Absorber Question

[lighterup]

I am having trouble with what I believe to be parasitic scatterers in the support structure of a circular antenna array. My proposed solution is to install a layer of tuned elastomeric absorber, such as Cuming C-RAM FLX inside the radome below the array to keep the transmitter from lighting up anything below the array.
My question is this:
I have been bitten before by metallic edges in the field causing reflections, and I am a bit leary of running the metalic backing on my absorber sheets all the way to the edge of the sheet. My system is an L-Band radar running at 1.3 GHz, and I am thinking of terminating the metal backing about 1/2" to 1" from the edge of the absorber sheets closest to the radiating antenna structure. Am I just being paranoid? Anybody had any experience with this stuff?

 

[VEBill]

I can't answer your question, but I've seen a totally different solution to a similar problem.

Sometimes the diffraction scattered from the edge of a parabolic dish can be a problem. The example that I read about was that some C-band downlink receivers were bothered by ground-based point-to-point links on the same band. They needed to improve the sidelobe performance of their satellite dish.

The solution was to change the shape of the edge of the dish from a simple straight edge to a scalloped form (a series of raised humps that have the shape of a rectified sine wave). The size of the edge humps is a function of frequency. Basically, edge diffraction can be reduced by getting away from straight edges.

 

[lighterup]

Thanks, VE1. My transmit antennas consist of a collinear vertical array of six dipoles. The elevation pattern looks good when the array is measured without the support structure, but evidently it develops some lower lobes when placed on the support structure, as we are seeing severe clutter problems when the system is operated higher off the ground than about 50'. All I am trying to do is prevent the illumination of the support structure by absorbing the energy which is directed downward.

 

[VEBill]

Is it a 'circular' or 'collinear' array? Your two posts differ.

Collinear means they're are all lined-up, typically within a single tube.

If it is collinear, you may need to decouple the feedline to achieve the cleanest pattern.

If it is vertical collinear, then there shouldn't be much power directed downward. The relative spacing of the antennas should result in a null down (and up).

 

[lighterup]

A more appropriate description would have been a 360 degree cylindrical array of collinear arrays. You are correct, in that the collinears do not show much energy at a negative look angle when measured in the chamber, but when integrated with the entire mechanical structure, there seems to be enough energy being directed downward to light up every water tower and metal roof within about 3 kliks.

 

[Higgler]

Lighterup,
A dipole has a null downward and upward for V pole in your array. If you put a dipole in a box, it has peak gains off the edges, exactly opposite free space.
Tuned absorber may help, but if the currents are generated on your back plane and run onto the support structure, the resonant absorber won't help since it expects a "far field planewave source", not a surface current running under it.

Do you have a photo of the setup. 1.3 Ghz is 9" and quarter wave chokes are small, I'd try that first for simplicity. Absorber can be a pain.

I would think you have about 8 elements in front of a 12" diameter ground plane and an 8" support mast. You can add chokes between your antenna and the support mast. 1/4 wave is 2.25 inches without dielectric filling and you may only need a few of them to cut your currents down. you could also add chokes to your mast, which may be easier. Let us know if you are familiar with 1/4 wave chokes.

kchiggins

http://www.toyon.com

 

[VEBill]

There's a bit of a discepancy in that if the antenna doesn't put much power downward, and the metal structure is at that location where there isn't much power density, then it shouldn't affect the pattern much. It should lack the necessary power to do anything.

In other words, putting a mirror in a dark corner shouldn't have any affect on the pattern of the bright light.

Thus, I'd look for a reason why the RF is getting onto the not-directly-illuminated structure.

The culprit might be currents on the outside of the feedlines. If so, then at HF you'd simply wind a common mode coil in the coax. For 1.3GHz, Higgler's suggestion of 1/4 wave chokes is probably best. Or maybe some suitable ferrite cores.

 

[lighterup]

Here is a link to a photo of a similar system.

http://www.defenselink.mil/transformation/factsheets/LCMR.pdf

I do not believe the feedlines are radiating significantly, as the input is only about 10dBm. The coax lines to the element columns terminate into microstrip, and the onboard PAs see less than 1.2:1 VSWR across the band of operation.

 

[Higgler]

Is the support structure just a 3 legged tripod with a small flat mounting place on the top as shown in the photo? Or is it another type of mast for mounting?

When testing with and without the mount; was the comparison full array without and full array with mount as shown in photo. How bad is the pattern degradation?

also, is the backplane behind the array reflective metal or lossy absorber?

kch

 

[Higgler]

better photo, page down near the bottom,

http://www.syrres.com/pdfs/FAJournal_article-afghan.pdf

kch
Looks like you spaced the antennas 0.7 Lambda. if you are scanning upward, your lower two elements may form their own baby grating lobe downward (which is a cone extending all the way back to your mount) and give you more energy than you'd expect towards your mounting legs. In a normal test without the base mount, the RF would go behind you, but adding the legs would reflect it back out forward. The ground will make it worse with the corner reflection bounce.

Is this a big problem you're dealing with, or a little annoyance?

Kch

 

[lighterup]

The scan is azimuthal. The elevation receive beamformer is a Blass coupler structure, and there is a known lobe in the receive pattern about -15 degrees elevation. My hope is, if we can eliminate some of the transmitted energy at negative look angles with absorber, that energy will not be picked up in the receive lobe. The mount for the new configuration is a tower. The tripod is mounted on a small platform atop the tower. The tower has no guy wires. A fifty foot tower gives marginal clutter performance. An eighty foot tower gives poor clutter performance. I can't discuss specific vulnerabilities, but I would classify it as no small problem at =>50' system height.

 

[Higgler]

Known lobe means grating lobe at -15 degrees (below the horizon) for one particular elevation scan angle?

I see from curves that 15 degree scanning with about 0.78 Lambda element spacing brings the grating lobe into real space in the downward direction. Hence, you've got alot of grating lobes in your patterns if your array scan's upward more than 15 degrees, which is what I think it does.

To prevent energy from hitting your mount, you may want to simply extend your groundplane behind the lowest element downward by 6 inches or more, it could be just some copper tape with aluminum foil to start. (this assumes you have a groundplane behind the antenna). Assuming it's the bottom two elements shining energy backwards, that may help control the bounce off your mounting mast.

As for the grating lobes, your elements could use antenna pattern controls added to them to minimize grating lobes downward. First things first though.

kchiggins

http://www.toyon.com

 

[lighterup]

Hmm...I think I can do the ground plane extension without too much form factor upset. Good idea.
Thanks
lighterup

 

[VEBill]

There's something funny about there being a difference in the pattern performance of a system mounted on an 80-foot tower versus a 50-foot tower (at 1.3 GHz!). Presumably the first (top) 50-feet of tower are pretty much the same in either case...

My gut tells me that there's something else going on other than pattern changes.

 

[lighterup]

Visualize this, ve1bll:
There is an large warehouse within 3 km of the site with about an acre of tin roof with a slight pitch which has a pretty good cross section as the height increases and the surface begins to become more normal to any lobes which may be incident on it.

 

[VEBill]

The more perfect the plane reflector, the less likely it is to be properly aligned. Also, an acre looks pretty small from 3km away. 3-corner reflectors, even unintended ones, are much larger.

A quick test might be to simply tilt your radar system slightly up in that direction. See if such a quick test validates your theory abotu the root cause of the problem.

 

[Higgler]

lighterup,
if you have a single direction problem, maybe you can add a simple antenna on the bottom of you unit and control it's amplitude and phase on the receive only portion and knock down some signal return. Starts to complicate your simple system though.

If it were easy to find solutions, we'd be out of jobs.

kch

 

[lighterup]

All suggestions so far have been valid, and I am grateful for your input. However, my original query was about a simple absorber inverted umbrella which could be fitted inside the rigid radome of the system. Has anyone had any experience with this stuff as a sidelobe suppression mechanism? Bear in mind that I am not considering the classic Salisbury screen, but the loaded, tuned elastomeric absorber material of the type I quoted originally, which has a much more forgiving angle of incidence and is not constrained to be used in the far field. I believe this is what the parabolic dish folks do for suppression.