Uniform Illumination of Microwave

I have a material testing application where I need to ensure the microwave energy is uniformly distributed over my test material (in the plane transverse to the direction of wave propagation). That is, if my material is 5cm by 5cm (ignore the thickness) and placed 10cm for from the antenna for example, I want to have constant E field distribution over the material CSA.
Does that mean?
1. I have to use an antenna (i.e horn) with very poor gain?
2. Ensure the sample is in the far field
Information would be greatly appreciated. the literature is rather confusing as some suggest wider horn aperture would ensure uniform illumination, but that would suggest greater antenna gain which seems wrong to me.

If you are using a single antenna, a very low gain one is needed.

If that is not good enough maybe a big aperture horn with some sort of plastic lens at the end to correct for the illumination falling off from boresight.

How uniform should it be and what is the allowable phase deviation across your DUT?

For reducing the influence of reflections, you want your DUT as close as possible to the transmitting antenna. So the actual distance is a trade-off between non-uniformity because of reflections and non-uniformity because of near/far field issues.

For using far field distances, you should treat your DUT as an antenna with its own far field distance also. So if you are testing at 20 GHz, your 5cm*5cm square has a far field distance of 0.33m. You have to add this to the far field distance of the horn you are using. Because of this distance, I would not use a low gain horn, but something with same far field distance as the device under test. The directivity of the horn reduces the effect of reflections.

Using the 2b^2/lambda formula assumes a deviation from plane wave of 22.5 degrees.

To make sure everything is OK I would check for uniformity by using a dipole (or other non-directional antenna) and scan the measurement volume. I would not use a horn as receiver for testing uniformity as this may mask reflections arriving from back and side.

It is very well possible that your illumination will be reasonable uniform for shorter distance (then based on the far field distances). This depends on the antenna you use.

WimRFP said:

How uniform should it be and what is the allowable phase deviation across your DUT?

Click to expand...

Well I was thinking more in terms of magnitude of the incident wave on the DUT as opposed to the phase variation. In my mind I am working in the far field where the phase variation is not a problem. What if distance from the antenna is not an issue i.e. I can somehow eliminate reflections by working in an anechoic chamber does that mean the only source of non-uniformity can be eliminated by simply placing the DUT at distance >> 2b^2/lambda from the antenna?
Related to that, if I can use a TEM antenna (yes they exist in some papers!) then I don't even have to worry about the distance from the antenna?

---------- Post added at 17:20 ---------- Previous post was at 17:14 ----------

biff44 said:

If you are using a single antenna, a very low gain one is needed.

If that is not good enough maybe a big aperture horn with some sort of plastic lens at the end to correct for the illumination falling off from boresight.

Click to expand...

You mean like a focussing lens? But then that means I have to place the material under test at the focus of the beam don't I? How does using multiple antennas make it illumination any more uniform? I thought that may cause standing waves and hence hot spots in the material if the antennas are on simultaneously.

If reflections are no issue, you can use the sum of the far fields for the antenna and the DUT, this will provide reasonable uniformity.

This does not automatically mean that when you are within the far field distance, the field is non-uniform. A horn illuminating a specially designed reflector can produce a uniform field within the far field distance for the horn-reflector combination. This is used in compact antenna ranges.

No matter what solution you are going to use, testing for uniformity is the safest way.

---------- Post added at 17:31 ---------- Previous post was at 16:41 ----------

The reason that the far field distance for the DUT must be taken into account, you can imagine from the example below:

When you have a horn with half power angle of 20 degrees (about 21 dBi) at 20 GHz, that would have a mouth of about 6.5*6.5 cm^2. The far field distance for such a horn would be around 0.56m.

Now you will illuminate a DUT with size 0.5*0.5 m^2. Will the illumination be uniform at 0.56m distance (from the horn's feedpoint)?

No, the half power beam width (in m) will be about 0.2m at 0.56m from the antenna's feed point, so you will not illuminate the large DUT uniformly.