A big problem on small satellites (cubesats) is how to match antennas.
Simply connecting a network analyzer, or MFJ antenna analizer or any other test equpment won't work. Connecting a cable to the satelltie model will completely upset the near-field and any useful measurements.
Especiallly, trying to match a 2 meter antenna to a 4" cubesat!
I wonder if the best way is to simply do the best you can with EZNEC or other modeling tool, then build a true fidelity model spaceframe with the fully designed and implemented mathcing network, but reserve one side of the model to have a field-strength meter on it connected internally to the actual TX/RX cable to the antenna.
Then using long insulated dielectric "tweakers" just tune the matching circuit for max. (Of coures getting out of the way after each tweak).. Also, then comparing the results with a similar field strength meter and tuned dipole as a reference at the same distance.
We tried lots of methods, including building in a directional coupler and having it report its reflected power via telemetery... but I think the one above is what I will do next time.
When the spacecraft is small, relative to a quarter wave, everything near the spacecraft and any wires/cables have drastic impact..
Has anyone had other successes? Just thinking what we should be telling these students that are building cubesats. A 6" whip for 70cm is not too bad on a 4" cubesat, but 2m is compltely different...
Bob, WB4APR
At 12:12 PM 5/4/2008, Robert Bruninga wrote:
A big problem on small satellites (cubesats) is how to match antennas... ... Has anyone had other successes? Just thinking what we should be telling these students that are building cubesats. A 6" whip for 70cm is not too bad on a 4" cubesat, but 2m is compltely different...
Hi Bob,
This problem is not new. Walt Maxwell, W2DU had the same problem when working on antenna for big satellites at RCA Astro-Electronics. His solution was to use ferrites to decouple the feed cable from the satellite body. Then you can just use a directional coupler outside the satellite to measure the return loss.
I have used this and it works pretty well as long as you keep the feed cables at 90 degrees from the antenna E field. You can use the EZNEC model to determine the expected E-field.
You need to make sure the ferrites provide at least 10x the impedance of the antenna at the satellite (i.e. if you expect 35 ohms for the antenna, you need at least 350 ohms from the ferrites.) Also, you MAY need to break up any resonances on the cable by putting ferrites every 1/8 wavelength on the feed cable if you cannot get the cable at a true 90 degrees.
Note that 90 degrees will not likely be directly broadside from the satellite especially for a 4" cube on 2 meters.
73, Tony AA2TX
participants (2)
-
Anthony Monteiro -
Robert Bruninga