AVS Forum banner

21 - 35 of 35 Posts

·
Registered
Joined
·
3,856 Posts
How do you feel about changing spacing on the director's lengthwise down the Boom? I've got plenty of Boom stock.
I finally found a way to get the boom location measurements for the elements from the model. They should agree with the stock 91XG you have.

I used this NEC file from holl_ands:
A-D 91XG - RevF 4nec2 File
No Errors or Warnings. AGT=1.0.
UHF 91XG Corner-FD-Yagi

which is on this page:
UHF 91XG Corner-FD-Yagi

I downloaded the NEC file, opened it in Notepad and pasted it into nikiml's NEC Viewer.

nikimil's Home Page:
nikiml's Antenna pages

91XG nikiml Home Page_1.jpg


nikiml's NEC Viewer:
nikiml's Antenna pages - Nec Viewer

91XG nikiml NEC Viewer_1.jpg


91XG nikiml viewer 3D_1.jpg


91XG nikiml viewer side view_1.jpg


91XG nikiml viewer side mm_1.jpg


91XG nikiml viewer side inches_1.jpg


If you decide to increase the element spacing, I calculated the new element locations:

91XG Element Locations 100pct2_1rev.jpg


but I suggest you try the extended directors with the original spacing first.

AD 91XG Balun_3.jpg


In the model
UHF 91XG - Wider Bowties, Better Ch14-51
UHF 91XG - Wider Bowties, Better Ch14-51
holl_ands said he shortened the director strip attached to the boom in front of the Driven Element:
To adjust SWR to acceptable levels, it ALSO was necessary to SHORTEN the FIRST Director-Rod[which primarily controls SWR] from an Overall Width of 6.28-inches down to just 6.0-inches (+/- 0.1-in).
but I don't know how important that is.

91XG Director Strip Front Edge_1.jpg


91XG Director Strip Rear Edge_1.jpg
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #22 · (Edited)
Wow, Tripleo and Rabbit I've got to thank you for your very, very informative posts/ comments. I thought I was going to be doing this on my own but I've got some support.

So for now I will modify my X elements the way I've been doing. I'll leave them in the stock locations and test what I've created against another 91XG.

And it looks like I would need to reduce the size of the first element.

I probably missed it but did he change anything with the driven element?
3122924


Driven element, any suggestions on that?

3122925
 

·
Registered
Joined
·
3,856 Posts
I thought I was going to be doing this on my own but I've got some support.
Well, you finally got me interested in the project. There certainly is a need for a high-gain antenna for the new 14-36 UHF band.
So for now I will modify my X elements the way I've been doing. I'll leave them in the stock locations and test what I've created against another 91XG.
Good; just change one thing at a time.
And it looks like I would need to reduce the size of the first element.
Yes, you should also try that. The model done by holl_ands says that reducing the length of the director strip in front of the Driven Element from 6.28" to 6.0" reduces the SWR, which means you will have more Net Gain.
I probably missed it but did he change anything with the driven element?
I don't think he did. He was looking for the easiest way to find more gain for the lower UHF channels.

Thank you for the photos of the Driven Element.

91XG DE by ChiwaukeeOTA_1.jpg


91XG DE by ChiwaukeeOTA2_1.jpg


On the long Canadian thread, tballister also wanted to increase the gain on the lower channels. He modified the DE, but there was no improvement. He did open up the balun and found out that the DE was not a folded dipole, just a dipole with a loop on each side. The total length of the DE is 15.75", making it a fullwave dipole.
Antennas Direct 91XG Hardware Hacks

91XG DE tballester_1.jpg


91XG DE Detail by tballister_1.jpg


tballister's website:
Antenna Comparisons, Antenna Reviews, DIY Antenna Modifications, HDTV Antennas

91XG nikiml viewer DE3_1.jpg


Driven element, any suggestions on that?
The only DE modification I would consider is a 3/4 wave folded dipole, which was called a Tetrapole in a Patent by John Winegard:

Winegard UHF Tetrapole2.jpg


The Tetrapole has two advantages: it has two current loops, instead of just one as in an ordinary 1/2 wave dipole or 1/2 wave folded dipole, and it can easily combine with a VHF antenna connected at the top of the Tetrapole. The two current loops match the dual directors at each director location of the 91XG antenna.

0.75 wave folded dipole2_1.jpg


You can see a Tetrapole on the original RCA ANT751 and on Winegard and Antennacraft antennas.

RCA ANT751 Rear View2_1.jpg


RCA ANT751 from below_1.jpg


But a fullwave dipole also has two current loops (peaks), so you should stick with the simplicity of the original 91XG DE; there is no need for the VHF combining ability of the Tetrapole.

fullwave dipole current distribution_1.jpg


The length of the fullwave dipole doesn't seem to be as critical as the lemgth of the directors, as tballister found out when he increased the length of the DE.
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #24 · (Edited)
Very good Rabbit !

I'm taking this project in baby steps. As you know, maybe we haven't settled on the exact length of the directors yet but I did start the groundwork. 121% ? Okay how long should my director extensions be ? I can do the math later.

Also for anybody who's reading this I am not destroying two perfectly good 91XGs. They were damaged in a storm Beyond repair. I don't even have enough X elements in plastic brackets to do what I have to do to so I'll have to steal some from a brand-new unassembled 91XG kit.

3123987


I already have my son involved. I'm teaching him life skills, how to work with his hands. He was having a little hard time sniffing the tips of the X elements so I had him filing the burrs.

3123990


3123994





I went bought a brand new pair of snips because my old Snips were rolling the edges too much. These are much more like cutting paper with scissors.
3123992



As it turns out made 89 tips, and my son made the observation that's good because we'll probably wreck one or two extending those directors!

3123993


I also bought a bag of 1000 3/32 aluminum rivets. I'll need at least 200 of them I'm sure ;) .

So I'm going to wait I'm making the directors until we settle on the size. I would like the antenna to Peak at 608 mHz. I wouldn't care so much but RF 36 in Milwaukee is fairly weak for me and I'm not sure why.

 

Attachments

·
Registered
Joined
·
3,856 Posts
I'm taking this project in baby steps. As you know, maybe we haven't settled on the exact length of the directors yet but I did start the groundwork. 121% ? Okay how long should my director extensions be ? I can do the math later.
Yurii and majortom have been helping me. Majortom did this rescale at 117%. It looks good; peaks at 608 and low SWR. That's pretty close to my original estimate of 118%.

91XG majortom at 117percent_1.jpg


He then did a rescale at 133%:

91XG majortom at 133percent_1.jpg


That doesn't look as good. The peak is below 608 and the SWR is too high. Jeff Kitz was too high. The max is at 121%, but that doesn't allow room for any errors, so I'm happy with 117%.

As far as actual size goes, the extension calculated by holl_ands of 7.65 inches from the center line is as good as we can calculate now. Maybe majortom has a better idea.

I have ordered a nanoVNA to try to measure the resonant freq of a stock director. Then maybe I can calculate how long the extension should be for a MAX at 608.
So I'm going to wait I'm making the directors until we settle on the size. I would like the antenna to Peak at 608 mHz. I wouldn't care so much but RF 36 in Milwaukee is fairly weak for me and I'm not sure why.
WMKE is only running 15 kW and its directional transmitting antenna is aimed west. You are out of its area of coverage and in a dead zone (no color):

ChiwaukeeOTAavsCovWMKE_1.jpg


and the direct signal is blocked by terrain and curvature of the earth:

ChiwaukeeOTAavsP3WMKE_1.jpg
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #26 · (Edited)
Thank you rabbit I did not think that 49 miles would put RF36 over the horizon.. I thought it was LOL like the rest of the Milwaukee stations. Apparently that's not the case though according to your diagram. It's funny, I used to pick it up pre-repack. 36 was one of the only channels that had 8 sub channels.

I think now, they re-broadcasted on RF10 and Consolidated some of the channels. It is PBS and they were notorious for taking the money and running on the reverse auction.
 

·
Registered
Joined
·
3,856 Posts
Thank you rabbit I did not think that 49 miles would put RF36 over the horizon.. I thought it was LOL like the rest of the Milwaukee stations. Apparently that's NOT the case though according to your diagram. It's funny, I used to pick it up pre-repack.

RF36 South..
If you only give the RF channel number and don't include the callsign, it can create a problem. In this case it did create a problem. A new report for your location shows WMKE in Milwaukee and WQRF in Rockford on RF channel 36:
RabbitEars.Info

Your link shows WRJK scheduled to move to RF channel 36, but the new report for your location shows WRJK still on RF channel 11.

WRJK has filed an application to move to RF channel 36, but it is listed as Pending, not Approved.

ChiwaukeeOTAavsWRJK ch36 Pending.jpg


Trip in VA might know what RF channel it is actually on now.

In the future, please give the RF channel number AND the callsign. I would rather spend my time on your antenna problem than chasing the wrong callsign.
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #29 · (Edited)
If you only give the RF channel number and don't include the callsign, it can create a problem. In this case it did create a problem. A new report for your location shows WMKE in Milwaukee and WQRF in Rockford on RF channel 36:
RabbitEars.Info

Trip in VA might know what RF channel it is actually on now.

In the future, please give the RF channel number AND the callsign. I would rather spend my time on your antenna problem than chasing the wrong callsign.
Hey Rabbit. I'm sorry about that. I was just merely stating I was hoping to receive RF 36 again. I didn't know how important it was to give the call sign. Anyhow, I don't care if I get RF36 or not because most of it is repeat of what's already available on RF8 which is virtual Channel 10.1 PBS.

Again, sorry about that.
 

·
Registered
Joined
·
3,856 Posts
So, with the 117% exactly how long should my extensions be?
That's still not yet easy to answer. I'll explain further down.

I was able to get the location of the elements on the boom for the 117% rescaled version. Majortom didn't post the NEC file, but he did post a link to the nikiml NEC viewer, so I was able to copy 50 measurements (mm and inches) for the new version:
nikiml's Antenna pages - Geometry viewer,

91XG DE location 117pct rescaled version_1.jpg


91XG Element Locations 3_1.jpg


This is not going to be easy to explain, but I will try. My goal is to find a way to give you an answer as accurate as I am able to do. I was going to do this on the Canadian forum, but since you asked, I'll do it here too.

The reason why it is difficult to give you an exact answer to how much longer the extensions should be is because adding extensions is not exactly the same as what happens to the model when it is rescaled to 117%.

When the model is rescaled, everything is made larger just like what the Xerox machine does when you set it to 117%. The arms are made longer but they are made wider, and the "V" is also made taller and wider. And, the distance from the center line to the bottom of the "V" is also increased.

You are just adding extensions to a director that has not been made larger as it is in the model. So, we know you must add extensions to make the director operate at a lower frequency AND compensate for the fact that the director is still its original size.

I'm a ham, so I'm used to designing a Yagi antenna for the center of the band that I will be using it for. That center frequency is the design frequency for the DE. The Director will be about 5% higher in frequency and the Reflector about 5% lower in frequency. In your case, the corner Reflector is non-resonant, so we don't worry about that.

14-69 Band
470 to 800 MHz, midpoint = 635 MHz

14-36 Band
470 to 602 MHz, midpoint = 536 MHz
635/536 = 1.18
The dimensions for the change from the 14-69 band design to the 14-36 band would have to be increased 18%.

The other method is to use the MAX gain frequency for the design frequency since you want channel 36 to have the highest gain:

holl_ands said
RevF: UHF Raw Gain = 11.3 dbi (470 MHz) to 16.6 dBi (698 MHz) to 17.5 dbi (758 MHz), F/R & F/B Ratio Min = 19.8 dB (Excellent) and SWR (300-ohms) Under 2.3 (Excellent).
To Widen the Bowties in the 4nec2 Model, THREE new Wires were added to each of the four tips on ALL 22 Bowties. Outer Width-to-Centerline parameter "Ymax = 3*Zmax = 3*2.55 = 7.65" (+/- 0.05-in) was adjusted to find the "best" combination of Raw Gain, F/B & F/R Ratios and SWR, which resulted in Max Gain falling on 626 MHz.
758/626 = 1.21

So, 121% is as high as you would want to go. Since that doesn't allow for errors, I'm comfortable with 117 to 119%, and the rescale at 117% that majortom did looks good to me.

The only way I can think of is to measure the resonant frequency of the stock director for the original design and scale that frequency to the 14-36 band. Would you be willing to send a stock director and a modified director to me so that I can measure them? If there is a more accurate way to do it, I would like to know about it.

My plan is to use a nanoVNA like a dip meter to measure the resonant frequency of a director. I used a dip meter to measure the elements of a 14 MHz beam. I had to replace the coil with a triangular loop that coupled to the center of the element as suggested by G6XN in his book HF Antennas for All Locations.

The NanoVNA, a beginners guide to the Vector Network Analyzer
Dip Meter at 43:05

Millen Grid Dip Meter_1.jpg


Millen Grid Dip Meter at dip3_1.jpg


The Millen Grid Dip Meter doesn't go high enough in frequency for this job.
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #31 ·
Rabbit, If you haven't ordered a nano VNA I have one and I would send it to you tomorrow along with a director and a modified director. PM me your address
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #32 ·
Rabbit, thanks again!

For anyone else follwing this thread, These pics are shared on the Canadian forum also, as some have taken interest there.

So after deciding to ditch the 133% increase I decided to go with the one inch extensions on my X elements:

11980



At the top is the 133% increase. Next one down is the 1" extension and below that is the standard stock 91XG X element. I'm hoping the Nano VNA will help with the tuning.

All of the above items plus the 91XG active element is on its way to you also. The Nano VNA should stay with you for fun and research.

Thanks again!
 

·
Registered
Joined
·
155 Posts
Discussion Starter · #33 ·
I agree that the just adding extensions doesn't change the scale of the shape as you increase the percentages.

11981



11983



11982


I'm currently playing around with splitting the bottom of the X element director by splitting at the bottom to spread the wings, Bad idea :


11985







Attachments
 

·
Registered
Joined
·
3,856 Posts
I agree that the just adding extensions doesn't change the scale of the shape as you increase the percentages.

I'm currently playing around with splitting the bottom of the X element director by splitting at the bottom to spread the wings, Bad idea:
It's an interesting idea, but an unnecessary complication that weakens the element. Let the increased length compensate for that.

The center of the element is where you have the greatest current, just as in a dipole. A crack there isn't desirable.

half wave dipole_1.jpg








Attachments
  • 20210421_095126.jpg
 

·
Registered
Joined
·
153 Posts
Rabbit, If you haven't ordered a nano VNA I have one and I would send it to you tomorrow along with a director and a modified director. PM me your address
I am about to order one from Nooelec.
From everything I have read they are the most reputable and support the originator and the software programmers.
 
21 - 35 of 35 Posts
Top