How to design a matcher-diplexer?

[hfmobile]

I need to use a biconical antenna for a broad frequency frequency range (ratio 1:15). I recorded the antennas impedance chart: for the high frequencies it is something next to 50Ohm but for the low frequencies it is something far off and very capacitive. The antenna is just too small. But I have to use this antenna. I tried to match it for the whole frequency range into VSWR better than 2:1, I even used some resistors. I know, this is not a nice idea. But matching such an impedance is like cranking a bathtub: the water flows out at the one or the other end. An idea is to create a kind of matching diplexer which matches different in two different frequency ranges. Is there some literature how to do this?

 

[Higgler]

I'm curious what the size and frequency range of your antenna is. I have a few hints for improving the low end frequency VSWR.
I recently made a 3:1 VSWR Bicone from 1 to 40 Ghz, at 3" tall and 5" diameter.

Is this antenna one that you designed or was it forced upon you?
It's difficult to make the antenna more "efficient" when it's electrically small. I've seen impedance transformers used for electrically small antennas, using actual transformers for a low frequency antenna (100-500 Mhz).
kch.

 

[VEBill]

Higgler wrote, "It's difficult to make the antenna more 'efficient' when it's electrically small."

The original poster's screen name is 'hfmobile'. If that is any indication, perhaps he already has experience with electrially small antennas. ;-)

 

[hfmobile]

I am forced to use this antenna which is a classic biconical shortwave vertical monopole antenna, originally specified to use for for 3.5-30MHz. I want (I should write "I am forced") to match it in the range of 2.0-30MHz or at least 2.0-28MHz by a broadband matching network. Usage of a classic narrowband (single frequency) antenna tuner is not appropriate because these motor driven or relay-switched tuners tuners are too slow or at least much slower than the transmitter.
I tried to match the antenna but when I was able to match the antenna on one end the VSWR always raises higher than the maximum acceptable value on the other end of the frequency range (see my first mail with the metaphor of the bathtub). This was the reason to ask for experience desingning diplexers or a matching circuit which behaves VERY different in two frequency ranges.
All in all the antennas impedance is somehow acceptable in the range of 3.5MHz and up. So a matching circuit which matches the antenna below 3.5MHz but does not affect the VSWR above that frequency is my aim. Maybe I have to add a resistor to add some real component to the antennas impedance but of course I would like to prevent this way...

 

[VEBill]

One could state that you're looking for an antenna matching network (an 'antenna tuner') that will work over a fairly wide frequency range, 2-30 MHz, without any requirement for adjustment (no knobs, or at least 'set and forget'). That would be nice if it existed (I'll buy one).

You could scale the 3.5-30 MHz antenna to cover 2.0 to 17 MHz. Perhaps then add a fast switch to automatically choose.

Another approach that can be used is to insert an attenuator. 3dB one-way loss (not too bad) gives 6dB return loss (pretty good match).

 

[Higgler]

The technique I saw was used on a 100-500 Mhz electrically short monopole. A simple transformer was used to transform the high impedance of the electrically short monopole to improve the antenna match to the 50 ohm cable. Short antennas have high impedance.

Is the diplexer needed to split the antenna to two receiver or transmitter ports? If not, you could use two separate matching circuits, (low and high band) using back to back identical diplexers and matching circuits in between the diplexer ports. Only around the diplexer split point would the match go wacky.

kch

 

[VEBill]

100 to 500 MHz is a 5:1 ratio.
2 to 30 is a 15:1 ratio.

 

[hfmobile]

Short antennas have a hígh impedance but if I have compensated all reactive parts of this impdance, the real part which is left is very small. This contains some resistive losses and the rdiation resistance. The latter is small because the antenna is short.
I developed some LC-networks for broadband antennas for smaller frequency ranges (up to 1:5 ratio) and sometimes for resistive loaded or terminated antennas for an even higher frequency ratio. These networks did not need further tuning but these antennas have not been so difficult, they were "calmer".
With this antenna I have the problem that the thing I have done good on the one end (of the frequency range) is made worse by the things I do on the other end. This was the reason that I asked for some kind of diplexer, a circuit or a way of solution (a helping software or literature?) to dimension two frquency ranges concerning the same antenna...
We work with broadband transmission line transformers for high power, add their S-parameters in a simulation software for exact calculation.
Generally this antenna is for a medium power application. Our solution so far was a resistive "calmed" LC-network. But I dislike the big resistor there which needs oil cooling in the lower frequency range. The efficiency and the effort are not in a good ratio.
Adding just an attenuator lowers the efficiency for the complete frequency range. Of course we had this idea too ;-)
The switching idea is a good idea. When the transmitter is frequency-hopping the two matching networks need to match the antenna with an overlapping frquency range which has to be far wider than the hopping range because the switch will be too slow for hopping (I suppose that).

 

[Scott231us]

Did you ever figure this one out? I'm a amateur, and I'm trying to design a broadband RF transformer for for the 2mhz-60mhz range.. Any ideas? I am a beginner, so please pardon the easy questions.