[Techno_Dweeb]

Regarding the slot antenna....
I don't know how the design found the feed point to be 1/3 of the way along the slot. I tried a feed point midway, but did find that the 1/3 position worked better. We would need Stephen Hawking to explain that one.

 

[ota_canuck]

I still have the antenna shown in post#8 mounted at my shop. It's still giving awsome performance. When I have some more time, I'll try moving the feed point down to the 1/3rd distance to see what happens.

 

[ota_canuck]

If I were transmitting with a 19.6 SWR , I'd probably blow something in the radio!

Two way radio SWR should be less than 1.0, by adjusting the physical antenna length to match the reflector/ground plane. How can you adjust the OTA antenna's SWR? I would think making the aperture longer if the ground plane is larger or making it shorter if the ground plane is smaller.

Would improving SWR necessarily improve reception gain figures?

So, with that though in mind I got my cutters out, I increased the length of the aperture of post#8 from 16" to 18" and I didn't notice any difference.

Any thoughts of what I could try?

 

[hkaye]

VSWR is caused by impedance mismatch between the antenna and your receiver. Perhaps moving the feed point as suggested will improve the VSWR.

 

[ota_canuck]

Yeah! That's the next step. I'll move the balun down 1" at a time until I'm about 1/3rd down on the aperture.

We are assuming that the raw feed point value is 300ohms. Maybe the 4:1 balun is not going to perfectly balance the impedance at any point. I believe that aperture antennas are normally geared to have 50ohm output for wifi and higher frequency use.

With some tweeking the 2.9-5.9 SWR may be improved somewhat on UHF, but VHF's SWR it way high. Though it still is doing a great job on VHF's RF7 & RF12 regardless of the high SWR.

 

[300ohm]

Would improving SWR necessarily improve reception gain figures?

Yes, high SWR affects digital reception way more than analog reception. Holl_ands has a long discussion of why some where in this forum.

We are assuming that the raw feed point value is 300ohms. Maybe the 4:1 balun is not going to perfectly balance the impedance at any point.

Impedance varies with frequency. The SWR is the impedance mismatch from 300 ohms (which is a setting in the 4nec2 program). The further away from 300 ohms, the higher the SWR, ie the Impedance mismatch. Because Impedance consists of real and imaginary resistances, it will take time to figure out the best impedance value for this antenna. Then if you wanted to, build a custom coax balun for it to connect to a 75ohm coax downlead. (Most ham coax is 50 ohms).

The SWR mismatch also effects Net Gain.
A quick way to look at the SWR figures to subtract from Raw Gain :

SWR     Mismatch Loss dBi
-----   -----------------
1.0     0
1.5     .18
2.0     .51
2.5     .88
3.0     1.25
3.5     1.60
4.0     1.94
4.5     2.25
5.0     2.55

Two way radio SWR should be less than 1.0

No, no, SWR is a ratio. 1 to 1 is the best you can get. If your meter had a zero setting, then that would have been 1:1, which would correspond to 1.0 in our context. A lot of older equipment used general purpose meters, whose scales would run from 0 to 10, 25, 50, 100 etc.

 

[300ohm]

I still have the antenna shown in post#8 mounted at my shop. It's still giving awsome performance. When I have some more time, I'll try moving the feed point down to the 1/3rd distance to see what happens.

Moving the feedpoint down to 5 inches above the bottom of the slot, the gain peak is 4.38 dbi 5.32 SWR at 488 mhz with a bumpy curve downward to 1.63 dbi 2.60 SWR at 698 mhz.

The vertical pattern (horiz is about the same) is modified to this :

So the peak gain of the antenna is 36 degrees upward, both front and back. That may not be a bad thing, depending on how the direction of the signal is coming in.

 

[loveota]

EDIT : Post #20 Should be " Director " NOT " Deflector "

 

[ota_canuck]

Does 4nec2 determine the actual feed point impedances or does 4nec2 assume that the antennas impedance is already designed at 300Ohms?

 

[300ohm]

Does 4nec2 determine the actual feed point impedances or does 4nec2 assume that the antennas impedance is already designed at 300Ohms?

You input the Characteristic Impedance as a global setting. It can be any figure so 4nec2 is fine for ham or wifi etc usage also. Since common TV cable is either 300 ohm twin lead or 75 ohm coax and the common TV baluns sold are 4:1, 300 ohm characteristic impedance is the figure to use for TV antennas. The 300 ohm standard was set in the 30's/early 40's. Many at that time wanted a 600 ohm standard for TV, as 300 ohm twin lead required higher tolerances to manufacture. 75 ohm twin lead was just impossible to make, heh.

Then if you wanted to, build a custom coax balun for it to connect to a 75ohm coax downlead.

It looks like a 6:1 coax balun (450 ohms to 75 ohms) is the best compromise for use with the antenna in post #8.
SWR is still high at 2.8 SWR at 470 mhz to 4.5 SWR at 698 mhz, but much lower than with a 4:1 balun.

 

[ota_canuck]

Here comes the silly questions.

Does the distance at the feed gap change the feed point impedance of the antenna? Would narrowing at the feed point improve the antenna's impedance?

It is my understanding that the entire balance of the element lengths in relation to the size of the ground plane/passive elements/reflector sets the impedance value. Is that correct?

This is a confusing animal because, in the case of a slot antenna, there is no ground plane/passive elements or reflector to tune the aperture to. It's all one piece with a hole in the middle [the hole is the driven element?] and it's all electrically bonded together with the feed points.

Hmmm! heh!

 

[hkaye]

When I was playing with nec simulations just about any dimensional change to the antenna changed the impedance - some dimensional changes were more sensitive than others.

As well since this is a broadband antenna the impedance will change with frequency as well. A smith chart is really handy to understand the characteristics of impedance changes with the different frequencies, this is what I look at while playing with the nec simulations.

Impedance match is not necessarily the holy grail as I've chased the perfect impedance match and ended up with an antenna with less gain than the one with an impedance mismatch even after factoring in the loss of the SWR. Or I've found a perfect match only to reduce the bandwidth of the antenna.

The slot antenna has two dimensions - the slot dimension and the outside dimension. These will affect the impedance. I guess only nec simulations will give you an idea of what exactly happens when these dimensions are changed.

------

The best analogy of moving the feedpoint in the slot to change the impedance would be to compare this to a gamma match and how that works.

------

I don't pretend to understand how all this works

 

[300ohm]

Basically yes. Changing anything or changing the relationship of any pieces relative to each other changes impedance.

Narrowing or widening the feed gap is a very common way of adjusting impedance.

 

[flcs3]

Some basic slot theory

The ideal slot theory is usually presented in terms of dipole theory, on the basis of the "Booker extension" of "Babinet's principle". The earliest reference in my notes is by Booker from 1946, "Slot aerials and their relations to complementary wire aerials."

Slots and (strip) dipoles are complementary by interchanging the electric and magnetic fields. To use the results for round dipoles for strip dipoles, use twice the diameter for the width of the strip. For horizontal polarization, the dipole will be horizontal while the slot will be vertical. For half-wavelength the impedance of the dipole is (72 +j42.5) Ohm, while the equivalent slot is (363 -j211) Ohm (i.e. the dipole is inductive longer than the resonant length, whereas the slot is capacitive). Wire dipoles have low impedance at odd half-wavelengths, while slots have low impedance at even half-wavelengths. It is possible to combine them for bandwidth improvement.

Moving the feedpoint off-center is a technique sometimes used for enhancing bandwidth. As usual, the theoretical results are based on simplifying assumptions, such as infinitely thin dipoles, sheets of infinite extent, perfect conductors, etc.

An interesting development is the use of a parasitic wire with a slot:
T. Morioka, S. Araki, K. Hirasawa
Slot antenna with parasitic element for dual band operation
Electronics Letters, vol. 33, no. 25, 4 Dec 1997, pp. 2093-2094.

Hopefully there are better web pages on the subject, but here is one:

http://www.antenna-theory.com/antennas/aperture/slot.php

HTH

 

[Techno_Dweeb]

Thanks for the insightful explanations. I did not think of the posibility of bandwidth changes due to feed points. I will have to try some more experiments with my test unit.

 

[ota_canuck]

Xauto rescaled the slot size to "18" by 1/8" attached balun to center" [see post #13.]

The top portion above your balun was 9" and the remaining 3" below the balun, so Xauto basically rescaled to the total slot length to acheive 9" above the balun and 9" below the balun, hence the avg. UHF frequency should be the same as your antenna.

Moving the balun down renders the longest portion of the slot as the being the determining factor of the frequency of the full slot [dipole equivelent] length. If you look at 300ohm's post #27, you'll see that moving the balun down below center seems to discard the bottom portion of the slot and redirects the vertical reception angle.

I would think that making the slot to the designated dipole length and then dividing it in half by having the balun centered would make more sense than moving the balun off-center.

I made one as you had described using a 4" dryer vent tube with a 17" x 3/4" slot and the [4:1] 75 Ohm balun is centered at 8.5". I'm not sure what the actual gain figure would be, but at 25 feet height I get most of my available UHF channels at 100%. The lowest at 14.1RF & highest at 50.1RF are very strong. This cylinder design absolutely seems to be 100% omni.

I believe nec2 is capable of analyzing cylinders, but not sure about slotted cylinders.

 

[FranKllr]

I believe nec2 is capable of analyzing cylinders, but not sure about slotted cylinders.

http://www.digitalhome.ca/forum/showpost.php?p=1026872&postcount=726

X-731 from bokakob is a slotted cylinder. Seems it works.

 

[flcs3]

I would think that making the slot to the designated dipole length and then dividing it in half by having the balun centered would make more sense than moving the balun off-center.

Fig. 7-20 (page label 92, pdf page 96) of this pdf shows some effects of moving the feed point.

http://www.cwc.tf.uni-kiel.de/lectures/antenna design/antenna-design_lecture_2009-04-23.pdf

I believe nec2 is capable of analyzing cylinders, but not sure about slotted cylinders.

These two messages from the NEC-list archives may be instructive:

http://www.devilsfoot.com/Nec/Archive1997/0064.html

http://www.devilsfoot.com/Nec/Archive1997/0069.html

The second is from Jerry Burke of LLNL.

 

[loveota]

ota canuck
Since I have a 50' roll of 1" x 2" mesh I gave this ant. a try . The dimensions I used was 20" x 28" , 14" x 4" slot , 45 deg. on the outer 2" , mounted 3" off the mast . No height tuning done , just a quick test . It pulled in VHF 12 and my only local digital 19 abit stronger than my SHGH setup . While the other channels were weaker it did pickup RF50 (54.1) . Not bad at all performance wise , unfortunately it doesn`t fit in with my line of "Alien builds" . See profile for pics. The SH on top of the SF isn`t active BTW. Impressive enough to do more testing another day .

 

[ota_canuck]

antenna bonded or non-bonded mounting?

Re-tests of the antenna that I described in post#6 & have shown in post#8.

Seeing that I believe that the screen portion of this design is a ground plane interacting with the aperture, I wondered if there was a need to isolate the mounting of this type of antenna from the mast.

I've done some testing with the idea of bonded and non-bonded mounting onto the grounded mast for this aperture style antenna. I have found no difference in performance, whether this antenna design is bonded or not bonded to the grounded mast. Mounting points are 2" from the top of the outer edge and 2" from the bottom of the outer edge. The feed points must be at least 1" away from the grounded mast.

I'm not sure if nec models can confirm the effect of bonding [or isolating] this antenna design to the grounded mast.


Two notes of interest from these bonding trials:

- plastic standoffs may not be required for mounting.

- the mast side [back side] of the antenna seems to become more directional than the front side that has the swept forward bends.