Use hybrids, rat races are narrow band.
Start with mincircuits or use werlatone if you have the money.

kch

Thanks Higgler, Will do..

Frank

I just noticed the patch antenna reference.

Patches aren't 2:1 bandwidth (not even if you say please), they're typically 1% to 20%.

Do you really mean patch antennas?
If yes, then you're probably operating within the 200-400 MHz band and only need a small percentage (like 390 MHz +/- 10 MHz), or you thought patches were wideband.


kch

I'm trying to stretch the desing to the limit 20%BW. My freqs are inside the 200-400 Mhz band. Any suggestion to stretching the limit on patch BW??

again thanks,

f

200 to 400 MHz? That frequency range overlaps more than one application. 225 to 400 MHz is Military UHF, but 200 to 225 normally isn't.

If you're forced to retreat to a narrower bandwidth, then you could use cable lengths to set the relative phases. For example, when making circular polarization with crossed dipoles, it's common to use cable length to set the 90-degree phase delta.

Is this an experiment or a high volume production job?

Tough to do at your frequencies, you'd need a patch at 400 MHz that's 6" tall and 14" long. Twice that at 200 MHz. Do you have the room? Your array will be 30" square x 6" tall for a four element array at 400 Mhz and 60" square and one foot tall at 200 Mhz, roughly +/- 15%.

We did some thick and double tuned patch designs last year. See Pozars book on thick patches. He says 20% vswr bandwidth with a thick patch.

kch

Depending on your application, you might be interested in this:

The ViaSat QDC-100 combines four antennas for enhanced reception of UHF MilSatCom around 300MHz. It also controls a 1-of-4 switch for the transmit path. There's info on the 'net.