radiation efficiency
I was wondering - Has anyone encountered a problem in calculating radiation efficiency of small antennas (~λ/20) in CST MWS?
The reason i'm asking, is a simulation of a λ/20 dipole I performed in CST which produced more than 90% radiation efficiency (!) - That kind of a result can't be true, right?
Another way to look at my question is:
Are there any -<Special>- settings for small antenna simulation in CST?
Regards,
P.
Hi Pushhead,
I guess it could be reasonable if the antenna is thick enough and the metal is sufficiently conductive. Of course, if/when you try to match the antenna to any real impedance (like 50 Ohms), the matching components will introduce a lot of losses that will significantly reduce your total efficiency. Especially, you will need a big series inductor that for sure will be quite lossy.
Regards,
Peter
Thanks for the reply, Peter.
Well, the radiation efficiency is above 90%, and the total efficiency is ~1% but this is because of the mismatch.
What I meant is that the radiation efficiency should be poor for small antennas. i.e. where Power_out=eff×Power_in, then power_out is very small!
If my CST simulation is accurate (which i'm sure is not) than all you have to do is connect a matcing network to my λ/20 dipoe, and it will radiate exactly like a λ/2 dipole....
I would be happy if anyone could comment on this issue.
Regards,
P.
Hi,
you are absolutely right, there is something wrong in your calculation. I pretty often calculated very short antennas. You have to use lossy material to get a right result. I don't know why but you get a wrong radiation efficiency with PEC. Just pick a slightly lossy material. You might think the difference should not be that big since the dimensions are so small but it has a huge influence...
Let me know if you were successful.
Ciao, Bodo
hi,
I think you should revise the definition of radiation efficiency
er = Radiation Resistance/(Radiation Resistance + Ohmic Resistance)
or
er = Power radiated from the antenna/(Total power that entered the antenna)
the total power that entered the antenna = the power radiated + THe power loss
if you used PEC material, then there is no power loss and the radiation efficiency will be incorrectly high no matter what the size of the antenna was.
I hope this might help,
Sherb
Hey Sherb,
sorry but I cannot totally agree with what you wrote. Lets say we have a dipol antenna with a length of 10mm (which is really very short...). We will calculate the radiation efficiency on e.g. 500MHz. Using PEC the radiation efficiency is higher than 90% (total efficiency is very poor due to mismatch). Then we are using a slightly lossy material instead of PEC, lets say lossy copper. The radiation efficiency then goes down to less than 5%. How much resistance can 10mm copper have on a frequency of 500MHz. No antenna would work anymore if you would have so much loss. The resistance at a Lambda/2 dipole would be even 35 times higher.
The radiation efficiency must be poor on very short antennas because of it's low radiation resistance. Almost all the input power is reactive power that does not radiate. Only active power can be radiated.
If somebody knows why CST does not calculate it right with PEC, let us know... I have no idea.
Wouldn't you agree? ...very intresting topic.
ciao, Bodo
Pushhead wrote:
--------------------
"If my CST simulation is accurate (which i'm sure is not) than all you have to do is connect a matcing network to my λ/20 dipoe, and it will radiate exactly like a λ/2 dipole.... "
---------------------
Yes, you are absolutely correct. If (and this is a big if) you can do the matching network lossless, then the performance will be nearly identical. I say nearly because of two things:
1) The directivity is slightly slightly higher for the λ/2 dipole compared to the λ/20 dipole. In practice, the difference is negligable.
2) Most importantly, the impedance bandwidth will be much higher for your λ/2 dipole compared to the λ/20 dipole. A clever matching network will help, but not that much.
The theoretical problem with small antennas is always, and only, the impedance bandwidth. Of course, in practice, you will also have more losses (e.g. in the matching network) in small antennas compared to full-size antennas.
Best regards,
Peter
hi elomatic,
Actually I see no problem with what sherb said and what you said. For small antennas you are correct their ohmic resistance will be small, but also their radiation resistance will be extremely small (note that at high frequencies the current flows on a small shell on the conductor), this will lead to the small efficiency which I, Sherb and you were talking about :)
I do not remember exactly what the ohmic losses resistance equation was but I think it was something like
Rohmic = Rs/(2 Pi a) * 1/I(0)^2 * Integration(I(z)^2,-L/2,L/2)
where a is the radius of the wire and L is its length
Best REgards,
Adel
Hi again and thanks for the replies - they have all donated to my knowledge...:D
I would like to concentrate now on the accuracy of small antenna calculations in CST MWS.
For example, I'd like to take a λ/20 (or smaller) dipole. I would expect a Directivity of ~ 1.7 dBi, poor radiation efficiency and poor ("flat") matching which will lead to a very low gain.
As for now I would be happy to receive your comments on the right setting for small antennas - Boundary conditions, meshing, port size etc.
Thanks again,
P.