On Thu, 2008-05-08 at 19:04 -0700, Loren Moline WA7SKT wrote:
Hello,
Back in the old school yagis were constructed with element spacing of say .15 wavelength and lots of times the elements were all spaced the same distance apart. A length was calculated for the driven element and the reflector was made 5% longer and the first director 5% shorter and if tapered the remaining directors were shortened by 5% progressively.
Now with todays computer generated optimization I notice the first director is very close in front of the driven element and also the spacing of elements gets progressively wider with each director.
My question is this. What effect did it have to move the first director so close to the driven element. Was this done to help the driven element impedance be closer to 50 ohms or did this have more to do with antenna gain?
Thanks!
Loren WA7SKT
If you check the director spacings you will find eventually they reach a spacing of 0.4 wave and stay at that as you go forward.
There have been yagi designs with constant length and spacing directors without taper, and some (like the longer NBS) with tapers in both directions, often a longer director furthest out front. (copied for the CC 3219 and 4215 or whatever it was). What this shows is that there are several ways to accomplish the slow wave structure that cause the focusing and so the gain of the yagi.
The works published in UKW Bericht and VHF Communications by DL6WU in the 80s (as I recall) on optimal yagi design are the first to design a yagi that works well by theoretical means. The characteristics of a DL6WU yagi include: The driven element Z is 50 ohms or 200 ohms with a folded dipole. The first director isn't too close to the driven element. Director spacings taper smoothly from about .08 or .1 wavelength up to 0.4 wavelength then don't change any more. Director lengths taper smoothly from the first to the last director. He bases director lengths on reactance change or phase angle. He emphasizes that wide first director spacing to get the feed impedance up.
K1FO and computer optimized yagis tend to squeeze the first director close to the driven element. Its been noted by K1FO and DJ9BV that such a close spacing tends to knock the impedance of the driven element way down, like 15 or 17 ohms which takes the T match to get up to 200 ohms for the balun. And some have noted that when the driven element is loaded down that far that the measured antenna efficiency never is as good as computed or as measured on a design where the first director spacing is greater. I'm sure the many muses published by K1FO will mention that occasionally.
NEC simulations of yagis don't show gain dependent much on the driven element dimensions. So the driven element details aren't modeled but determined experimentally.
The performance of an original DL6WU yagi beats the NBS yagi by maybe half a db per wavelength of boom. A computer optimized yagi beats a DL6WU by about that much, at some costs. First the driven element impedance tends to go down which can cause that efficiency drop from the close spaced first director, and then the element spacing, boom effect, and lengths become more critical making construction a bit more difficult. The DL6WU yagis are most tolerant of construction details. The most critical of 432 yagis were developed by K2RIW, his 19 element design. He demanded element length precision to +/- 1/64th inch, and when that care is taken that yagi comes up at 15.05 to 15.1 dBd at antenna measuring contests. Its so consistent that its the US standard gain reference for 432. With a pair of them at about 75 feet, I've heard CW signals off the moon at moon rise by ear without a preamp on the radio.
73, Jerry, K0CQ
On Thu, 2008-05-08 at 19:14 -0700, Loren Moline WA7SKT wrote:
One correction..the driven element might be balance fed so substitute
200 ohms for 50 ohms in that case or say 50 ohms with a gamma match.
Loren WA7SKT
Gamma matches above 100 MHz or so have tended to be parts of yagis that missed their gain claims by several dB as if the asymmetry of the gamma match upset coupling from the driven element to the long array of directors. But then Kent uses a J shaped asymmetrical element to save on a balun and makes them work for 8 or 10 element yagis.
73, Jerry, K0CQ
WA7SKT, K0CQ & The Microwave Group de K2RIW 5/09/08
Dear Loren, Jerry and The mw Group,
Loren asked a great Yagi Design question concerning the First Director spacing, Impedance Matching, and Gain.
PAST YAGI DESIGNS – A number of Yagi designs have used a close spacing of the First Director as an Impedance Matching Device for the Driven Element. I've been told this is one of the claims in the patent for the F9FT antenna designs. That was considered to be an effective method. However, the use of modern computer modeling programs (with experimental confirmation) has shown there is a better way.
SEPARATE THE REQUIREMENTS – I say the adjustment of the Parasitic Elements should be done to optimize the Gain, or a particular Sidelobe Pattern, or a particular G/T performance, or a particular Front-to-Back Ratio (the adjustments are all slightly different). The adjustment of the Impedance Match to the Driven Element is a separate and independent function, and that function should not be allowed to compromise the other performances.
WE DID IT THE HARD WAY – In the past, many amateurs were lulled into mis-using the First Director as an Impedance Matcher. They believed achieving a good Impedance Match to the Driven Element would required a considerable amount of mechanical intervention, such as using a particular Delta Match, using a Folded Dipole Feed with different diameters for the Folded Structure, a T Match with a precise setting of the T Length, etc. We now know this is not true.
DELTA-MATCH GAIN LOSS? – By the way, a Delta Match that is directly connected to a coaxial transmission line is not a good Balun, it creates an Imbalance that will allow the transmission line to radiate. This may be a reason such Yagis sometimes do not perform as well as expected.
A BETTER WAY – We recently learned that achieving “a perfect impedance match” to the Driven Element (such as a VSWR of 1.05:1 or better) is rather easy. To summarize, it simply requires three actions:
WHERE IS THE INFORMATION? – You will find a few thousand words that better explains this Yagi (or other Dipole-like) Impedance Matching technique if you go to: www.wa1mba.org
At the bottom of the first page you will find, “Yagi Impedance Matching Info”, click on it. On the next page you will find, “A description of the concepts of Yagi matching”, (as well as the other two Monopole Reactance Charts) click on it.
WHY THE RELUCTANCE – When you follow this procedure, often you will end up with a Driven Element that is slightly longer than the Reflector. This seems to horrify some amateurs. They mistakenly believe a Yagi will only perform correctly if the Director is the longest Element. The Parasitic Elements do all the work – the driven element only has to radiate in a Dipole-like manner, and present a good impedance match to the transmission line.
Jerry, K0CQ has said, “NEC simulations of Yagis don't show gain dependent much on the driven element dimensions. So the driven element details aren't modeled but determined experimentally.”
WHAT IS THE BENEFIT – Achieving a “perfect impedance match” is very desirable, particularly for EME and Satellite operations. In each case it is desirable to acquire the lowest possible Noise Figure. Your LNA will only produce the Noise Figure you saw on the Automatic Noise Figure Indicator if your antenna ALSO looks exactly like 50.0 ohms resistive.
1/64 INCH YAGI REQUIREMENT? – This may be excessive. My good friend George, W2KRM was the manufacturer of the “RIW Products” 19 Element Yagi kits. George wasn't sure of the sensitivity of the element lengths, therefore he (or someone else) may have stated this requirement. I believe that a (+)(-) 1/16 inch accuracy is adequate for Element Lengths, as long as the error is not additive. The Element Spacings are even more forgiving – if it is not additive. However, the 3/16“ Element Diameter is critical, and mounting the Elements with Insulated Shoulder Bushings is critical. I recommend an approximate 1/64” chamfering of the ends of the Elements. A lack of chamfering will change the apparent electrical length of the Elements, and that deficiency would be accumulative.
73 es Good UHF/SHF/EHF/EME/Satellite DX,
Dick, K2RIW