Yop all,

in order to realize some RF measurments, i need to find an antenna with a very high cross-polar isolation. I suppose the best choice is to take a logarithmic antenna. Do someone know a very good antenna or a reseller where i can find one?

Thanks a lot

Hi
If you are interested in frequencies high enough for a horn antenna, there is a design published for an octagonal horn with better than -40dB CP level on bore-sight and around 35dB off bore-sight.
It uses a short rectangular to square section that excites a higher mode, then a transition into the flared section where the square corners are chopped off to form an octagon. I have built one for 13GHz that performed very well over a band of about 1GHz.

At VHF thru L Band, log-periodics can achieve >30dB CPI on bore-sight, but as so often happens, the "rabbit ears" off bore-sight (around where the first side-lobes on the copolar pattern would appear) reduce the CPI to around 20dB.
So, what part of the spectrum are you interested in?

Ok, so i work in fact in an industrial environment. In our project we use a satellite link and and some wireless local network such as WiFi for example.
The idea is to be able to analyze the signal of the WiFi (check on the power level, the coverage...) network and the satellite one (if the link is well up, analyze power level, norms compliance etc). So it would be great if the antenna is able to "see" signal from 2,4 GHz (the lower frequency for WiFi) up to 16 GHz (the higher frequency for the satellite link).

Why do i need an antenna with a so high CP isolation? That's a requirement to be compliant to the EN50383 (Europe) norm: all the measurement must be taken with an antenna with a CP isolation of, at least, 30 dB.

Hi Oli
OK, for the Ku Band 12 to 16GHz, I can definitely recommend the octagonal horn patented in 1978 by UK ERA, Dr.Alan Rudge. The details are in the UK Patent (now expired) 26226/78, and an article in the UK IEE (now IET) conference publication 169.
I have dimensions for 13GHz, and I made a model for a test range (for study purposes only of course) and it worked very well. Dr Rudge claimed to achieve -50dB cross-polarisation over part of the band.

The key to obtaining low CP levels is to ensure (somehow) that the E and H plane radiation patterns are as nearly identical as possible. This is possible to achieve with a horn because the currents are constrained by the walls.

At 1.8-2GHz it might be possible to construct a similar horn scaled in frequency, but the dimensions would be large. It is rare to find a design, for example a patch antenna, that has a CP level any better than about -15dB relative to the co-polar beam on boresight. It might be better to contact a company like www.tdkrfsolutions.com
Unfortunately their data sheets on the web do not quote the CP performance, usually a warning sign they are not good. If something is good, marketeers are never slow to blow the trumpet.

Thank you for the information. I will try to contact them and know more about the antenna.