Hello Everyone
Has anybody checked out, or better still built, one of G0KSC's OWA (optimised wideband arrays) for 6m, 2m, 70cm?
Website is: http://www.g0ksc.co.uk
They look to be a robust and quite simple design although I would use Teflon rather than perspex insulators. He uses square tube, which suits me because I have some already, and a direct feed to a split DE with no "hairpin" and just a coax balun.
Please have a look and let me, and others, know what you think. My big rebuild is planned for April - August and I am exploring as many ideas as I can.
73
Wayne VK4WDM
G0KSC OWA yagis
Re: G0KSC OWA yagis
I'm impressed! He uses very thick elements for 2M which is interesting.You will probably use perspex for insulators after you price 15mm thick teflon ...
Bob, ZL1RS in the Bay of Islands at RF64vs
-
VK4QB
Re: G0KSC OWA yagis
Hi Guys
Don't forget, Lo Q elements, wide bandwidth, lo gain ! You can't have them all. Plus there is the added weight !!
Brian 4QB
Don't forget, Lo Q elements, wide bandwidth, lo gain ! You can't have them all. Plus there is the added weight !!
Brian 4QB
Re: G0KSC OWA yagis
I built a 70cm OWA from a Cebik article some time back. It works as expected. Whilst it is true that there are compromises as detailed above - you have to weigh up if these bother you. From my point of view the direct 50 ohm match and the resultant simplifying of feed is a tradable benefit - especially in wet and harsh climates where weatherproofing is an issue. I would recommend Lexan as apposed to perspex as it stands up to UV and is more robust - I have a 6m beam with Lexan insulated feed that shows no sign of deteriation after 15 years in the sky.VK4QB wrote:Hi Guys
Don't forget, Lo Q elements, wide bandwidth, lo gain ! You can't have them all. Plus there is the added weight !!
Brian 4QB
-
VK4QB
Re: G0KSC OWA yagis
A further thought and this is a big thing, is that there is no earth on the driven element and feedline EXCEPT AT THE TRANSCEIVER . Not a good idea for lightning, or near misses..But I guess there are fores and againsts in most things.
Brian 4QB
Brian 4QB
Re: G0KSC OWA yagis
An alternative set of designs can be found here - DK7ZB yagi designs
I've built 6m and 2m versions, which perform quite well, and are easy to construct. They are not, in general, wideband arrays however I've used both antennas at different portions of the respective bands with only moderate degradation in SWR and performance.
I've built 6m and 2m versions, which perform quite well, and are easy to construct. They are not, in general, wideband arrays however I've used both antennas at different portions of the respective bands with only moderate degradation in SWR and performance.
73
Andy VK6OX
If you can't explain it simply, you don't understand it well enough. Albert Einstein
Andy VK6OX
If you can't explain it simply, you don't understand it well enough. Albert Einstein
-
VK4WDM
Re: G0KSC OWA yagis
Certainly interesting points! The weight issue was certainly something I picked up. I had not thought about the "only grounded at the transceiver" point - certainly important up here in the tropics.
I have had a look at both the YU7EF and DK7ZB designs. Plenty of options there, and I can't forget the designs of local guru VK4ABW, who gave me the design for my present 7 el 6M antenna.
The first thing I need to do is get the tower down and see how good a condition the existing elements and booms are and whether they can be used for a rebuild. Then there is the Com Antenna option as per my previous post.
73
Wayne
I have had a look at both the YU7EF and DK7ZB designs. Plenty of options there, and I can't forget the designs of local guru VK4ABW, who gave me the design for my present 7 el 6M antenna.
The first thing I need to do is get the tower down and see how good a condition the existing elements and booms are and whether they can be used for a rebuild. Then there is the Com Antenna option as per my previous post.
73
Wayne
-
VK4QB
Re: G0KSC OWA yagis
Wayne,
A thing that puzzles me is, we all know that the reactance at the feed point changes with frequency. He does not say how he overcomes or compensates for that, and seems to assume it is of no consequence and the balun takes care of that ..MMMMMM
I think that I would like Owen 1OD to make an analysis of this antenna. Something does't seem right. The insulated elements are Ok but the DE and balun /matching seem a bit sus. Maybe I'm wrong , but ???
73
Brian 4QB
A thing that puzzles me is, we all know that the reactance at the feed point changes with frequency. He does not say how he overcomes or compensates for that, and seems to assume it is of no consequence and the balun takes care of that ..MMMMMM
I think that I would like Owen 1OD to make an analysis of this antenna. Something does't seem right. The insulated elements are Ok but the DE and balun /matching seem a bit sus. Maybe I'm wrong , but ???
73
Brian 4QB
Re: G0KSC OWA yagis
"Lo Q elements, wide bandwidth, lo gain ! You can't have them all".
These antennas are computer designed/optimised for wideband performance. Yes, G0KSC has compromised gain (and F/B and sidelobes) to get this ... BUT the interesting thing with these designs are that the compromises are in the order of 0.1dB in gain and a few dB in side lobes and F/B ratio compared to narrow bandwidth, "high gain" designs on the same boom length. None of these compromises will have an appreciable effect on overall antenna performance, even for some demanding application like EME ... except in the most extreme situations e.g. very high man-made noise environment (in which case, one should sell up and move!!)
The benefit is an antenna that can be built simply by following the given dimensions (without a micrometer!) to give comparable performance, with "instant" low SWR without time spent adjusting a (potentially narrow band) driven element matching system like a T-match or the horrible gamma match! The performance is "guaranteed" due to it's computer optimisation and very simple no-loss feed system ... in short, all the effort has gone into the design using the power of the latest computer antenna software to provide a real simple antenna that "just works".
Designs like DK7ZB's offerings are fine, but in my opinion the "DK7ZB match" (impedance transformation balun) introduces as much loss as the antenna provides in extra gain in comparison to these G0KSC OWA designs. In this case, it's a matter of what you gain on the swings, you loose on the roundabouts ... but we are only talking 1/10th or so of a dB gain here and that won't gain or loose many QSO's
"we all know that the reactance at the feed point changes with frequency. He does not say how he overcomes or compensates for that"
I thought he did? These particular multi-element yagi designs have been computer optimised to create wide band yagis that have little change in feedpoint impedance/reactance (and gain, and F/B ratio) within the frequency range of interest. See the graphs on G0KSC website with each antenna (I only looked at the 2M and 70cm designs) ... the feedpoint impedance/reactance only changes an ohm or so over 144-145MHz.
A similar approach is taken by YU7EF and VE7BQH with their wide bandwidth "50ohm" yagi designs and after building about ten yagis (for 2M) designed with this way over the past couple of years, all I can say is "the approach works ... and works well". With one 13 element design on a 6.3M boom it was a case of cut -drill - assemble - attach feedline - test, and I had a "perfectly acceptable" 13dBd gain 2M antenna built in one afternoon. The SWR was less than 1.2:1 from 143 to 146MHz and I put many EME QSO's in the log with the eventual 2 x 13 element array and just 200W output ... including (with a little patience) Rex VK7MO. Thanks Rex, that was one of the most satisfying moments of that trip
As a matter of interest, feedpoint impedance in a yagi antenna can do some "unbelievable" things depending on the spacings and lengths of the elements used. It is actually possible to design a yagi antenna whose SWR dip LOWERS in frequency as you SHORTEN the driven element! This had me stumped when I built a W1JR 8 element yagi and put the SWR meter on it. SWR was best below 144.0, ah I can fix that by chopping off a few mm from each end of the split dipole. Sorry Bob, SWR dip moves even lower in frequency Adding extra length to the tips bought the SWR up in frequency and into the 2M band again! I went back to the antenna software looking for answers and found the cause was a "backwards" reactance curve for that particular design 
I've kept a close eye on this effect ever since and found several yagi designs exhibit it ... it's one of the things I find fascinating about this hobby.
"... and seems to assume it is of no consequence and the balun takes care of that ..MMMMMM"
As above, the feedpoint impedance/reactance vs frequency comes from the "structure" of the yagi element lengths and spacings along the boom, and will not be taken care of in the balun. The simple coiled choke balun he suggests just does what a balun is supposed to do ... going from the balanced split dipole feedpoint to the un-balanced coaxial cable feedline (in a basic sort of way) and is there to minimise current flow on the outer of the coax that would otherwise radiate and ruin the sidelobe suppression and possibly introduce radiation pattern squint. There are other baluns that might be "better", like the sleeve balun or (maybe?) the ferrite choke balun. None of those are going to take care of the antenna's feedpoint resistance change ...
"I had not thought about the "only grounded at the transceiver" point - certainly important up here in the tropics"
I've used this split dipole approach on all 2M yagis in my EME arrays since 2003 and operated in several lightning prone countries with no problem ... including one place where we experienced power line/phone line strikes that took out telephone and computer network equipment in the same building that I had the antenna array mounted on. After going through several lightning strike disasters at work I came to the idea that electronic equipment lightning protection/survival is somewhat a matter of planning, but mostly a matter of good luck!!
If lightning is a concern, the split dipole driven element on the G0KSC OWA yagis can be replaced with a folded dipole so the middle of the folded element can be grounded to the boom and 4:1 halfwave coax balun with the braids grounded to the boom at the feedpoint ... and introduce some tenths of dB loss due to the coax in the balun ... and some hours of extra construction time. Alternately, the split dipole no-loss feed system can be retained using a sleeve balun with the outer coax jacket grounded to the boom at the point 1/4wl back from the feedpoint (it provides a ground to one side of the coax at least).
"Plus there is the added weight !!"
I grabbed a piece of 12mm perspex I had out in the shed ... it was big enough to make 13 insulators for a 2M yagi and weighs about 700gm. After saw cuts I estimate about 45 grams per insulator ... so maybe 55-60 grams per insulator for 15mm material. 12.7mm diameter aluminium tube weighs 30 grams per meter more than the typically used 9.6mm tube with a 1.2mm wall thickness used in VK for cockie proof 2M yagi elements, so about 26 grams extra per element. G0KSC is recommending 1 and 1/4" square aluminium tube boom material for the longer 2M booms. That won't sag appreciably and will certainly survive the elements ... and the cockies
... and after all, these G0KSC antennas are just small ones
But, in the end, which ever way the individual ham decides in the best one to follow, an antenna in the air will work more DX than one on the computer file
73, Bob
These antennas are computer designed/optimised for wideband performance. Yes, G0KSC has compromised gain (and F/B and sidelobes) to get this ... BUT the interesting thing with these designs are that the compromises are in the order of 0.1dB in gain and a few dB in side lobes and F/B ratio compared to narrow bandwidth, "high gain" designs on the same boom length. None of these compromises will have an appreciable effect on overall antenna performance, even for some demanding application like EME ... except in the most extreme situations e.g. very high man-made noise environment (in which case, one should sell up and move!!)
The benefit is an antenna that can be built simply by following the given dimensions (without a micrometer!) to give comparable performance, with "instant" low SWR without time spent adjusting a (potentially narrow band) driven element matching system like a T-match or the horrible gamma match! The performance is "guaranteed" due to it's computer optimisation and very simple no-loss feed system ... in short, all the effort has gone into the design using the power of the latest computer antenna software to provide a real simple antenna that "just works".
Designs like DK7ZB's offerings are fine, but in my opinion the "DK7ZB match" (impedance transformation balun) introduces as much loss as the antenna provides in extra gain in comparison to these G0KSC OWA designs. In this case, it's a matter of what you gain on the swings, you loose on the roundabouts ... but we are only talking 1/10th or so of a dB gain here and that won't gain or loose many QSO's
"we all know that the reactance at the feed point changes with frequency. He does not say how he overcomes or compensates for that"
I thought he did? These particular multi-element yagi designs have been computer optimised to create wide band yagis that have little change in feedpoint impedance/reactance (and gain, and F/B ratio) within the frequency range of interest. See the graphs on G0KSC website with each antenna (I only looked at the 2M and 70cm designs) ... the feedpoint impedance/reactance only changes an ohm or so over 144-145MHz.
A similar approach is taken by YU7EF and VE7BQH with their wide bandwidth "50ohm" yagi designs and after building about ten yagis (for 2M) designed with this way over the past couple of years, all I can say is "the approach works ... and works well". With one 13 element design on a 6.3M boom it was a case of cut -drill - assemble - attach feedline - test, and I had a "perfectly acceptable" 13dBd gain 2M antenna built in one afternoon. The SWR was less than 1.2:1 from 143 to 146MHz and I put many EME QSO's in the log with the eventual 2 x 13 element array and just 200W output ... including (with a little patience) Rex VK7MO. Thanks Rex, that was one of the most satisfying moments of that trip
As a matter of interest, feedpoint impedance in a yagi antenna can do some "unbelievable" things depending on the spacings and lengths of the elements used. It is actually possible to design a yagi antenna whose SWR dip LOWERS in frequency as you SHORTEN the driven element! This had me stumped when I built a W1JR 8 element yagi and put the SWR meter on it. SWR was best below 144.0, ah I can fix that by chopping off a few mm from each end of the split dipole. Sorry Bob, SWR dip moves even lower in frequency
Adding extra length to the tips bought the SWR up in frequency and into the 2M band again! I went back to the antenna software looking for answers and found the cause was a "backwards" reactance curve for that particular design I've kept a close eye on this effect ever since and found several yagi designs exhibit it ... it's one of the things I find fascinating about this hobby."... and seems to assume it is of no consequence and the balun takes care of that ..MMMMMM"
As above, the feedpoint impedance/reactance vs frequency comes from the "structure" of the yagi element lengths and spacings along the boom, and will not be taken care of in the balun. The simple coiled choke balun he suggests just does what a balun is supposed to do ... going from the balanced split dipole feedpoint to the un-balanced coaxial cable feedline (in a basic sort of way) and is there to minimise current flow on the outer of the coax that would otherwise radiate and ruin the sidelobe suppression and possibly introduce radiation pattern squint. There are other baluns that might be "better", like the sleeve balun or (maybe?) the ferrite choke balun. None of those are going to take care of the antenna's feedpoint resistance change ...
"I had not thought about the "only grounded at the transceiver" point - certainly important up here in the tropics"
I've used this split dipole approach on all 2M yagis in my EME arrays since 2003 and operated in several lightning prone countries with no problem ... including one place where we experienced power line/phone line strikes that took out telephone and computer network equipment in the same building that I had the antenna array mounted on. After going through several lightning strike disasters at work I came to the idea that electronic equipment lightning protection/survival is somewhat a matter of planning, but mostly a matter of good luck!!
If lightning is a concern, the split dipole driven element on the G0KSC OWA yagis can be replaced with a folded dipole so the middle of the folded element can be grounded to the boom and 4:1 halfwave coax balun with the braids grounded to the boom at the feedpoint ... and introduce some tenths of dB loss due to the coax in the balun ... and some hours of extra construction time. Alternately, the split dipole no-loss feed system can be retained using a sleeve balun with the outer coax jacket grounded to the boom at the point 1/4wl back from the feedpoint (it provides a ground to one side of the coax at least).
"Plus there is the added weight !!"
I grabbed a piece of 12mm perspex I had out in the shed ... it was big enough to make 13 insulators for a 2M yagi and weighs about 700gm. After saw cuts I estimate about 45 grams per insulator ... so maybe 55-60 grams per insulator for 15mm material. 12.7mm diameter aluminium tube weighs 30 grams per meter more than the typically used 9.6mm tube with a 1.2mm wall thickness used in VK for cockie proof 2M yagi elements, so about 26 grams extra per element. G0KSC is recommending 1 and 1/4" square aluminium tube boom material for the longer 2M booms. That won't sag appreciably and will certainly survive the elements ... and the cockies
... and after all, these G0KSC antennas are just small ones
But, in the end, which ever way the individual ham decides in the best one to follow, an antenna in the air will work more DX than one on the computer file
73, Bob
Bob, ZL1RS in the Bay of Islands at RF64vs
Re: G0KSC OWA yagis
Today I built a modified version of the G0KSC 7 element 2M design (3.2M boom version)...
Retained the spacings but modified the lengths to use 3/16" dia elements through the boom ...
New lengths modified/verified by using 4nec2 software ...
(lost 0.1dB in the process, due to the small element diameter)
Through boom corrections taken from a table on YU7EF's website (rounded to the nearest mm)...
The driven element is a folded dipole the same length as the "dipole" specified ...
Fed with a 4:1 half-wave balun to a very short feedline ...
SWR dipped to 1:1 (50ohms X=0) at 144.4MHz on my MFJ259 and was low across the bottom end of the band.
(I'll get some better figures tomorrow).
... that was a satisfactory result, so I made another 3
... will do the stacking frame tomorrow
These are ultra-lite versions on a 16mm square aluminium tube boom with a 12.5mm square tube end extension ...
... for another Pacific Island EME DXpedition
73, Bob ZL1RS
Now as a new topic at viewtopic.php?f=11&t=8423&p=6341#p6341
Retained the spacings but modified the lengths to use 3/16" dia elements through the boom ...
New lengths modified/verified by using 4nec2 software ...
(lost 0.1dB in the process, due to the small element diameter)
Through boom corrections taken from a table on YU7EF's website (rounded to the nearest mm)...
The driven element is a folded dipole the same length as the "dipole" specified ...
Fed with a 4:1 half-wave balun to a very short feedline ...
SWR dipped to 1:1 (50ohms X=0) at 144.4MHz on my MFJ259 and was low across the bottom end of the band.
(I'll get some better figures tomorrow).
... that was a satisfactory result, so I made another 3
... will do the stacking frame tomorrow
These are ultra-lite versions on a 16mm square aluminium tube boom with a 12.5mm square tube end extension ...
... for another Pacific Island EME DXpedition
73, Bob ZL1RS
Now as a new topic at viewtopic.php?f=11&t=8423&p=6341#p6341
Bob, ZL1RS in the Bay of Islands at RF64vs