antenna-theory.com

[ecdowney]

Hello. Regarding the LPDA by Mr. Enrique Ayala, what is the reason the two booms are angled at 7.5 degrees? Could they just as well be parallel and, if so, what would be the effect of different spacing?

Thanks, Elwood, WB0OEW

[Schubert]

It's all optimization. Decreasing that angle would likely affect the bandwidth somewhat, but would probably still be ok.

[KY5G]

I have precisely the same question, and "optimization" is simply not a sufficient answer.

Is it to vary the impedance along the transmission lines?

Based on my observations of how the value changes using the calculator and playing around, the separation between the two booms (halves of the element TLs) determines the TL impedance at a given point and wider spacing increases the impedance, so the TL impedance starts low and increases as you move toward the longer elements...

Is this to try and match the TL impedance to the "predicted" impedance of each pair of dipole halves?

This is my SWAG thusfar. I'm not sure if this is what I want to do...

I'm using a 0.98+ tau and the associated optimum sigma for my LPDA that I am cutting to for a UHF antenna... 7.155 foot boom and 24-25 dipoles.

Absent a compelling explanation/argument to play around with a constantly variable TL gap, I will use other means to optimize the structure... like a director out front, or circularizing the forward and aft elements, et al.

Was the angle computed or determined experimentally?

[KY5G]

I'm becoming convinced that the inter-boom half angle is an attempt to create a constantly variable impedance along the boom/TL to try and match EACH dipole impedance more accurately INSTEAD of using a single "average impedance" for the entire structure...

There is a formula at the link below that calculates the "average characteristic impedance of the shortest element N"

Then THAT value is used along with chosen input impedance to calculate what I can only assume is the AVERAGE impedance required along the TL/Boom

https://hamwaves.com/lpda/en/index.html

I suspect that one could manipulate these formulas and the ones here:

https://hamwaves.com/zc.square/en/index.html

...to calculate the TL separation spacing for EACH individual dipole and plot them on a graph...

I have not done this, so I don't know if the result would be linear, BUT, I suspect it will be "linearesque"...

In any event, one could use the statistical formulae (sum of squares? I forget) to reduce the samples to the "best" linear fit for the spacings so that they reduce to two straight lines AND reveal the arc angle described by the boom halves when the impedance match per dipole...

I will build my proto to allow for varying the angle from the parallel to a full arc angle to test experimentally (VNA sweeps, et al).

My design is a lot different from the one on the site, so no real reason to expect the angle to be similar.

f= 470-608 MHz
tau = 0.98(+)
sigma = optimal
about 24 elements and a 7.2 foot boom NOT including an optimization director, et al.

IF one circularized the design, you'd need to recalculate the required spacings individually AND then redo the linear regression calcs to verify/find the boom-half arc angle again.

[Galoxlos]

I think it shouldn't have an effect. Wait for people to know.