On 11/18/21 12:58 PM, Tim Duffy wrote:
There is a reason DX Engineering/Comtek does not sell Comtek 4 square boxes
above 14 MHz. The inductor and capacitor values are very low and losses can be
a real problem. We are still trying - but it is not as easy as dividing down
the values. Believe me - we tried.
73
Tim K3LR @ DX Engineering
At some point, some other quad hybrid approach becomes more practical.
And harder as percentage BW increases. I don't know that I would want
to build a Lange Coupler for 10 meters (although you might be able to
build on on flex substrate and coil it up?
But there are other techniques that might lend themselves to coax. A
branch line coupler is made of 4 transmission lines, with impedance
0.707*Z0. If your Z0 is 71 ohms (maybe 75 ohms is close enough?) then
the lines are 50 ohm lines that are 1/4 wavelength long.
Or a rat-race, with lines that are 1.414 * Z0 - again, 50 ohm Z0 would
use 71 ohm lines.
https://www.microwaves101.com/encyclopedias/hybrid-couplers
and various pages linked from there. All kinds of cool ideas.
The nice thing is that these are modelable in NEC (unlike the Collin
hybrid), since they're just transmission lines. (Yes, you can model the
Collin as an idealized 90 split, but you can't model the coupled
transformers and capacitors)
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|