PDA

View Full Version : Comparing Two Big 2m Yagi Designs



K0FK
06-19-2008, 04:20 AM
Are there any rules-of-thumb for the relative efficiency of two comparable-gain beam antennas based only on the matching devices they use? Here are the two specific antennas in question.

The first is an 11-element design by G4CQM (http://myweb.tiscali.co.uk/g4cqm/g4cqm/netlon/test.htm) using what he calls a direct feed driven element, which just seems cool.

I compared his 11 element design with a Hy-Gain 13 element antenna, because the two beams have about the same gain claims (15.3 dBi for the Hy-Gain and 15.7 dBi for the G4CQM design). THAT led me to look closer at the physical designs, with my attention drawn to the beta match used on the Hy-Gain--is there anything that can be said about the efficiency of the beta match in general, or is it totally specific to the particular, complete antenna design?

Has anyone here built a beam using the G4CQM-style driven element?

WB2WIK
06-19-2008, 05:02 PM
Are there any rules-of-thumb for the relative efficiency of two comparable-gain beam antennas based only on the matching devices they use? Here are the two specific antennas in question.

The first is an 11-element design by G4CQM (http://myweb.tiscali.co.uk/g4cqm/g4cqm/netlon/test.htm) using what he calls a direct feed driven element, which just seems cool.

I compared his 11 element design with a Hy-Gain 13 element antenna, because the two beams have about the same gain claims (15.3 dBi for the Hy-Gain and 15.7 dBi for the G4CQM design). THAT led me to look closer at the physical designs, with my attention drawn to the beta match used on the Hy-Gain--is there anything that can be said about the efficiency of the beta match in general, or is it totally specific to the particular, complete antenna design?

Has anyone here built a beam using the G4CQM-style driven element?

::G4CQM and some others live in a world of delusion regarding "directly fed" driven elements. You're still feeding a balanced load with an unbalanced line, and if you optimize all dimensions to create a perfect, non-reactive 50 Ohm load from a split dipole, you've now officially screwed up gain optimization.

"Look at what the pros are using" is generally good advice. The "pros" in this case, who are top-of-the-heap VHF weak signal operators who win all the contests, work all the meteor scatter and moonbounce, and have 2m WAS, 2m 500-grid VUCCs, and dozens of countries confirmed on 2m, use "the good stuff," by and large. Here in the States, the majority are using (for 2m) yagis by M2 (http://www.m2inc.com) which are computer optimized long boom yagis having a T-match and coaxial balun at the feedpoint. This system is lossless, except for the loss of 26-1/2" of high-grade RG6/U, which at 144 MHz measures 0.05 dB, or 1/20th of a dB. You can't measure that, it's too low -- so you can only calculate it.

Getting a directly-fed dipole to properly balance on 2m is a difficult task. I wouldn't bother trying. None of the "pros" bother trying. The K1FO, K2RIW, K6MYC, et al. designs all use a DE matching system to optimize performance and maintain pattern cleanliness.

I like the T-match & balun system used by M2 (which was also used for decades by Telrex, the previous highest-performance VHF antenna manufacturer): It's low cost, simple, foolproof, easy to adjust and has no loss at all, while providing a method of matching a driven element that is part of a highest possible gain assembly.

WB2WIK/6

K0FK
06-20-2008, 04:39 AM
Your first couple of comments make sense. In spite of not knowing why, intuitively at least, I think I'm tracking you. I wish I was sharp enough that seeing coax go into a dipole would have at least made me scratch my head in wonder...of something. It didn't even raise my eyebrows--gotta a LONG way to go in understanding this stuff.

Of course, the guy (G4CQM) has been designing antennas for I think a decade or two, and I've been 'looking' at them for a few months, so naturally I assume there's a lot to learn there.

WB2WIK
06-20-2008, 10:43 PM
The problem is, unless you can perfectly model the design (which at this high a frequency becomes problematic), the only way to know what you've done is to measure it. And the only way to do that is on a large antenna range that's reasonably anechoic for the frequency being used, which very, very few people have.

I can build a reasonable antenna range for use at 13cm, but not for 2m -- everything's too big.

However, what G4CQM describes has been tried many times, and measured. It doesn't work. The problem is, to balance the DE is difficult with direct attachment because the distance from feed attachment to transmission line split must be absolutely identical for both conductors of the line, and this alone is hard to accomplish with what he's shown. It's much easier to accomplish with a coaxial receptacle in a machined housing, using machined parts of tightly controlled dimensions such as M2 does it.

Then, if there's no matching device used, there's no way to adjust the impedance of the DE other than to do that by adjusting element lengths and parasitic element spacing. When you do that, to match impedance to a 50 Ohm line (e.g., creating a perfect 50 Ohm load by mechanical tweaking), you have to space the reflector and driven elements in non-optimum positions. That upsets the pattern and detracts from maximum gain, compared with spacing those elements for optimum gain and then adjusting the DE for a match, using a tunable matching device.

This is the reason people use tunable matching devices! It's not just to make antennas more complicated, it's to make them work better.

The Yagi is a 60+ year old design and it's been tweaked to death by hundreds of qualified engineers over several decades. There's nothing left to do with it. Making it "simpler" by directly feeding the driven element with coax is not any kind of improvement.

WB2WIK/6

W8JI
06-20-2008, 11:42 PM
Your first couple of comments make sense. In spite of not knowing why, intuitively at least, I think I'm tracking you. I wish I was sharp enough that seeing coax go into a dipole would have at least made me scratch my head in wonder...of something. It didn't even raise my eyebrows--gotta a LONG way to go in understanding this stuff.

Of course, the guy (G4CQM) has been designing antennas for I think a decade or two, and I've been 'looking' at them for a few months, so naturally I assume there's a lot to learn there.


It is just as easy to make a dipole balance on two meters as it is any of the T matches and other things. There is no difference at all. There are a lot of very simple baluns that work quite well. The most simple is to just use 1/4 wave of coax from the feedpoint and ground it to the boom 1/4 wave (about 18-19 inches) away from the feedpoint. I used that in commercial antennas and the measured balance was very good.

G4CQM increases the feedpoint resistance to 50 ohms by adding an additional element spaced very close to the driven element. He calls this element a "director", but it is really functionally NOT a director. It simply serves to increase the DE impedance to 50 ohms.

I see nothing at all wrong with his design, it looks fine to me. It is simple and it is a standard way to do Yagi antennas. Look up the OWA antenna and you will see what I mean. His is the same basic principle as the OWA. He just has a little wider spacing of the mutually coupled second driven element.


73 Tom

K9KJM
06-21-2008, 08:53 AM
My .02 on VHF yagi type antennas. IF you want to build your own antenna, Make a "Quagi" A really easy to build and match homebrew that works every time. http://commfaculty.fullerton.edu/woverbeck/quagi.htm

If you want to just buy an antenna, Get the largest M2 your tower/support/budget will allow.

And be sure to use very low loss coax feedline. Times LMR-400 for runs up to about 70 or so feet, Anything longer, Get some good used large size Heliax.

G4CQM
02-15-2010, 01:51 AM
WB2WIK please take time to read...My yagi website has undergone many changes and NEW antenna designs added since your comments back in 2008. 'Direct Feed' driven elements are here to stay, they are a reality and DO WORK, in fact incredibly well! I've been using them since late 1994 and they have been used in commercial yagis for the professional market here in the UK from 1999 onward. I know because I designed them!

My website now explains in full how my 'Coaxial Dipole' works and the Vector Network Analyzer plot included demonstrates just how remarkable they are.

The JN53 Contest Team account is a completely independent account of just how effective 'Direct Feed' is and details their move away from unreliable matching devices.

URL: http://myweb.tiscali.co.uk/g4cqm/

73's from

Derek Hilleard G4CQM

EI2GLB
02-15-2010, 02:10 AM
M2 are too expensive over here in EU. I can homebrew 5 proper antennas for the price of one middle of the Road M2.

I've never heard of G4CQM but I am sure his designs are very good.

DK7ZB and G0KSC are 2 others to have a look at along with YU7EF

WB2WIK
02-15-2010, 02:56 AM
WB2WIK please take time to read...My yagi website has undergone many changes and NEW antenna designs added since your comments back in 2008.


Very nice! If I had time, I'd replicate and test but this is more time consuming than I have.

However if you can provide a sample antenna of any length I'd be happy to test it on a commercial range against an identical boom length VHF yagi made using a T-match and balun, or a Quagi. Easy to accomplish, since I already have the range and it would be interesting to see how it fares.

Tonna F9FT made "directly driven" VHF yagis (with a very fat driven element also) that worked fairly well (I had several back in the 1980s when they were distributed in the U.S.; unfortunately they no longer are) but never could compete with the K6MYC, K1FO and K2RIW designs at the calibrated range antenna competitions held by CSVHF and other organizations; so they lost their following and also their distributorship and now are not well known. I thought that was a pity, as I would certainly sacrifice 1 dB gain for the ease of assembly they offered (square booms, wing nut assembly, requiring no tools in the field).

Can we test one? I'd be happy to document/photograph/have independent witness the results.

73

Steve WB2WIK/6

WB2WIK
02-15-2010, 02:58 AM
M2 are too expensive over here in EU. I can homebrew 5 proper antennas for the price of one middle of the Road M2.



They're pretty cheap here in the States; the difference must be transport costs and perhaps import duty, etc.

Here they cost about as much as it costs to homebrew a yagi using good materials.

KC8VWM
02-15-2010, 03:25 AM
They're pretty cheap here in the States; the difference must be transport costs and perhaps import duty, etc.

Here they (M2 antenna) cost about as much as it costs to homebrew a yagi using good materials.

I strongly disagree.

I just constructed a DK7ZB 12el longyagi on a 8.00m (http://www.qsl.net/dk7zb/2m-longyagi/12-ele.htm)(26ft) on a square boom for less than $150.00

Material costs breakdown:


$15.00 per 96" square boom x 4 = $60.00

Stainless hardware (nuts bolts washers) $25.00

10 ft - 10mm rod x 4 @$10.00 ea = $40.00

A piece of 12mm rod for constructing the DE. $$Free. Had a 3 ft piece in stock otherwise we are looking at 3-5 dollars if purchased.

75 ohm cable for constructing the balun = CATV wire - free.

Female N - connector = $4.00

A bag of 25 electric fence insulators converted to "element standoffs" =$4.99 (Available at farm supply)

1/2 inch electrical conduit box for mounting DE / balun cable, N connector assembly etc = $4.00


Total cost of homebrew construction = $137.99


.... So.. Where can I get an M2 antenna of comparable boom size, number of elements and similar gain parameters for that cost?

Just sayin :)

WB2WIK
02-15-2010, 03:31 AM
.... So.. Where can I get an M2 antenna of comparable boom size, number of elements and similar gain parameters for that cost?


Have you measured the gain? If so, where'd you do that?:)

KC8VWM
02-15-2010, 03:48 AM
Have you measured the gain? If so, where'd you do that?:)

I'm not comparing "gain" per se,

What I am comparing are the costs of purchasing similar materials of similar quality (6061-T6 etc), having similar boom dimensions (3.83λ) , and/or having a similar number of elements (aluminum rods) in the commercial design construction and identify their respective comparative costs.

You can compare gain parameters of various antenna's using this chart if you like:

http://www.dk7zb.org/ve7bqh.htm

WB2WIK
02-15-2010, 04:13 AM
I'm not comparing "gain" per se, I am comparing similar "stated gain" parameters specified according to EZNEC to keep everything compared on a level playing field.

For example: http://www.dk7zb.org/ve7bqh.htm

I am comparing the costs of purchasing similar materials of similar quality (6061-T6 etc), having similar boom dimensions (3.83λ) , and/or having a similar number of elements (aluminum rods) in the commercial design construction and their respective comparative costs.

I understand; but it's not a realistic comparison unless you have a way to measure the results (gain). I can build an antenna that "models" at 20 dBi on 2m for $50. But it won't measure that. The M2 antennas have consistently measured their stated gain at numerous commercial ranges, including CSVHF, NEWS, SBMWS, etc. where provisions are made to actually measure real gain within 0.1 dB or so. If we are comparing "apples to apples," that has to include real world gain measurements.

WB2WIK
02-15-2010, 04:14 AM
I understand; but it's not a realistic comparison unless you have a way to measure the results (gain). I can build an antenna that "models" at 20 dBi on 2m for $50. But it won't measure that. The M2 antennas have consistently measured their stated gain at numerous commercial ranges, including CSVHF, NEWS, SBMWS, etc. where provisions are made to actually measure real gain within 0.1 dB or so. If we are comparing "apples to apples," that has to include real world gain measurements.

I should append this to state that real world gain at VHF and UHF are often quite different from the models due to losses incurred in practically "everything." At HF, it's a lot less critical and the 4nec2 or EZNEC predicted gains are much closer.

EI2GLB
02-15-2010, 05:55 AM
Can I ask you to give more info on this point.

In the real world how much gain could be lost compaired to the computer models.

what factors cause the reduction in Gain.

I want to upgrade my 6m yagi and dont want to waste my time if its not going to make an improvment.


Trevor
EI2GLB



I should append this to state that real world gain at VHF and UHF are often quite different from the models due to losses incurred in practically "everything." At HF, it's a lot less critical and the 4nec2 or EZNEC predicted gains are much closer.

AI3V
02-15-2010, 02:20 PM
Longer boom = more gain.

# of elements, or the type of driven element/matching system have VERY little to do with gain.

Longer boom = more gain.

Rege

EDIT: here is a typical chart:

http://www.sm5bsz.com/antennas/figa3.gif

Its from SM5BSZ (http://images.google.com/imgres?imgurl=http://www.sm5bsz.com/antennas/figa3.gif&imgrefurl=http://www.sm5bsz.com/antennas/cpudes01.htm&usg=__U0ehPGqQN4n2MIlR_GUvy-z4bic=&h=390&w=500&sz=10&hl=en&start=1&itbs=1&tbnid=K_UJhHcpRyITeM:&tbnh=101&tbnw=130&prev=/images%3Fq%3Dgain%2Bof%2Byagi%2Bvs%2Bboomlength%26 gbv%3D2%26hl%3Den) But there are many more out there.

Rege

G4CQM
02-15-2010, 10:20 PM
Steve,

Many thanks for getting back!

Yes, as you say Tonna's were a form of 'Direct Feed' yagi. However, they were designed before computer antenna modeling software was available. And there is a BIG difference because the Tonna is a LOW IMPEDANCE design requiring matching. This was achieved by using a HAIRPIN MATCH, the loop around the feed point. It is a highly inductive and reactive method and with a low impedance design can give rise to higher Ohmic losses and inefficiency.

My yagi designs however offer an alternative, 50-Ohms natural impedance 'Direct Feed'. This is achieved by careful manipulation of the element positions and lengths to produce an antenna that requires NO MATCHING. Meanwhile the 'Coaxial Dipole' deals with the unbalanced to balanced issue. It takes a great deal of skill and care to produce a 50-Ohm yagi and deliver the other desired criteria. There are other designers also offering high performance 50-Ohm designs, G0KSC (http://www.g0ksc.co.uk/), YU7EF (http://www.yu7ef.com/) and even DK7ZB (http://www.qsl.net/dk7zb/) now has some 50-Ohm designs.

ALL of our designs have been publicly scrutinized by Lionel H. Edwards VE7BQH (http://www.sm2cew.com/gt.htm) in his now famous G/T Analysis Table published regularly which is available widely on the Internet.

The BIG SURPRISE is that M2 has just produced a 'Direct Feed' yagi the NEW 6M8GJ (www.bigskyspaces.com/w7gj/6M8GJMAN1_BW.pdf), I wonder why? However, from the images currently available it would appear that they have adopted a HAIRPIN MATCH arrangement, remember the Tonna!

One thing is for sure, all this competition is good for our fellow Radio Amateurs, producing higher performance antennas. Designers are currently striving for that ultimate design and previously ignored criteria such as the H-Plane performance are now coming under the spotlight. My NEXTGEN YAGIS and G0KSC's LFA's are a leap in that direction!

73's Derek Hilleard G4CQM

WB2WIK
02-15-2010, 10:44 PM
Hi Derek,

Yes I know Lionel and I know Lance W7GJ for whom the new 6m beam is "named." And of course I also know Mike.

I'll be looking forward to "trying out" some of these new designs, for sure.

Since 1977 or so (up until right now) the most effective DE system I've come across for VHF-UHF beams is the Quagi design using loop elements for the DE and REF, then planar elements for the directors. It's also "directly fed" with no matching device of any kind and has been demonstrated to yield better results than any sort of planar driven element system on Yagis for 144-222-432-902-1296 MHz. After that, almost everyone goes to loop yagis for 2304-3456 MHz primarily to avoid raindrop detuning at the ends of linear elements.

Actually interlaced Quads on short booms seem to work better than anything for that application. My friend Wayne N6NB (with whom I work some VHF contests, like a few weeks ago in the ARRL VHF SS) has designed a tri-band interlaced quad for 144-222-432 MHz which is very well optimized empirically and works incredibly well for only 2L on 144-222 and 3L on 432 MHz. Very short, easy to use "mobile" (as we do) and no interaction between the three sets of elements. It took a great deal of experimentation to derive this design which can now be easily reproduced by anyone using just household copper wire and some wooden sticks for spreaders!

I drove around with that and 9 other antennas on the roof of my van at speeds up to about 130km/h for an entire weekend, with no damage to anything. When you have beams for 10 bands on the roof of your vehicle, everything holding together at high speeds becomes an important factor.

73

Steve WB2WIK/6

G4CQM
02-16-2010, 05:23 PM
Steve,

I forgot to mention probably the most important issue you have raised in your posts and that is of antenna testing. You are right, it's all very well designing a yagi and having the best design but it's no damn good if the software is giving erroneous predictions and believe me it can happen!

Here in Devon we are fortunate to have a first class testing facility ETC (http://www.etcal.co.uk/) possessing an anechoic chamber and an outdoor test range all calibrated to traceable national standards. I personally have used their services and because of my job I also work with antenna manufacturers who have regularly calibrated test equipment.

It is for this reason that I can say with a high degree of certainty that the software and methods I use are proven and tested. In addition my colleague Richard Mason G6HKS is a keen DXer and regularly tests my designs working 2M and 70CM EME with remarkable success. Richard has a small garden and is surrounded by hills so his yagis have to be good, he was amazed at the results with my 2M-NEXTGEN8 (http://myweb.tiscali.co.uk/g4cqm/powabeamantennas/2M-NEXTGEN8/) in the autumn last year.

Finally, I guess that you and I are not that far apart in our thinking and its been a pleasure having this public debate with you!

73's Derek Hilleard G4CQM

WB3BEL
02-16-2010, 06:49 PM
Making good 2 meter yagi antennas is not really rocket science any more. Lots of excellent designs have been published. Use of antenna modeling software makes it easy to experiment with high confidence that the design will work. Yes, if you do not know what you are doing, you can get garbage out of these analysis tools. This says more about the skill of the craftsman than the quality of the tools.

But, I would hazard to say that if you:
1. Mount the elements considering the influence of the boom.
2. Precisely model the element diameters, lengths and spacing.
3. Use a matching method that is low loss and weatherproof.
4. Avoid feedline radiation.

The antenna will work as modeled. If the antenna is very high Q like minimum element long boom designs optimized for the weak signal part of the band, the degradation in side-lobe performance will occur before major gain reduction or VSWR impairment. It is possible to do null depth and other pattern measurements at home with rudimentary test equipment. You don't need a full blown antenna range. This is not to say that there is no innovation remaining. There are always interesting developments.

In many locations ultra low side-lobe performance is worth a tradeoff in maximum gain. Suburban locations have been transformed over the last decade into cesspools of RFI that kill weak signal work at VHF and above. If you can't hear them you can't work them...

I like direct feed. Lots of folks are using it and it works. No voodoo is needed. If you have been on EME recently you probably have heard this station KB8RQ:

http://www.spasalon.com/kb8rq/

Ferrite beads on the feedline close to the connection to the driven element work perfectly without need for coaxial baluns. That is not to say the coax balun, gamma match, tee match or folded dipole designs are bad. They are not. They work very well but they are a bit more complex, and provide a place to do something wrong if you are not careful. They are also a point where weather effects can cause corrosion or losses.

For UHF and SHF antennas much more precision is required. Making good antennas at VHF is pretty easy in comparison.

WB2WIK I enjoyed reading about your rover exploits in the Jan VHF. Great job. Here on the east coast 2m is the money band. I would not use 2 el quad for sure. I'd find a place for a longer yagi 8 ft boom min. Here its easier to drum up traffic on 2m then move to other bands. But everything on a small rover is a tradeoff...

WB2WIK
02-16-2010, 06:51 PM
Hi Derek,

Excellent, OB!

I do find, as you probably have, that "boom length" alone can't be depended upon to yield high gain antennas and at VHF-UHF many things become quite critical that aren't factors at all when dealing with lower frequencies. While one, two or four long-boom Yagis will surely make e.m.e. contacts especially using WSJT and especially when working others with much larger systems, the most effective e.m.e. antenna (for 144 MHz) I've ever used was 16 short-boom quads where the entire array could be rotated on a central axis to continuously change uplink/downlink polarity. That worked even for old-fashioned moderate speed CW, "ear" copy, and working much smaller stations.

I also use commercial ranges with NIST traceable calibrated antennas to make field gain measurements, and if one has a lot of time (I never do), it's interesting to play around and see what works. At one of the major VHF conferences several years ago, a commercial, calibrated range was set and homebrewers entered all sorts of designs, many just for fun. A 60" long aluminum ladder was fed at a match point and had as much forward gain at 144 MHz as some "commercially sold" 5-6 element Yagis.

Anyway, "home measurement of antenna gain at UHF" is also quite possible if done properly, and I provide all the details on this in two successive articles I published in CQ magazine for December 1985 and January 1986. The article begins, "Measuring Antenna Gain: Anyone Can Do It," and goes on to detail everything about the system and the calculations to achieve excellent accuracy using relatively inexpensive test gear. The articles are loaded with photographs of my actual system as used.

If you're interested these magazine reprints are "on line" at http://hamcall.net/cqcgi

Unfortunately, the download is not "free," but I think for $5 you can download as much as you want for "one day," and there's a lot of issues there.

73

Steve WB2WIK/6

WB2WIK
02-16-2010, 07:01 PM
WB2WIK I enjoyed reading about your rover exploits in the Jan VHF. Great job. Here on the east coast 2m is the money band. I would not use 2 el quad for sure. I'd find a place for a longer yagi 8 ft boom min. Here its easier to drum up traffic on 2m then move to other bands. But everything on a small rover is a tradeoff...

Nice comments. I used to work the VHF contests "back east" when I lived in NJ and still hold some records back there. WB2WIK/4 (FM27) set a record in 1987 for the W4 call area, then we went back and did it again as KC2PX/4 (same group, just switched call signs) from FM27 in 1988. That record held as the highest score ever made in a VHF contest from the W4 area for ten years. Under my own call and from my home stations in the early 70s through early 80s I "won" the VHF SS for the Division several times and nationally twice. My group out west set the all-time high score in any VHF contest as N6CA/6 where I was the 6m station owner and operator at this large multiop, and our record has held for almost 20 years and remains unbroken today.

My "money bands" were every one I could get on! It was always fun working somebody on six or two and then taking them "up the bands," as far as we could go...almost anybody workable on 50 or 144 tropo was also workable on 1.2 GHz, where path losses are higher but so are antenna gains and they just about cancel each other.

The short antennas are good for "roving" because we use ten antennas per vehicle, and they are installed for the entire contest weekend (and usually much longer than that) and must withstand freeway cruising speeds of 80 mph or so without damage. The 10-band mobile setup includes a dual feed dish for 5.7 and 10.3 GHz, which must be "aimed" away from the wind (i.e., towards the rear of the vehicle) to prevent damage while cruising. But once "parked," they all rotate on a common platform, so one rotator is used to turn them all and one needn't ever exit his vehicle to do any of this.

For us, the "money bands" are 1.2 GHz and above where contacts are worth 4 or 8 points each. One contact on 10 GHz is equivalent to 10 contacts on 2m, so a lot of us scramble trying to find open sites where 10 GHz contacts can be made over long distances.

73 & happy VHFing!

Steve WB2WIK/6

KC8VWM
02-17-2010, 08:54 AM
Delete dupe..

KC8VWM
02-17-2010, 09:06 AM
I understand; but it's not a realistic comparison unless you have a way to measure the results (gain). I can build an antenna that "models" at 20 dBi on 2m for $50. But it won't measure that. The M2 antennas have consistently measured their stated gain at numerous commercial ranges, including CSVHF, NEWS, SBMWS, etc. where provisions are made to actually measure real gain within 0.1 dB or so. If we are comparing "apples to apples," that has to include real world gain measurements.

Huh?

A boom is a boom. They are sold on a per foot (not on a Db gain) basis. That's about as "realistic" as it gets isn't it?

A 10mm element is sold by the length, not by the "dB" they will gain.

Stainless steel screws, are well.. Stainless steel screws...

"Results," are about the bottom line costs to construct such an antenna in the performance criteria selection. (lol, I have a business background and hence I am not some Joey Baloney when it comes to this stuff. :) )

We can compare "apples to apples" but yet strangely I am not in the farmers market at the moment because I am actually comparing "dollars to dollars" in this equation.

Since when are aluminum booms, stainless steel parts, antenna elements sold to consumers on a dB gain basis? (other than commercially?)

I doubt you could purchase any commercial antenna design for the same amount of money as it would cost to homebrew one.

I clearly laid out the actual construction costs of building a 12 element, 14.28 dBd, DK7ZB homebrew antenna in an earlier post.

Yet, this is what you seem to suggest and I quote:

"Here they cost about as much as it costs to homebrew a Yagi using good materials."

Now please provide me with a commercial antenna design by M2 of similar physical dimensions for the same costs?

I will only be happy to send them a check today if M2 can somehow produce such an antenna that would equal or exceed these physical dimensions and yet cost the same amount of money to homebrew your own.

Well, can we do that? I have my checkbook ready and pen in hand.

My Best,

WB2WIK
02-17-2010, 05:36 PM
I clearly laid out the actual construction costs of building a 12 element, 14.28 dBd, DK7ZB homebrew antenna in an earlier post.




I understand that, Charles, and it sounds like you did a good job with reasonably priced materials.

The part I questioned was the "14.28 dBd." That's the gain by design, but have you measured it to determine that's what you're achieving?

If not, you don't know what it is. The M2's measure what they claim, sometimes a tad over, on well calibrated ranges. Many homebrew assemblies do not; not that they're bad designs -- most are excellent designs -- just assembly processes vary, and when a lot of homebrewers arrive at a calibrated range to test their antennas, they leave quite disappointed.

I referenced earlier a way to measure VHF antenna gain with the only requirements being the knowledge to do it, some wide open space, and a few hundred dollars worth of readily available test equipment. It's a real eye-opener.

73

Steve WB2WIK/6

KC8VWM
02-17-2010, 11:15 PM
Just to be clear, Steve and I are having a technical discussion comparing the economics, technical merits, advantages and disadvantages of homebrew antennas vs. commercially manufactured long boom Yagi antenna designs. Steve is a great guy, is an excellent elmer, and no I don't pretend to know half of the knowledge base he has acquired over the many years he has been an active ham. :) One thing is clear, I am confident Steve will have an answer to this equation and hopefully we can all learn something about this in the process.

No Steve, I have not measured how this 8 meter hulk of aluminum metal with 12 elements performs on an antenna range compared with a hulk of aluminum of similar size and dimensions made by a commercial antenna manufacturer. (I would love to do this at the VHF conference :)

However, when comparing the construction techniques of both antenna's used (homebrew vs. commercial) I have concluded the following.

It seems that placing the elements over the boom, results in a cleaner and more predictable pattern than the idea of placing elements through the boom like M2 antenna's do.

I also concluded that antenna modeling software has a great deal of difficulty predicting elements that are placed through the boom accurately because of the various densities, thickness, and metal compositions used in the boom construction.

We should also be aware that aluminum is not only comprised of aluminum. For example the composition of aluminum "should" mean we could accurately calculate it's density as (g/cm3) = 2.6980E+00 and it should exhibit a mean excitation energy factor of (eV) = 166.000000

Aluminum alloy 6082 is composed of:

Si: .7-1.3%, Mn: .40-1.0%, Mg: .6-1.2%, Fe:.50% max, Cu:.10%max, Cr:.25%max, Zn:.20%max, Ti:.10%max, Others:each:.05%max,total:.15%max

As we all know, various metals exhibit different performance characteristic criteria when RF is flowing through them. For example copper is a better conductor than aluminum and steel is a poorer conductor than aluminum would be etc. So how can we calculate such variables into the final performance equation?

Another unknown variable that antenna modeling hasn't yet figured out is how to calculate eddy currents flowing along the boom and how they might affect element lengths. This is yet another reason to place elements above the boom in an environment where the behavior and flow of currents can be better "predicted" by the software.

Of course HF is less critical of such issues and more forgiving of deviations in construction but yet it's absolutely paramount to consider such variables into the equation when constructing antenna's intended for VHF and into the microwave region of the bands.

I have also concluded that using 10mm elements in the construction results in better bandwidth and antenna pattern characteristics on VHF. M2 seems to make compromises in this area. Why compromise performance when antenna performance "could" be designed better?

I have also concluded that using square booms results in better mechanical strength and provides for improved accuracy in element placement especially in a homebrew construction environment when compared to placing holes along a round boom.

I have read reviews of such commercially manufactured antenna's using round booms indicating the holes don't even line up with one another. :eek: So, how accurate can such antenna's be in terms of predicted performance levels when the physical construction isn't even accurate?

For example:

W4CX states:

"Instead of instructing the builder to "equalize the lengths of each element side of the boom", the manual should state what that actual lenght should be. I know...it's small stuff, but it makes a big difference in ease-of-construction to not have to do trial and error measurements between the two sides."

HCLAGG states:

"Out of the 8 holes that screw the boom sections together, only 2 of them aligned. The 6 others had to be filed out. Some of the holes were misaligned by 1/8". Quite sloppy.This is disappointing for an M2 product."

Maybe these are isolated incidents?

So while measured antenna performance using an antenna test range is one consideration in determining a long Yagi's engineered antenna performance, Yet, I don't feel it's the final conclusion which accurately determines engineered performance. How does an antenna test range measure beamwidth, antenna temperature, suppressed sidelobe pattern, bandwidth and/or mechanical construction considerations?

My Best,

KL7AJ
02-17-2010, 11:25 PM
Are there any rules-of-thumb for the relative efficiency of two comparable-gain beam antennas based only on the matching devices they use? Here are the two specific antennas in question.

The first is an 11-element design by G4CQM (http://myweb.tiscali.co.uk/g4cqm/g4cqm/netlon/test.htm) using what he calls a direct feed driven element, which just seems cool.

I compared his 11 element design with a Hy-Gain 13 element antenna, because the two beams have about the same gain claims (15.3 dBi for the Hy-Gain and 15.7 dBi for the G4CQM design). THAT led me to look closer at the physical designs, with my attention drawn to the beta match used on the Hy-Gain--is there anything that can be said about the efficiency of the beta match in general, or is it totally specific to the particular, complete antenna design?

Has anyone here built a beam using the G4CQM-style driven element?

"Efficiency" and "Gain" of antennas can be two conflicting things...especially with LARGE arrays. YOu have power lost with each subsequent element. Hopefully the gain you get out of the thing is a fair tradeoff for the loss of efficiency.

The difference between matching systems should be insignificant...as long as they're functioning.

Most people don't realize that a Rhombic is a horribly inefficient antenna..but the gain makes up for it. No different from a long Yagi in that respect.

eric

WB2WIK
02-17-2010, 11:34 PM
Hi Charles,

Wow, I agree there's really no excuse for bad machining/workmanship/QA on commercially manufactured products. I expect to see more of this on homebrew ones!

I have several M2's and their machining and workmanship has been flawless (so far, so good -- let me find some wood to knock) but it's a small sample of maybe 8-10 beams out of I'm sure thousands shipped.

I have definitely seen some rough edges on commercial antennas from Cushcraft, Hy-Gain, Tennadyne. Not so much with Mosley. I bought a $350 vertical (Hy-Gain DX-88) that came in a sealed carton with one wrong tubing section in the kit and spent 20 minutes trying to figure out why nothing fit together! In 100 degree heat, that wasn't much fun.

As for the material variations, yes "modeling" doesn't deal perfectly with many things, which is one justification for actually measuring results rather than going purely by predictions. I like the square booms on my Tennadyne LPDAs, they do make element alignment quite perfect and nearly impossible to screw up.

Re modeling vs. empirical optimization: Telrex (long defunct, but made some great VHF beams in the 50s-60s-70s) used empirical optimization that worked extremely well, building VHF yagis way before anyone had a computer. They used the antenna range at the U.S. Army Signal Corps facility at Ft. Monmouth, NJ. Your tax dollars in action, helping to perfect VHF beam designs.

Their model 2M8, which was an 8 element optimized Yagi on a 14' boom, was unbeatable for years and would blow away anything its boom length made by anyone except someone who precisely copied it. The element spacings were odd and the directors did not taper, they seemed almost haphazard in length.

When NEC first became available for hobbyists, I "designed" an 8L Yagi for 144 MHz, using a hundred reiterations to zero in on the absolute most forward gain possible on a 14' boom, consistent with a non-reactive match to the driven element. After dozens of hours of screwing around, I printed the final design and compared it with the 1950-designed Telrex 2M8.

There's no difference, at all. The elements came out the same lengths and spacings as Telrex used. And they had no computers.

BTW, the Telrex VHF beams all had "through the boom" elements. The element lengths are adjusted to compensate for that. An advantage of "through the boom" is the elements cannot possibly twist or become out of alignment in any way once they're installed, and there's no brackets or insulators or anything to break with age. Once they're there, they'll be there until the aluminum all turns to powder.:p

AI3V
02-18-2010, 03:32 PM
I do find, as you probably have, that "boom length" alone can't be depended upon to yield high gain antennas Steve WB2WIK/6

Boom length is THE parameter that makes antenna gain possible.

Without boom length you have NO CHANCE at gain.

Anybody that tries to convince you otherwise is doing you a disservice.

It's just that simple.

And another thing, anybody that tries to sell you a antenna based on computer MODELING, and not on actual measured performance, say on a EME path, -not a measurement made out in a parking lot somewhere with a ladder and still more "modeling"- doesn't have your best interests in mind.

You do not become a "Big Gun" on ANY band by changing a Gamma match to a quad element, or by adding a ferrite bead or two to an antenna,Or making a element a fraction longer.

These type changes are measured in fractions of a DB in improved G/T of a antenna. They will NOT chang your signal from from S3 to S9+20, you will be lucky to go from S3 to S3+a half needles width.

You become a BIG GUN by putting a LOT of aluminum up HIGH in the air.

And the wholesale price of aluminum is about $2.40/LB in hundred pound lots, simple weighing and multiplication will tell you how much your antenna is "worth"

Rege

WB3BEL
02-18-2010, 06:45 PM
...
And another thing, anybody that tries to sell you a antenna based on computer MODELING, and not on actual measured performance, say on a EME path, -not a measurement made out in a parking lot somewhere with a ladder and still more "modeling"- doesn't have your best interests in mind.
...


I agree with almost everything you wrote with a minor exception to the above paragraph.

It is true that it makes good sense to do some proof test to show that a prototype meets it's objectives by measurement.

But, it is not actually a requirement! The thing might actually work and your testing it does not change how well it works.

I have more faith in computer modeling tools than most folks I guess. Sure, if you don't understand how to use them or some of the limitations you can get "off in the weeds".

But the counterpoint is this. Most folks don't know how to make good antenna measurements either. So which is more important? Maybe it is good to have a simulation and see how it matches to real world measurements. Neither is absolutely essential, but if both look good you may have a higher degree of confidence.

I know that I can simulate certain kinds of antennas and build it with no mistakes and they will just work. 2meter yagis of many types fall into this category. Ones that have simple feed, simple element mounting have fewer places to introduce errors.

So, if you had to choose which to have faith in a model or a measurement which would you rely on? I say it depends solely on the quality of either. How would a non-expert know the difference?

73, Harry WB3BEL

KL7AJ
02-18-2010, 06:49 PM
I agree with almost everything you wrote with a minor exception to the above paragraph.

It is true that it makes good sense to do some proof test to show that a prototype meets it's objectives by measurement.

But, it is not actually a requirement! The thing might actually work and your testing it does not change how well it works.

I have more faith in computer modeling tools than most folks I guess. Sure, if you don't understand how to use them or some of the limitations you can get "off in the weeds".

But the counterpoint is this. Most folks don't know how to make good antenna measurements either. So which is more important? Maybe it is good to have a simulation and see how it matches to real world measurements. Neither is absolutely essential, but if both look good you may have a higher degree of confidence.

I know that I can simulate certain kinds of antennas and build it with no mistakes and they will just work. 2meter yagis of many types fall into this category. Ones that have simple feed, simple element mounting have fewer places to introduce errors.

So, if you had to choose which to have faith in a model or a measurement which would you rely on? I say it depends solely on the quality of either. How would a non-expert know the difference?

73, Harry WB3BEL


I have a great deal of confidence in both NEC-2 and Mininec3 antenna modeling....which only comes about with a great deal of experience. You have to know what the limitations of the programs are...but within those limits, they are VERY trustworthy.

Eric

AI3V
02-18-2010, 10:43 PM
So, if you had to choose which to have faith in a model or a measurement which would you rely on? I say it depends solely on the quality of either. How would a non-expert know the difference?

73, Harry WB3BEL

A very good question!

For me, The measured value, without a doubt.

And that's why I mentioned a EME path as the test range.

Now I have nothing against computer modeling, There has been some very impressive work done by the EME crowd, Surprisingly enough not to maximize forward gain, but to reduce the sidelobes, and therefore improve the G/T of the antenna.

I have read about arrays that deliberately sacrifice a db or so of forward gain for several db improvement in the received s+n/n.

After all, we can either increase s or decrease n!

But in every case the computer modeling was backed up by building, and measuring the array.

And the difference is very small between computer optimized, and "old school" antennas, not enough to matter for 99.99% of stations.- The difference in G/T is only significant when your antenna is not looking at the "warm" earth, but the cold sky of deep space.:D And even then you need arrays of the antennas to measure this.

Rege

WB2WIK
02-18-2010, 10:54 PM
And the wholesale price of aluminum is about $2.40/LB in hundred pound lots, simple weighing and multiplication will tell you how much your antenna is "worth"



That's a bit irrelevant since it has nothing to do with the cost of high strength extruded aluminum alloy, the only type really appropriate for boom material.

1.50" x.065" extruded aluminum 6061T6 "shippable to your door" is limited to 72" lengths unless you want to pay horrendous freight fees for oversized lengths and costs $32.47 per 72", wholesale. If you want to build an 18' boom antenna, this is about $100 worth of HD extruded tubing.

For lighter duty stuff, 1.375" x .058" extruded 6061T6 costs $28.08 per 72" length. An 18' antenna made of this is still almost $90 in material.

http://www.onlinemetals.com/merchant.cfm?pid=4360&step=4&showunits=inches&id=71&top_cat=0

The antenna manufacturer's aren't extruding tubing, they buy it. Wholesale costs are slightly lower, but "aluminum cost per pound" really means nothing.

WB2WIK
02-19-2010, 01:49 AM
And that's why I mentioned a EME path as the test range.


C'mon, Rege, that's silly. I've made e.m.e. contacts with 100W and a single Yagi on a tripod in the backyard, working superstations like W5UN and KB8RQ using ordinary CW. Using JT65, more and easier contacts are not difficult to make.

My method of "measuring" gain is about 95% accurate (within measurement equipment tolerance) and although it takes more than a stepladder and a backyard (I use tall poles and an open field, typically 100WL long) and is fully described and published. Hundreds have used the same system. I can validate manufacturer's claims within about 0.2 dB or so, and in many cases the claims are exactly validated; sometimes not.

Setting up a homebrew range is interesting and allows one to witness what the effects are of bending the boom of a long yagi, slightly canting or kiltering elements off plane, and making quick changes in driven element feed system without other changes. The classic example of N6NB's demonstrations at CSVHF back in the late 70s, where he could take an 11L Cushcraft (a non-optimized beam in many ways!) and simply replace the DE and REF with quad loops then re-test the same antenna on the same range 10 minutes later are startling. Those are all well documented as well, with lots of witnesses. Picking up 1.5 dB by making a 10-minute change in the design -- without changing boom length at all -- at a cost of maybe $2 was an eye-opener for many.

The old 15 element Telrex (28' boom) was tested with the boom perfectly straight, and then the boom "sagging" just a couple of inches on each end, as it might if the supporting guys stretched. Amazing.

Wish we had YouTube back in those days!:p

KC8VWM
02-19-2010, 07:17 AM
Man, this is one heck of a big yagi I am building here at the moment. I have to build it in what "seems" like 4 seperate 10 foot sections. :)

But you have me thinking about changing the reflector and DE to a quad element design to sqeeze out every last possible db in this equasion.

Would I have to remodel the entire antenna or do you think or can I simply exchange one DE for another without incident?

Just wondering what your thoughts and past experiences are on that idea. The 10 minute antenna conversion done on the antenna range Steve was talking about gave me some food for thought.

My Best.

WB3BEL
02-19-2010, 02:38 PM
Charles,
No you can't just change the DE and Ref to quad loops and leave the rest the same.

I think that you are building the DK7ZB 12 ele. I took a quick look on 4NEC2 and it looks bad. You lose 1.5 dB gain if you put the loops at same spacing as the rod version.

I am sure you could re-space the elements and get some back. I have not seen the gain increases claimed by quad loop in long yagi designs. If it really works great...

But 1.5 dB gain is equivalent to 40% longer boom if everything was ideal. Hmm

73,
Harry WB3BEL

AI3V
02-19-2010, 03:13 PM
That's a bit irrelevant since it has nothing to do with the cost of high strength extruded aluminum alloy, the only type really appropriate for boom material.

1.50" x.065" extruded aluminum 6061T6 "shippable to your door" is limited to 72" lengths unless you want to pay horrendous freight fees for oversized lengths and costs $32.47 per 72", ($13.50 at TT,)wholesale. If you want to build an 18' boom antenna, this is about $100 worth of HD extruded tubing.

For lighter duty stuff, 1.375" x .058" extruded 6061T6 costs $28.08 per 72" length. ($12.30 at TT)An 18' antenna made of this is still almost $90 in material.

http://www.onlinemetals.com/merchant.cfm?pid=4360&step=4&showunits=inches&id=71&top_cat=0

The antenna manufacturer's aren't extruding tubing, they buy it. Wholesale costs are slightly lower, but "aluminum cost per pound" really means nothing.

The price I gave in my post above is for 6061-T6 extruded tubing.

The prices you gave are for drawn tubing, that is, tubing that has been extruded, and then subsequently drawn thru additional dies to bring it to final shape and size.

I have added (in red) the TT price for drawn tubing.

There is a radio supplier from Texas (http://www.texastowers.com/aluminum_.htm) that can beat Online Metals price by a large margin.:D

By wholesale, I mean buying thru a metal "service center" as we refer to them in the biz, Typically these folks do not want to fool with orders much under $500 or so.

OTOH, there are a dozen service centers in the Pittsburgh area, all of us run delivery trucks, and charge far less than a LTL CC (Less than truckload common carrier).

You are correct that UPS will only accept 6' lengths, and that LTL truck freight can be expensive, For the home brewer finding a local source within reasonable driving distance might prove to be more economical.

Most suppliers will cut the tubing into UPS lengths, Just like the manufacturers of smaller antennas do.

Rege

AI3V
02-19-2010, 03:35 PM
Charles,
No you can't just change the DE and Ref to quad loops and leave the rest the same.

I think that you are building the DK7ZB 12 ele. I took a quick look on 4NEC2 and it looks bad. You lose 1.5 dB gain if you put the loops at same spacing as the rod version.

I am sure you could re-space the elements and get some back. I have not seen the gain increases claimed by quad loop in long yagi designs. If it really works great...

But 1.5 dB gain is equivalent to 40% longer boom if everything was ideal. Hmm

73,
Harry WB3BEL

I agree, The only thing the quad DE has going for it is ease of matching (just connect the coax to the loop, and nothing else), and that only works out for antennas of modest boom length....The popular Quagi.

The Quagi take advantage of the fact that the impedance of the antenna goes down as the boom gets longer, By using a loop, with it's higher impedance than a dipole, and then only using enough boom so that the impedance drops to 50Ω and not lower. A very clever design in my opinion.

OTOH, this make for a "foolproof" antenna. The difference between 13dbd antenna in a sea of 16dbd antennas will be hard to notice on a "normal" path, But mess up the Driven element/matching system and it's not hard at all to have a nice dummy load.

And how many times has a station installed a new antenna array, and fouled up the phasing? I had that problem on 1296, I got on the band with 4 of the F9FT Yagi's, Found out quite to my dismay that 4 antennas improperly phased are nowhere near as loud as a single antenna:eek: Took a afternoon of experimentation to get those 4 driven elements all going the same way!

I had the same problem on a commercial 435 mhz circular polarized Yagi, The factory phasing lines were just not right, and all the power went into one polarity instead of CP.

What did Scotty say on Star Trek? Something about the more complicated the plumbing the easier it is to clog the pipes?:)

Rege

KC8VWM
02-20-2010, 12:35 AM
Charles,
No you can't just change the DE and Ref to quad loops and leave the rest the same.

I think that you are building the DK7ZB 12 ele. I took a quick look on 4NEC2 and it looks bad. You lose 1.5 dB gain if you put the loops at same spacing as the rod version.

I am sure you could re-space the elements and get some back. I have not seen the gain increases claimed by quad loop in long yagi designs. If it really works great...

But 1.5 dB gain is equivalent to 40% longer boom if everything was ideal. Hmm

73,
Harry WB3BEL

Thanks Harry.

Yes, you are correct I am building the DK7ZB 12 - (27ft long)

I didn't plug the quad loop dimenions into EZNEC but just wanted to confirm my suspicions it was a bad idea.

I'm not so sure if extending the boom 40% longer than it already is to gain an additional 1.5 dB is going to be such a good idea because it will cause many sleepness nights when it gets windy outside.

This photo should give you an idea of the physical "scale" of the DK7ZB antennae project :)

http://i629.photobucket.com/albums/uu18/kc8vwm/dk7zb/MVC-005F.jpg

My Best,

KC8VWM
02-20-2010, 12:42 AM
"UV protected electric wire fence insulators" (microwave oven proof :)) used as antenna element supports. (Mentioned in an earlier post in this thread - "Bag of 25 for $5.00")

http://i629.photobucket.com/albums/uu18/kc8vwm/dk7zb/MVC-008F.jpg

http://i629.photobucket.com/albums/uu18/kc8vwm/dk7zb/MVC-007F.jpg

Aluminum inside almuminum to join the sections together.

http://i629.photobucket.com/albums/uu18/kc8vwm/dk7zb/MVC-006F.jpg

Hopefully, I will have this VHF flamethrower up in the air soon.

Gud DX :D

EI2GLB
02-20-2010, 08:18 PM
This is what it will look like when finished.

http://i154.photobucket.com/albums/s261/fatboymotorsport/th_12eldk7zb.jpg (http://s154.photobucket.com/albums/s261/fatboymotorsport/?action=view&current=12eldk7zb.jpg)

http://i154.photobucket.com/albums/s261/fatboymotorsport/th_DSCF1288.jpg (http://s154.photobucket.com/albums/s261/fatboymotorsport/?action=view&current=DSCF1288.jpg)

Click both to see them larger


Thanks Harry.

Yes, you are correct I am building the DK7ZB 12 - (27ft long)

I didn't plug the quad loop dimenions into EZNEC but just wanted to confirm my suspicions it was a bad idea.

I'm not so sure if extending the boom 40% longer than it already is to gain an additional 1.5 dB is going to be such a good idea because it will cause many sleepness nights when it gets windy outside.

This photo should give you an idea of the physical "scale" of the DK7ZB antennae project :)

http://i629.photobucket.com/albums/uu18/kc8vwm/dk7zb/MVC-005F.jpg

My Best,

KC8VWM
02-20-2010, 09:34 PM
http://i154.photobucket.com/albums/s261/fatboymotorsport/th_12eldk7zb.jpg

Very Nice!

Thanks for sharing. :)

W4TMW
02-21-2010, 03:49 AM
I'd like to ask a question here if I may....

I'm right in the middle of building a G0KSC 28 el LFA antenna ( 25' boom) for 70cm.

My question to anyone is...what size tubing is being used for these very long boom antennas to maintain good rigidity? Is it possible to use a tube that is so large that it interacts with the elements? The G0KSC design requires the elements to be insulated from the boom. I'm using my own design insulators from Teflon for best possible electrical insulation, so elements 'through the boom' is not a factor.

I see the M2 antennas use 1 1/2" round tube and most Europen builders are using a little smaller square tubing with round tube inserted inside. I've always been under the impression that round tube is stronger than sq tube if the sizes are the same. I'm drilling my boom tubes on a milling machine so accurate positioning on round tubing isn't a problem.

Anyone have any suggestions about boom tubing sizes ( both OD and wall thickness) on these long boom antennas.

Thanks,
Tom
wb3jhp

KC8VWM
02-21-2010, 05:21 AM
Hi Tom,

You can use a round boom to build the G0KSC 28 el LFA antenna but you would have to redesign the antenna somewhat.

The antenna is modeled for use on a square boom, so I would assume G0KSC has applied the dimensional changes (A 25 -45% correction factor is the general rule!) necessary for element lengths, spacing etc when he's publishing his designs intended for use on square booms.

Otherwise, you might email the designer (Justin) and ask if he has a comparable design for use on your specified round boom.

In the old days, the actual boom was never really much of a consideration other than a support structure. Old designs did not consider the boom size as it related in terms of element size, etc but recent advances in antenna technologies involving next generation Yagi's now account for such variables in their design which contributes to their improved efficiency.

Some readers may be interested in reading what AntenneX (Jan 2010 Issue) has just published in a .Pdf article pertaining to how booms (both square and round for that matter) can have a direct influence on element lengths, gain and pattern characteristics when designing them. It a very well written article which directly addresses the issue:

AntenneX - Jan. 2010 Article (http://yu1aw.ba-karlsruhe.de/yagi_sq_boom_corr.pdf)

Also, I noticed G0KSC is using 1/4" rod in the design you are building. I understand when elements are thin like that, they exhibit a very high Q (for higher gain) but yet, the problem with high performance designs like this one is in the fact they are not very forgiving or tolerant when it comes to any dimensional changes contrary to the originally published design. Therefore, it's very important to follow every dimension exactly as published for the antenna to perform as desired. There is absolutely no room for errors, not even a 1mm +/- tolerance in materials or element lengths is considered acceptable when building high Q antenna designs like this one.

Some people believe when the antenna's don't turn out as expected, that compensating the DE and perhaps R and D1 for best VSWR will equate to mean it somehow corrects the problems encountered, but it doesn't because the intended design has now changed "Q" and therefore the desired pattern, gain, F/B ratio and/or sidelobe suppression has been compromised regardless of how you manage to "compensate" the VSWR for the better.

Antenna's using thicker elements (say 3/8" - 10mm etc.) tend to be more broadbanded, are more forgiving and tolerant of variations in their constructed design. These designs (DK7ZB design I am building for example) are typically easier and somewhat more forgiving to construct for most people at their homebrew antenna factory.

Admittedly, you do tend to give up a little gain in exchange for a little more bandwidth, improved accuracy in terms of expected outcome and more predictable pattern results etc.

Incidently, you can get 8' - 1"x1" (25mm x 25mm) sections of square aluminum boom at home depot for about $15 if that should help.

Booms (round or square) that are constructed this long will typically require additional supports to assure maximized mechanical integrity.

These are sample diagrams indicating how your long Yagi should be supported:

Support Diagram #1 (http://www.qsl.net/dk7zb/2m-longyagi/construction_1.gif)

Support Diagram #2 (http://www.qsl.net/dk7zb/2m-longyagi/construction_2.gif)

Hope that helps Tom and good luck with your build.

My Best,

W4TMW
02-22-2010, 02:07 AM
Charles,

Thanks for all of the info. You have given me a great tip !!!! Square tubing at Home Depo....I never thought about that. I originally wanted to go with square tubing but I couldn't find anything.....at a reasonable price in small lots. I have ordered round tubing from DX Engineering, however, they no longer carry 1/4" rod material so I went to SpeedyMetals for the elements.

The only thing that concerns me about the Home Depo tube is wheather it's 6063 or not but I hope to find out tomorrow.

I suppose it would be ok to go with ANY tempered 6063 as opposed to the T832 temper which Texas Towers and DX Engineering sells expressly for antennas. And for these long booms, I would not consider thin wall such as .049" but prefer .o58' as a minimum.

As far as boom sag is concerned, do you think it would be smarter to use a tapered boom design to reduce the additional weight at the outer ends?

Thanks again

Tom
wb3jhp

KC8VWM
02-22-2010, 03:29 AM
As far as boom sag is concerned, do you think it would be smarter to use a tapered boom design to reduce the additional weight at the outer ends?

Thanks again

Tom
wb3jhp

These square booms are very strong. They are often used to construct much heavier Yagi's for operation on 6m.

Naturally the ends may sag because of the length, but there are several ways to deal with that.

One idea is as you suggested by tapering the ends of the boom.

Home depot also carries smaller diameter square tubing that fits inside the larger square tubing, although I think these sections are limited to 6 feet in length.

However, this will not usually work if you are following any "through the boom" high Q element antenna designs unless you make changes in the prediction software to accommodate the tapered boom changes. However, this idea works perfectly for most isolated element - above the boom element designs especially where the boom is not "as critical" in terms of "Q" in the intended design. Of course, you would have to ensure the elements line up on the ends in terms of height placement above the boom for this to work correctly.

You might also study these photos for a second idea:

http://www.qsl.net/dk7zb/2m-longyagi/19-Ele-HB9YIT.jpg

http://www.qsl.net/dk7zb/2m-longyagi/12-El-SW6IED.jpg

http://www.qsl.net/dk7zb/2m-longyagi/11-Ele_DJ8MS.jpg


Basically it's a shorter boom placed directly below the longer boom. You need to cut 1" x 1" squares and place them between the 2 booms at various point which gives it a "ladder" like appearance.

Bolts are placed between the two booms to "build the ladder" between the two booms.

The lower boom can now be "pushed up" and is fully independent of the upper boom when bolted to the mast. Pushing the lower boom results in elevating the upper boom in place at the desired alignment. The fact the lower boom extends directly under the upper boom serves to eliminate sag and does not affect or interfere with the antenna pattern because it's placed along the "zero point" of the antenna.

The "ladders" are like placing hands directly under the antenna boom to support it into position if that makes sense.

Guess you have to study the photo in the link to get the idea in clearer detail. :)

The square tubing boom material you are looking for is located in the "angle iron and long threaded bolts" hardware section. (usually not far from the stainless steel nuts and bolts!) They also carry some round aluminum tubing that seem suitable for Yagi element construction but they may not always have the correct diameters or you may encounter a limited supply stock in some cases. But then again, sometimes you can get everything you need in terms of aluminum elements if they had a recent delivery and no one has bought them out yet. This approach also saves on shipping costs.

My Best,

W4TMW
02-24-2010, 04:24 AM
Charles,

Well, I visited Home Depot and found the 1" sq. you mentioned but the wall thickness is only .50" which makes it good as 'doubler' and I don't see the temper. It seems that most hardware store type tube is 6063 but temper is T5...softer than T8 I think.

I've looked and looked for 1 1/4" sq., which is recomended as a minimum per Justin, but only found 1/8" wall tube which i think is over-kill and too much weight in the air.

So, I'll be ordering from DX Engineering and go with 1.375"x .058" wall and taper down from there.

73 and thanks for your comments.
Tom
wb3jhp

W4TMW
02-24-2010, 09:24 PM
Just checking back, I found a mistake in my last post. The Home depot wall thickness should read .050'. Actually the label on the tube reads 1/20".

Tom
wb3jhp

ad: dxeng