| Author |
Message |
    
Tech808
Moderator
Username: Tech808
Post Number: 2530
Registered: 8-2002
| | Posted on Sunday, May 30, 2004 - 8:09 am: | 
|
Bouk,
First Let me Welcome you to the Copper Forum.
Our Antenna Expert Tech833 will not be back until Tuesday due to the Holiday MEMORIAL DAY here in the US.
I am sure that Tech833 or another Forum Member will be able to offer you some Helpful Advice.
We have a lot of Copper Forum Member's here who just Love to build there own Antenna's.
OK, People let's jump in to help our Newest Forum Member here with his Question.
Lon
Tech808 |
Highlander
Advanced Member
Username: Highlander
Post Number: 518
Registered: 12-2001
| | Posted on Sunday, May 30, 2004 - 8:54 am: |
|
Is the quad fed vertically or horizontally?
Also, were you comparing with actual radio signals, or just looking at the noise floor? Quads are less prone to some forms of noise than an end-fed vertical radiator. Also, a 4 element quad will have a pretty tight pattern, so if you are not pointed directly at your intended station, an omni might get a better signal.
It could also be a feedline issue, like a bad solder connection, lossy coax, too long of a run, etc. |
Dindin
Intermediate Member
Username: Dindin
Post Number: 136
Registered: 2-2004
| | Posted on Sunday, May 30, 2004 - 9:57 am: |
|
Welcome Bouk,
it sounds like polairity is giving you the drop in signal you see.I have built a few quads and put my feed point half way between the spreaders instead of on the end of one spreader.like the photo (fig8 on the website you gave).believe Scotts drawings (fig 4)are all backwards.think when Bullet sees this he will verify if I'm right or wrong.Info i used when i built quads came mostly from one of the ARRL antenna building books which i no longer have to check. |
Marconi
Intermediate Member
Username: Marconi
Post Number: 299
Registered: 11-2001
| | Posted on Sunday, May 30, 2004 - 10:28 am: |
|
Bouk, not to worry, for a while now, in my area of Texas at least, my beams have not worked well for skip either. I currently have up a 4 element yagi and a new 3 element yagi I will be testing as soon as skip conditions get a little better. I do not have a vertical side to test aginst my vertical A99, but I suspect that when conditions are like they are right now, the beams may put too much RF too high in the sky to really be considered effective. I do not know what causes this, but this is what I see with DX at least almost every year. It is like why do certain areas of the world come in at certain times of the year? I don't know why, but I know it happens.
This has been evident to me over some years now, so I would not worry much. I am also not sure about the wire quads, but after I build a yagi it always seems to take a while, blowing around in the wind, to get to settle down so-to-speak.
If I had a vertical and or some stations around with horizontal polarity, that I could test against locally, maybe I would be a little disappointed with the local response you are getting, but maybe the conditions I mention above affects local stuff just as well. My beams just are not working like I know they can when conditions get better than they are right now.
I have not done so yet, but I will go look at the site you noted. Describe a little of how you are feeding using the gamma. Are you sure you are feeding vertically?
I note your reference is a Signal Engineering site. They typically feeds their quads at a voltage node. This effectively changes the traditional feed point noted for traditional quads from 3:00/9:00 for the vertical feed point and 6:00/12:00 for the horizontal feed point to the very opposite polarity. They are unique in their design and doing so may provide them some additional gain if the impedance can be controlled. Doing it their way also takes real good care of the balance/unbalance problem and this problem can be a killer for a dual polarity beam which their feeder controlls quite nicely using only one driven element.
I see SE design as a 1/4 open stub feeder where basically the hot side is attached to a continious wire loop (not insulated) at the top end of the feeder. Instead of appearing as a folded dipole to the feed point, feeding at some insulator known as a current node in a traditional quad, electrically their quad feeder, a 1/4 wave stub simply sees the loop appear as a very fat 1/2 wave extension and thus the change in affective feed point polarity is accomplished by design.
This is how I see their feeder connecting directly to the loop without and insulator at the connection.
Good luck and be patient, |
Bouk
New member
Username: Bouk
Post Number: 2
Registered: 5-2004
| | Posted on Sunday, May 30, 2004 - 12:14 pm: |
|
Ok i will try to give you all more info, many thanks by the way for all the answers.
The quad is vertically polarized.
It is mounted like a + and not like a x.
The feed element is fed "three o clock" (gammamatch).
The feed element is also a continious loop with the coaxshield directly to the wire and the centerlead of the coax to the gammamatch wich is placed on the backside of the feedloop (between feed and reflector)
Could the gammamatch have something to do with this not working?? Building a gammamatch was the only way we managed to get down the swr.
We measured the s-readings against another local cb-user (thanx niefem).
Any ideas?? |
Dindin
Intermediate Member
Username: Dindin
Post Number: 138
Registered: 2-2004
| | Posted on Sunday, May 30, 2004 - 4:14 pm: |
|
ok Bouk,sound to me like you are polorized horizontal,try rotating your boom 90 degrees so the feed is at 6 or 12 o'clock and check reading again.
|
Marconi
Intermediate Member
Username: Marconi
Post Number: 300
Registered: 11-2001
| | Posted on Sunday, May 30, 2004 - 10:45 pm: |
|
Dindin, I discussed this in my previous post and you may be right if he is doing something like SE does in the way he feeds his quad. SE does not use a gamma matcher however, it is an open ended stub which is basically a 1/4 wave antenna that works against the boom or another wire in the smaller versions of their quads as the ground. This make for feeding at a high voltage point and that is what changes the location of the feed point relative to the polarity.
A gamma matcher requires a pretty good support and it doesn't seem practical for use on a wire antenna the way I invision it at this point.
Avanti used to build the PDL II and they used an orbital tuner what used a small gamma matcher to reach out and tape part of the orbital tuner at a 50 ohm point and that is how they matched their antenna. However, their loop had an open end an insulator that broke the wire between the hot and the ground, and Bouk tells us that he is tapped to ground using the shield on the continious wire loop. SE does something like this, but they feed the hot to the wire loop and the ground is attached to the boom and is left open at the loop end.
So, I not sure, at this point, whether he is feeding a V or a H. Maybe we will learn more soon.
Hang loose,
|
Dindin
Intermediate Member
Username: Dindin
Post Number: 140
Registered: 2-2004
| | Posted on Sunday, May 30, 2004 - 11:13 pm: |
|
Marconi,I also am having trouble picturing a gamma match on his quad.although from his discription i envision more of a beta match which is really not suitable for use on a quad.had one computer modeled once and lobes were terible,pattern looked more like an explosion in cartoon.we'll see. |
Ca346
Advanced Member
Username: Ca346
Post Number: 799
Registered: 8-2003
| | Posted on Monday, May 31, 2004 - 12:13 pm: |
|
BOUK:
I have the 'real' SE 4 element quad. (Lightning 4+) Their gamma match is a killer to get a flat SWR for me, but I like the dual polarity and a tuner eliminates the high swr on one problem side. I have not noticed a big difference between horiz and vert polarity on my antenna. I notice the diff, but it's not huge. What is huge is the receive diff between it and my IMAX2000 (same height). Signals jump 5 to 10 S units on the quad.
I think I agree with both MARCONI and DINDIN on several points. There is a big argument out there about the orientation of the feed point on the SE Quads. However, I am very happy with the Quad set-up. |
Marconi
Intermediate Member
Username: Marconi
Post Number: 301
Registered: 11-2001
| | Posted on Monday, May 31, 2004 - 8:13 pm: |
|
CA346, if you want to check out their patent, you will see that there feeder is what makes the difference in the feed point as apposed to what we normally see in a quad feed point. Their polarity is 180 degrees different than the traditional and that has to do with their feeding at a high voltage point instead of a low voltage point.
If I can find the address I will post it soon. |
Marconi
Intermediate Member
Username: Marconi
Post Number: 302
Registered: 11-2001
| | Posted on Tuesday, June 01, 2004 - 8:33 am: |
|
Ca346 says; "I have the 'real' SE 4 element quad. (Lightning 4+) Their gamma match is a killer to get a flat SWR for me, but I like the dual polarity and a tuner eliminates the high swr on one problem side."
I have two friends with SE Quads and they both have the same problem. I understand from one bud that he had to make the vertical tuner connection at the 12:00, position instead of at 6:00, in order to get it to tune right on his 4Plus Lighting. I am not sure why, but that seemed to cure his problem.
Seems I recall I had a thought from some special note that SE made about the two polar sides of their dual polarity, common driven element AT, being physically connected. I was wondering why they would say that. Saying, "...that the opposite side (polarity) was transparent on operations." Hopefully this is worth noting to you guys with SE quad beams.
Maybe I read this in the SE patent notes. Not sure if this was discussed in the docs or not. Their docs are such a mess in reading anyway.
Think about it. You're feeding a single driven element here and the feed line, attached at the opposite polarity point may well be making the antenna, being tuned, appear to be much longer than it should. This would only happen during the process if you don't physically remove the other feeder from the loop. I say that, but I'm not sure if that would mean removing just the coax at the pigtail connection to the brass rods, or remove the tuner wire connected to the wire loop itself out at the end of the spreader.
You have to see this antenna as being direct feed. There is no gamma matcher on a SE Qaud beam. There is no inductance feeding here like with a gamma matcher. A gamma matcher physically isolates the opposite side affects in dual a polarity beam, because the two polar sides are not electrically connected, except via the ground and that presents no problem.
|
Bouk
New member
Username: Bouk
Post Number: 3
Registered: 5-2004
| | Posted on Tuesday, June 01, 2004 - 10:43 am: |
|
Ok, you all have good ideas about what could be wrong, maybe it could be the fusion SEquad/Gammamatch.
Could i use the same spreaders and the wire and feed it in some other way on another point wich seems to be used more often by quadbuilders?? If that is the case, where and how? I will soon have some pics on the feed for you to look at. |
Ca346
Advanced Member
Username: Ca346
Post Number: 801
Registered: 8-2003
| | Posted on Tuesday, June 01, 2004 - 6:51 pm: |
|
Marconi:
I must have talked with you on channel 38LSB? I don't know enough about the design of the SE Quad to be able to even discuss it intelligently, but I sure understand when it works and when it doesn't! Thanks for the info. The next time I climb my tower, I'm going to tweak the antenna adjustment on the horizontal side and really mess it up....
BOUK:
Best book I've seen on antennas is the ARRL Antenna Book ($30). It covers it all... A little pricey tho... |
Dindin
Intermediate Member
Username: Dindin
Post Number: 142
Registered: 2-2004
| | Posted on Wednesday, June 02, 2004 - 5:04 pm: |
|
Bouk,you should be able to use the hardware to construct a good quad with no problem.experimenting is half the challenge of this hobbie!the book Ca346 mentions is a wealth of excellent building info.shop around and you can find a bargain on one.I got mine off e-bay less than 10 dollars with software!Marconi mentioned a direct feed and that would be the simplest form to go with.I made a drawing Sunday evening to post but without zipping or converting to pdf it was to large to post.it shows both currant feed and voltage feed points.drop me an e-mail if your interested and I'll will fordward it to you that way. |
Marconi
Intermediate Member
Username: Marconi
Post Number: 303
Registered: 11-2001
| | Posted on Wednesday, June 02, 2004 - 7:27 pm: |
|
Bouk, if you go with a four element you may be able to get by with direct feed. You will just have to try it. Two elements will probably require a feeder to get the impedance down enough. A single element quad will work using direct feed even without a feeder. BUT!!!!!
Any way you go, and if you want to have a good multi-element horizontal you will need something to balance the line (balun) or you will have so much line radiation that the line itself will be about as hot an antenna as the quad elements and that will skew badly the pattern of the horizontal, making it appear as a vertical. As far as transmit it should work well either way.
Just read Scott's words I sent you. In looking back here, it looks like you already have read his words. He makes a lot of sense, Bouk |
Tech833
Moderator
Username: Tech833
Post Number: 723
Registered: 8-2002
| | Posted on Saturday, June 05, 2004 - 9:17 pm: |
|
Just saw the thread. I hope I can help.
Gamma feeding a Quad works great when used as Vertical. When fed for Horizontal, the lobe tends to be off to one side slightly (side opposite the gamma).
If you use too thin ga. wire, you will not just lose bandwidth, you will also lose power handling capability and also the element will act differently. Velocity factor caused by the signal spiraling around the wire is the reason for this.
Make sure that if you gamma feed your quad, you only use 1 polarity. You cannot attach 2 gammas for different polarizations and simply switch them on a true quad unless voltage fed like the Signal Engineering quads. Also, your gamma will still present some inductive reactance at the feedpoint, so you will need a capacitor in series between your coax center conductor and the bottom of the gamma. Somewhere between 2 and 20 pF should work. It depends on a lot of things, so I cannot give you an exact value. Use a tunable type.
Let us know how you do! |
Bouk
New member
Username: Bouk
Post Number: 4
Registered: 5-2004
| | Posted on Monday, June 07, 2004 - 4:01 am: |
|
Ok.
Then i will change wire to a thicker one, and use a tuneable capacitor between coaxcenter and the gammamatch. Question: How will i know when the capacitor is tuned right?? Will it affect the SWR or the signal??
|
Marconi
Intermediate Member
Username: Marconi
Post Number: 306
Registered: 11-2001
| | Posted on Monday, June 07, 2004 - 11:08 am: |
|
Bouk, Tech 833 may be right on the mark about a gamma working fine on a wire dipole, but I tend to disagree without having ever used this type of setup. You are using RG8/u I think to serve as the inductive cap in the gamma you now have acting against the wire element. That means to me that the center conductor of this piece of coax is likely serving as the tuning rod, so-to-speak, and is attached directly and feeds directly at some tap point on the wire loop. This is typically one of the tunable parts (rod) of a gamma. The other being the tap connector (dog bone) attached to the rod and the loop.
If I am right to this point, then I make my argument against a gamma in this case. It is my understanding that in order for a gamma to work properly, there must exist some limited range of sizes (usually given in a ratio) between these parts, the rod (center conductor of the coax in this case) and the driven element (the wire loop in this case) and the gamma length. Typically the ratio for the gamma rod size is on the order of element/rod with the rod being about .25% to .50% that of the element. The gamma should be on the order of .04 to .05 of a wavelength in length. Also there should be some other similar spacing ratio for the spacing between the gamma and the element itself. This measurement needs to be pretty percise and remain pretty consistent the entire length of the gamma. You may have the spacing problem right, but the element/rod difference is for sure in error unless you are using a much heavier wire on the loop around the area of the attached gamma. It that is the case, then how do you support all of this out there on the wire and how do you get out there to tune?
I may be wrong again, but I get the impression from your words that this gamma is out at the end of one of the spreaders where the loop attaches, and I believe you stated that that point is at 6:00 where you are hoping to be feeding a horizontal, right?
How are you coming with pictures. You never did tell us how many elements you are working with.
Continued good luck in your challenging work. |
Tech833
Moderator
Username: Tech833
Post Number: 731
Registered: 8-2002
| | Posted on Tuesday, June 08, 2004 - 5:12 pm: |
|
Bouk, Yes, it will affect the SWR. Mostly, you will notice the difference as increased bandwidth. In other words, your SWR will be lower at the ends of the band. The SWR at the center of the band will be affected too, but not as much as the edges.
Mike, I am sorry to have to tell you that you are incorrect. A gamma spacing is not critical at all. It is merely a carrier of current to the feedpoint. It can be at any angle or spacing. The coax is not a factor with a quad, only a yagi type.
Disagreeing without ever having used this type of setup (your words) did make me laugh a little though! That is like disagreeing that fire is hot without ever having a fire yourself. Innocent as you meant to be, thank you. |
Bullet
Intermediate Member
Username: Bullet
Post Number: 340
Registered: 7-2003
| | Posted on Thursday, June 17, 2004 - 6:19 am: |
|
hum...ive read so much ive forgot what the question was..lol
ive never have wanted to use the gamma match for feeding quads so i have nothing to add here except thier are other feeds that are simple to make and are more efficient for single polarity purposes.
im guessing this is for cb band so wire of 14ga will work well for this.
a simple way to do this is with a 1/4 electrical wavelength of 75 ohm coax as your matching transformer and add a few snapp on ferrite cores at feed point as your balun.
this will keep your signal from skewing to one direction and keep tvi down.
his spacings were same as a moonrakers
65/56/65 that should work ok if swr is high play with reflector and driven spacing to find a happy place.
some normal dia.
reflector= 1030/f
driven= 1005/f
directors= 900/f
youll find better performance with
reflector 1.1wave spacing .165~
driven 1.1 wave
directors .9 wave spacings .2 +
this provides great gain and a good front to back |
Marconi
Intermediate Member
Username: Marconi
Post Number: 311
Registered: 11-2001
| | Posted on Tuesday, June 22, 2004 - 10:44 am: |
|
Hey Bouk, how are you coming along with your antenna work? If your antenna plans came off of the Internet I sure would like to know where you got them using a gamma match setup. I am curious how this thing works and we haven't heard anything more from you in a while.
BTW, again how many elements do you have? |
Marconi
Intermediate Member
Username: Marconi
Post Number: 312
Registered: 11-2001
| | Posted on Tuesday, June 22, 2004 - 11:34 am: |
|
BTW Bouk, you say the following "...Ok, you all have good ideas about what could be wrong, maybe it could be the fusion SEquad/Gammamatch."
Where did you get this information from. I don't seem to find such on the web page that you reference above for the quad antenna described there. In fact Scott talks about using a balun at some insulated point on the driven element, namely at a point where the wire is split so as to be insulated in order to accomadate the coax connection.
If I understand the description of your setup correctly, you indicate that the shield is attached directly to the continous wire loop and that the center conductor is attached to the feed side of the gamma device. I am not saying this will not work. I am asking how did you know where on the wire loop to place this device in order to find a 50 ohm feed point?
It is my understanding that if you have 4 elements then the driven element should show approximately 50 ohms at several points where the spreaders and the driven element wire connect. If you are using a gamma, then it would seem the 50 ohm tap point on the wire loop could not be at this same point, because I see that point as where you have the shield attached. So the gamma connection (dogbone connection) to the wire has to be farther out on the wire somewhere towards another spreader.
Am I still confused as to my view of your setup? Wish I could draw you a picture of what I am thinking.
BTW 833, do you really think I have to have personal experience with something before I can possibly understand how it works or not. If that is so, then what does that tell us about study, understanding, or reasoning? I didn't think my remarks were that funny. I am trying to understand here, and of course I could be all wet. Seems Bouk has some similar questions also, but since this is his first try at this maybe he should quit. Is that really your contribution to this discussion? You seem to have a good handle on all of this, why don't you explain the questions we have on our minds about this gamma feed device, instead of telling us it is working fine? |
Bouk
New member
Username: Bouk
Post Number: 5
Registered: 5-2004
| | Posted on Friday, July 09, 2004 - 6:09 am: |
|
OK, it has been a while since i wrote here.
The idea of using a gammamatch to this quad is 100% homebrewed, the construction of the match is partially taken from internet and partially taken from a yagi-gammamatch that we had. As i wrote,this getīs us in control of adjusting the swr. But maybe the construction of using a gammamatch on this antenna just fix the swr and destroy the antennareception/transmitting. |
Marconi
Intermediate Member
Username: Marconi
Post Number: 321
Registered: 11-2001
| | Posted on Friday, July 09, 2004 - 9:48 am: |
|
Oh! now I see. What you meant was the fusion of ideas lead you to use the gamma.
I don't believe you can blend the ideas that SE uses with those ideas applied to Yagi design and the use of a gamma without confusion of ideas rather than the fusion of ideas, but I have been wrong before. I don't know if I can help, but I would like to understand what you are doing and I am feeling I am just missing something in your words.
833 suggest to us that he has some understanding of exactally what you are doing and he says the following; "Mike, I am sorry to have to tell you that you are incorrect. A gamma spacing is not critical at all. It is merely a carrier of current to the feedpoint. It can be at any angle or spacing. The coax is not a factor with a quad, only a yagi type." Well, I don't understand this either and I have asked him to explain a little more with no word. I am not arguing that he is wrong. I just don't understand what he means that coax is not a factor. I was in error in mentioning this, but my remarks about coax above were directed towards the SE Quad tuning only, because of their dual polarity feature. I don't think Bouk has dual polarity.
Bouk also states the following, "The swr is ok and around 1,4 using a gamma-match feed. We have a really good "beam function" and 5-9 signal drops when we rotate the antenna. Now to the big problem. We compared the signalstrength with a GPa antenna (halfwave , no radials, same hight) using a coaxswitch, and we almost began to cry when the GPa stick outperformed the 4 el beam with around 3 s-units. What can be wrong, we have check the solders and they seem to be ok, please help me this just isnīt the way it is supposed to work." This sounds like to me that maybe you don't have the beam pointed at the other station and thus the difference. How can you say on one hand that you get an S5-S9 signal drop on rotating the beam and on the other say the TX/TR is destroyed? What if you are simply horizontally polozarized while thinking that you are vertical? Do you think this might give you the results that you are seeing if this is true?
|
Bouk
New member
Username: Bouk
Post Number: 6
Registered: 5-2004
| | Posted on Sunday, July 11, 2004 - 4:44 am: |
|
quote: "This sounds like to me that maybe you don't have the beam pointed at the other station"
When the signal-meter is at the top peak when we rotate the antenna, i guess thats the most we can get out of the antenna (the direction against the other station is correct).
quote: "How can you say on one hand that you get an S5-S9 signal drop on rotating the beam and on the other say the TX/TR is destroyed?"
Ok i meant that the gain was destroyed, sorry. The directional function was great, but the signal gets weaker
quote: "What if you are simply horizontally polozarized while thinking that you are vertical"
Yes the possibility that we are going horizontally has also crossed my mind it could explain the massive signaldrop. Or as tech 833 wrote: "If you use too thin ga. wire, you will not just lose bandwidth, you will also lose power handling capability and also the element will act differently." This is also a possibility.
If it is so that we canīt solve this problem i am open to suggestions of other ways to feed the quad. |
Bullet
Intermediate Member
Username: Bullet
Post Number: 354
Registered: 7-2003
| | Posted on Friday, July 16, 2004 - 3:51 am: |
|
even if it was working witch you should be able to do, i wouldnt feed it with a gama.
thiers several good ways to feed a quad. the one i posted above is one of the easier more direct methods and works well with metal objects in near field such as tower ect.
if seriously interested i can post other options but |
Quickdraw
New member
Username: Quickdraw
Post Number: 2
Registered: 12-2004
| | Posted on Sunday, December 26, 2004 - 9:57 pm: |
|
read my post under signal engineering beams just under this one...... signal e. has a great antenna with a bad feed system.....they split a full wave element in half to create dual polarity.... take there feed system throw it the trash and feed this antenna through a 1:1 balun and choose horz. polarization.im in the process and will let you know the outcome.....thanks and see ya....R. |
|
