calculated and tested, but never fully modeled as a complete unit. If one of you would like to model this for us, I would appreciate that.

Certain elements of this antenna are more critical than others. The total length of the wire element is critical, but the ¼ wavelength side bends are not as critical. The length of the impedance transformer is critical and so is the height of the cross fed element. The height of the primary is not as important. The reflector size affects the gain. Larger reflector = more gain. A 1 wavelength reflector will result in 9 dbi gain where a 1.5 wavelength reflector will result in 10.2 dbi gain. In this tutorial I used a 1.25 wavelength reflector.

Flight performance is similar to the 5 turn helical antenna. Multipath rejection is slightly less, but efficiency is significantly higher. I RECOMMEND USING A LONG RANGE RADIO SYSTEM WITH THIS ANTENNA. THE VIDEO RANGE WILL BE SIGNIFICANTLY GREATER THAN THAT OF A STANDARD 72MHZ OR 2.4 GHZ RADIO. Additionally, although you will be able to fly behind objects with this antenna, I highly recommend NOT doing it. That is a good way to lose your aircraft.

[IBCrazy]

To build this antenna several calculations need to be made. I will explain the theory on how this works in a later post for those who wish to understand just how complex the system really is. However, simply following the calculations is all that is needed to make a good antenna.

Wire type: Any solderable metal will do. Copper, brass, and steel all work fine. I use 14 AWG wire because it’s easy to bend.

Wire length:
Wire length in inches = 11732/f in MHz
Or for you metric folks
Wire length in cm = 29800/f in MHz

Reflector dimensions (square sheet):
Size in inches = 14600/f in MHz
Or size in cm = 37250/f in MHz

¼ wavelength:
¼ wavelength in inches = 2933/f in MHz
Of for you metric folks
¼ wavelength in cm = 7450/f in MHz

Phase shifting impedance transformer (covered part) where Vf is the cable velocity factor:
Cable length in inches = 8200 * Vf/f in MHz
Or in cm = 21336 * Vf/f in MHz

Primary element height:
Height in inches = 1465/f in MHz
Height in cm = 3724/f in MHz

Cross fed element height:
Height in inches = 900/f in MHz
Height in cm = 2286/f in MHz

For the following frequencies these are the measurements pre-calculated for you
Wire length
910MHz = 12.89”/32.75cm
1280MHz = 9.166/23.28cm
2.4GHz = 4.79”/12.96cm

Reflector dimensions:
910MHz = 16.11”/40.92cm
1280MHz = 11.45/29.1cm
2.4GHz = 6”/15.2cm

¼ wavelength
910MHz = 3.22”/8.185cm
1280MHz = 2.29”/5.82cm
2.4GHz = 1.198”/3.0417cm

Phase shifting transformer made with RG62A/U
910MHz = 7.66”/19.46cm
1280Mhz = 5.44”/13.835cm
2.4GHz = 2.85”/7.228cm

Primary element height:
910MHz = 1.615”/4.09cm
1280MHz = 1.145”/2.909cm
2.4GHz = .598”/1.52cm

Secondary element height:
910MHz = 0.98"/2.51cm
1280MHz =0.70"/17.85cm
2.4GHz = 0.367”/.933cm


-Alex Greve
Video Aerial Systems LLC

[IBCrazy]

This step is fairly straight forward. Simply bend your wire into a square whose sides are equal to your ¼ wavelength calculation. Do not make the bends a sharp point, but let them be rounded a little as this will help antenna performance.

Make two elements

[IBCrazy]

For this operation I used ¼” poplar. I cut the wood down on a table saw. A band saw also works, but I find a table saw to have a much straighter cut. You want to make your wooden supports long enough to hold both elements. Rip them down so their width is equal to your primary element height. Now cut a rectangular section from one side halfway don so that it forms an “L” shape. The bottom of the “L” should be equal to your secondary element height.

[IBCrazy]

For this I used 26 gauge galvanized steel. Find the center of the reflector and drill a hole big enough to fit your phase shifting transformer down through it. Now place your elements on the reflector as they would sit when assembled. Drill your second hold where one of the diamonds falls. Now feed your phase shifting transformer through the holes.


A small reflector of one wavelength will yield 9 dbic gain. A large reflector (as described above) will yield 10dbic gain.

[IBCrazy]

First, glue your spacer blocks down to the reflector. Make sure the split level in the “L” is lined up with the center hole. Now solder on your elements to your phase shifting transformer being certain not to short it out prematurely.



Polarization is determined by how the phase shifting transformer is connected. the best way I can describe it is this way:



Set the "V" of the phase shifting transformers to accept the elements.



RHCP:



If you would rotate your primary element COUNTER CLOCKWISE 90 degrees and the active elements of both ends of the transformer point in the SAME DIRECTION.


If you would rotate your primary element CLOCKWISE 90 degrees and the active elements of the transformer are NOT IN THE SAME DIRECTION.



LCHP:

The opposite of above.


The antenna in the photos is LHCP.


Now glue the elements down to the wooden spacers. Do not let them overlap as this will cause a shift in impedance.


Once that is done, simply solder in your coaxial cable and you’re done!

Now go fly that thing and have fun!

-Alex

[IBCrazy]

To understand this, we must first understand the 1 wavelength loop antenna. The single wavelength loop antenna has a characteristic impedance of 120 Ohms. When this antenna is placed over a ground plane 1/8 wavelength away, the impedance drops 17% or 100 ohms. Since two 100Ohm loads in parallel is 50 Ohms, we can use two wavelength loops 1/8wavelength over the ground plane to obtain a 50 ohm impedance. This is the case in the standard BiQuad antenna.

However there is a problem… You need to phase shift one elements 90 or 270 degrees. Since a 90 degree phase shift is very difficult, 270 degrees was chosen for this design. The best way to do this is with a section of coaxial cable. The unfortunate side of this is that there is no 100 Ohm cable available to match our 100 Ohm loop. Technically I could make my own transmission line for 100 ohms. However I am certain that I could not keeop the tolerances required for such a line. Not only that, it would be far too difficult for the average builder. Thus 93 Ohm cable was used as it is the closest to the 100 Ohm load. Unfortunately the cable also acts as an impedance transformer as it is not ½ wavelength, but closer to ¾ wavelength. Thus the second loop would have an 86 ohm impedance at the beginning of the coaxial feed. This imbalances the load and hurts SWR.
The second loop impedance at the transformer is given by the equation:
Z load = Z cable^2/Z loop

We need Z load to be 100 Ohms. This can be adjusted by moving the loop closer to the ground plane which will reduce the loop impedance. We actually need the loop impedance to be 86 Ohms BEFORE the transformer. This will make the transformer increase the effective impedance to 100 Ohms. The height at which this happens is approximately 1/20 of a wavelength away.

Great that’s it, right? Well, no. Now you have a new problem… The elements are now different lengths away from the target and thus the wave reaches them at different times. This disturbs the antenna circularity. Since the wave hits the second element about .075 wavelengths late, we can shorten the cable by .075*Vf wavelengths and achieve perfect circularity.

Are we done yet? No. This goes right back to problem #1 – the impedance is no longer correct! Why? The coaxial cable is shorter and therefore the impedance shift is no longer a nice algebraic equation, but now a second order differential equation to solve for the proper loop impedance.

I will refrain from posting the equation for solving this, but it is the inverse tangent of a second order differential equation. Needless to say, I have no idea how to solve it. I can post it later if one of you is a math genius and wants to figure it out. Perhaps you could use iteration using an MS Excel spreadsheet similar to an FFT function.
So how exactly do I design this part? Admittedly, I guessed. Yep, the good old fashioned guess, check, revise method. I knew the impedance of the loop changed greatly with small variations in height over the reflector once you get very close. Thus I left the 1/20^th wavelength height constant. Eventhough I need to transform the impedance here which involves changing the height, the height change is minimal compared to the impedance shift resulting from this small change. So now since my height is approximately constant, I can get my cable length, which I already calculated above.

I though you said you weren’t done? Correct. I’m not. I still need to change the impedance of that second diamond loop. While changing the length could actually do this, it would also result in a phase shift which will inhibit circularity. Circularity is the key to an FPV antenna working properly so now the height must again be adjusted to obtain proper SWR.

This is where my trusty friend Daiwa SII comes in handy. I know that when I hit a perfect 1.0 SWR, this antenna will be very close to perfection. Thus, all I did was slowly adjust the height and take SWR readings. The great thing is that the height actually increases, not decreases making the adjustment less fickle! Once I hit perfect SWR I have a choice: I can call it quits and figure it’s close enough, or I can make a new cable a little longer to compensate for the new height and do the calculations all over again. Since I chose a 93 Ohm coaxial cable which is very close to the 100 Ohm load I need, I stopped here. Another iteration would come up with a measurement that is so close that I can’t accurately measure it. However, if I used 75 Ohm cable for the phase shift transformer, another iteration would be necessary.

There is one last problem if you guys who are geometry savvy might have figured out by now: the two elements will only receive the signal at the same time when the airplane is right out in front. As the planes ventures out of the beam, the circularity rejection drops as the wave now doesn’t hit the element in perfect sync. The -3db point on this antenna is only 25 degrees. It is a narrow beam long range antenna. If you do your trigonometry assuming the center of each element is where the signal emanates from and draw your lines out to point in space along this imaginary -3db line, you will find the phase shift is minimal. At 75 degrees, the antenna loses all traces of circularity and becomes a dual polarized antenna and results in a -3db drop. However the signal strength here is already -25 db. Once you venture this far out of the beam loss of circularity is the least of your worries, you’ve lost signal well before this. That’s why this is a long range antenna. Getting to the sides is ok at very close range only.


-Alex Greve
Video Aerial Systems LLC
www.videoaerialsystems.com

[troynh]

Alex,

I noticed that the high element is positioned differently in the video vs your still pictures. Which should it be? Does it matter?

Thanks,

-Tim

[BloomingtonFPV]

Does the wood backing play any role in the antenna?

Would these be appropriate for the antenna tracker if I used two at 50 degrees from each other?