log-periodic toothed antenna

[neval]

Hello everyone!
Made this antenna in HFSS. Need s11 to be under 10db at 450-970 MHz and 1800-1900 MHz.
Anyone have idea how to achieve it?

Thanks!

 

[VEBill]

In general, the image shows an antenna that almost seems too simple. Not enough elements. Is there a way to increase the density of the elements in the design so that there are additional elements to accept power at the middle of the lower band?

You need an element or two with a length matching quarter wavelength of about 700 MHz.

 

[neval]

This is "original" antenna.
I removed some elements to eliminate 1-1.8 GHz bandwidth.
Was that ok?

 

[VEBill]

There is a parameter 'Number of elements' (example: 'Nelem' in the link provided).

http://www.ewh.ieee.org/soc/es/Nov1998/13/LPARRAY/LPDA.HTM

If the chosen number of elements is too low, then you'd expect to see holes in the bandwidth; especially at the low end where the percentage bandwidth is worst. For example, if somebody designed a log periodic antenna with only two or three groups of elements (an extreme example to make the point clear), then you wouldn't expect it to work very well. As a general rule, using a larger number of elements should assist in smoothing things out over the bandwidth.

If you're using a log periodic antenna, then you shouldn't bother removing elements to eliminate bandwidth.

You might also be able to tweak the element lengths to move things around a bit. If there's a reflection at 700 MHz, then adjust things so that there's a pair of elements accepting power at that frequency.

Thing is, you're trying to achieve a bit higher than 4:1 bandwidth (450 - 1900 MHz). That might be a stretch. If you can't make it work, then you might have to find a parallel arrangement of two antennas, each covering their own bandwidth.

Disclaimer: I've never designed a log periodic antenna, so this is all based on other somewhat related experience.