"Top Hat" GH with NARODs for VHF-HI: Major GH Improvement

[gcd0865]

Karyudo:

One set of dimensions is found in the .pdf file in post #92 (at page 7) of the following thread:
http://www.digitalhome.ca/forum/showthread.php?t=99907

This design appears to be among the best in excellent gain at reasonable size.

The rest of that thread has many helpful designs and tips for a DBGH with NARODs. The newer "top-hat" version of NAROD design appears to be helping out as well, although I'm not sure if the top-hat design and measurements have been optimized for this particular version of DBGH (others can chime in on that; I think it's being optimized for other GH versions right now).

[nikiml]

Karyudo,

Those models are mods of the JED's antennas.
You can just use his dimensions
and I have posted dimensions for the additional elements for two of the three models
in the thread.
Check out the 3d view to get an idea of how the dimensions are to be applied.

[Autofils]

=================
Note to Stampeder: This discussion should really be moved to the Software 4nec2 modeler's thread, but I leave that to your discretion
=================

300Ohm,

In reference to your post #39 http://www.digitalhome.ca/forum/show...2&postcount=39

Autofils:
I can tell you that I find a difference of a couple of dB, very surprising for a model with the min of 10segs per half wavelength vs say 22 segs per half-wavelength.

300Ohm:
OK, 1 1/4 dbi, I exaggerated a bit,heh.

A couple of comments on the plots by Snowman53:

1. The plots of Gain vs auto-seg per 1/2 wave for SBGH, show a significant drop in gain over a narrow range from 10 to 13 segments per1/2 wave.
Above 13, the gain results are reasonably constant (ie good convergence). In the case of DBGH plots, I wonder if those models were collinear or mesh screen or both? There was no discussion in the thread that I could find, that discussed how auto-segmentation actually calculates the segments for each element and what that implies for larger models.

As all modelers who post results on this forum should know, the minimum segmentation recommended by L. B. Cebik in his "Beginner's Guide to NEC Modeling" is 10 segments per 1/2 wave at the design frequency of the antenna. So I ignore the Snowman53's plot results for < 10 segments per 1/2 wave.

2. As I mentioned previously, my experience with convergence tests on manually segmented models at 10 -20 segments per 1/2 wave, have never shown anything close to a 1 1/4dB difference.

I suspect that Snowman53's auto-segmentation results might actually be characterizing the auto-segmentation algorithm, rather than antenna model differences at the 10 -13 auto-seg range. It is unfortunate, that a manual segmented model was not compared with an auto-seg model, since that would answer my suspicion. However your GH6.3Narod_with_Screen model presents an opportunity for any modeler to investigate that difference !

I propose this Modeling Challenge:
Take the GH6.3-Narod_with Screen model by 300Ohm in post #31 of this thread and make a manual segment version using the design freq of 584mHz with 10 segments per 1/2 wave. You can use a smaller segmentation density for modeling the mesh screen reflector. If you are really keen, you could show the effect of varying the segmentation of the mesh screen from a very small density up to 10 segments per 1/2 wave.

(1) Run a sweep 461 to 707; step 12 and compare the gains. What is the max difference between this manual segmented model vs the auto-segmentation 22 at 800Mhz model?



A final thought:
In his post summary, Snowman53 stated:

Originally Posted by Snowman53

In summary I think NEC is OK for comparing variations of the same antenna only if a consistent segment sizes are used.

I am not sure one can trust NEC to provide consistent values between different types of antennas which may use different segment sizes.
I would welcome a discourse on this topic with the resident modeling wizards.

Reference post: http://www.digitalhome.ca/forum/show...&postcount=422

"I am not sure one can trust NEC to provide consistent values between different types of antennas which may use different segment sizes."

Now if the segmentation is not done correctly due to modeler's in-experience, or other cause, this statement is true, but if the models follow all the the segmentation rules correctly, I don't believe this statement to be true. If it was, it would mean that all the excellent work by Ken Nist comparing commercial antennas would be in doubt as well.

I think the trust that Snowman53 is unclear about concerning different types of antennas needs an answer and I totally agree with his statement "I would welcome a discourse on this topic with the resident modeling wizards."

Perhaps the "resident modeling wizards" will seriously take a look at my modeling challenge.
I am reminded of that old expression .... "You can lead a horse to water, but you can't make him drink" The challenge has been presented .... any takers?

Cheers

[ Oh,Oh .... I hear Zapperman calling me...so gotta go and get back to work ....Cya later]

[300ohm]

I propose this Modeling Challenge:
Take the GH6.3-Narod_with Screen model by 300Ohm in post #31 of this thread and make a manual segment version using the design freq of 584mHz with 10 segments per 1/2 wave. You can use a smaller segmentation density for modeling the mesh screen reflector. If you are really keen, you could show the effect of varying the segmentation of the mesh screen from a very small density up to 10 segments per 1/2 wave.

I think its best to use Snowman53s original NEC file (Area 51 post 847) instead of the failed mesh experiment above. Mesh and manual segmentation dont mix well, heh. And we want to keep the graphs and data equal.
Snowman53 did present a good argrument.

I also think its best to get more modelers into this discussion , rather than just you and me alone. holl_ands will probably be back, and maybe Snowman53 too. mclapp, since he's very good with EZNEC can provide some insight too. And maybe we can get some more Mana_gal modelers into the discussion. And maybe Arie Voors too. Hes posted on this website in your absence, and likes us too. So has Dr. Natan Cohen, the Fractal Guy, the originator of the fractal antenna.

In the meanwhile, its not as important as you finishing your website for example. For now, Ill stick with the convienent and safe auto-segmentation of 21 or 22 except for antennas of very high segmentation, like the 91XG or CM4251, DBGH Mesh or Quad or modeling vhf oddballs like stampeders CM1111 or my CM1221, heh.

[Autofils]

With reference to 300Ohm's post #49 -> http://www.digitalhome.ca/forum/show...5&postcount=49.

Originally Posted by 300Ohm

... its best to get more modelers into this discussion , rather than just you and me alone. holl_ands will probably be back, and maybe Snowman53 too. mclapp, since he's very good with EZNEC can provide some insight too.

That would be great. I did mention "Perhaps the "resident modeling wizards" will seriously take a look at my modeling challenge."

I would suggest that you enlarge the scope of the discussion to think about collectively generating a more detailed modeler's guideline that Stampeder could make into a "sticky"

For me, there are really two important questions that snowman53's auto-segmentation analysis raises. I had mentioned them in my previous posts on this subject, but I guess I wasn't that clear. So let me have another go and provide some points that might start the discussion with a wider viewpoint.

Question 1. How to determine an antenna model's "reliability" for posted results of Gain/Swr etc

The late L. B. Cebik (W4RNL) pioneered the use of MoM nec programs in the ARRL as a better way to compare commercial antennas for Radio Hams, and to provide a software tool to better design "Home-Brew" Radio Ham antennas covering a wide range from 80m up to "Moon-Bounce" frequencies.

His recommendation for determining a model's reliability is to run two tests: not only AGT at at least one frequency (within the designed bandwidth), BUT one should also run Convergence Tests, to determine how increased segmentation affects the results of a specific nec model.
The value of these two tests is simply this: It is applied to a specific antenna model, hence comparisons of models of different antennas that have been found to be "reliable" via these two tests, provides the best and only assurance that the comparison is valid.
The AGT test should be done at at least one frequency inside the designed bandwidth and Convergence tests should be done from the minimum recommended density of 10 segments per 1/2 wave up to the max limit of segmentation that meets the segmentation rules.

In 4nec2, Arie has developed a Convergence test that does exactly that. It is the "Conv-test" function in the Calculate->Start optimizer (F12) tab of the main window of 4nec2. I would recommend that you ensure your Auto-segmentation setting is disabled before running the Convergence and AGT tests. If you have never run the Conv-test function in 4nec2, take a close look at it. It is an extremely powerful tool that Arie has built into 4nec2 !

I really believe that running both tests will provide the answer to Snowman53's query "I am not sure one can trust NEC to provide consistent values between different types of antennas which may use different segment sizes."



Question 2. Why is Auto-segmentation used ? Is it ever Disabled ? ...and other questions

Auto-segmenation is a useful computer aid for 4nec2 modelers, especially when the model is of an antenna with complex geometry. Snowman53's analysis shows that setting auto-seg to >13 gives good convergence on the specific SBGH and DBGH models that he considered in his analysis. It does not necessarily imply that all models will be reliable if you use auto-seg at a setting >13 (say 21 or 22). Instead, you should test each specific model using AGT and Covergence tests. That will give a measure of the model's reliability; using auto-seg does not give such assurances.

My modeling challenge was geared to explore what the auto-seg algorithm in 4nec2 actually does for complex antenna geometries, including a mesh screen. I was not familiar with the details of the snowman53 models, hence my suggestion for the GH6.3-narod with Screen. Feel free to explore with any other model.

Personally I don't use auto-seg. From what I understand, it is very useful when first setting up your model segmentation and should be used when running sweep optimizations. I suspect that most other times your should disable auto-segmentation. There might be some other strange effects that others have noticed, using auto-seg. You probably have the most experience of anyone on this forum with mesh-screen models; would you recommend using auto-seg with mesh screens? From Arie's help file, it's interesting to note that you can selectively turn-off auto-segs for specific wire elements in your model via a preceding + <sign>. This would allow the modeler some control over the auto-seg algorithm.

I see more and more new comers to 4nec2 modeling on this forum. I know from my journey that there is lots to learn and even more to continue to learn - for all of us. I really believe a more detailed 4nec2 guideline "sticky" that provides some explanation on these subjects would be very useful and would further grow the reputation for DigitalHome.ca as the world's best modeling forum.

Anyway...time to go... I hope folks find this helpful in better defining my thoughts. I'm way over my allotted time... you know ...Zapperman is calling ...so it's back to "kompozer" and more html stuff.


4NEC2 Auto-segmentation Help Notes

Auto-segmentation

When optimizing and/or sweeping antenna performance, it is likely thatdimensions and/or frequencies are changing, so it is possible that certain NEC requirements, such as the minimum number of segments per half-wave are not met any more. To overcome this problem, Auto-Segmentation is used. 

If this feature is enabled (See Settingssettings on Main window), each calculation/optimization/sweeping step, the wire length for each wire is checked against the actual wave-length and the specified number of segments per wire, and the position for voltage/current sources, RLC-loading and transmission lines is updated.  

 If a certain number of segments is already specified in the input file, at least this number of segments is set with auto-segmentation. This to avoid auto-segmentation decrease the number of segments specified for critical regions.  If you do not want to specify a minimum, set the number of segments to one.

If the number of segments is preceded by a “+” sign the number of segments is fixed, meaning, the segments for this wire are not affected by the auto- segmentation process.

 When enabling the Auto-Segmentation feature, you are asked to specify the number of segments per half-wave. Although, mostly accuracy improves if more segments are specified, also the computation time is increased. For ‘simple’ structures, without sharp angles, a value from 10 to 20 will be sufficient. If precise impedance values are required or if wires joining at sharp angles are used more segments could be required. If you are not sure what amount to specify, please use increasing values and compare the results. (You could do this using the Sweepereval function). Also abrupt current changes on the structure might indicate an insufficient number of segments. One place more segments aren’t better is if wires of different diameters are connected.

If computer performance is an issue, you could consider not to the use the auto-segmentation feature. If this feature is disabled, the processing especially for large input files with many wires is speed-up.

[nikiml]

A slightly conservative mod. This is because I fixed the horizontal legs length to be the same and I fixed the spacing between the active element and the narods to be 12mm. I believe this model won't produce the unequal segmentation warning.

Nec code:

CE
SY radius=0.003175
SY hat_rad=0.003175	'0.0010265
SY l1=0.300
SY g1=0.018
SY z1=0.128
SY l2=0.312
SY g2=0.045
SY z2=0.444
SY x=-0.084
SY a1=0.196
SY a2=0.202
SY a3=0.171
SY b=0.122
SY feed=0.075
SY a1ma2=a1-a2
SY a1pa2=a1+a2
SY a1ma2pa3=a1ma2+a3
SY a1pa2pa3=a1pa2+a3
SY n_top_l=0.20747
SY n_b_l=0.122
SY n_z_sp=0.0105
SY n_h=0.053322
SY n_b_z=a1pa2pa3*.7071+n_z_sp+hat_rad-radius
SY n_top_z=n_b_z+n_h
SY n_b_ystart=-n_top_l	'-feed-0.70711*a1ma2pa3-n_b_ysp
SY n_b_yend=n_b_ystart-n_b_l
SY n_back=-0.3034
SY n_refl_size=0.424505
SY n_refl_z=0.459371
SY n_refl_size1=0.475897
GW	1	23	0	-feed	0	0	-feed-0.70711*a1	0.70711*a1	radius
GW	2	25	0	-feed-0.70711*a1	0.70711*a1	0	-feed-0.70711*a1ma2	0.70711*a1pa2	radius
GW	3	21	0	-feed-0.70711*a1ma2	0.70711*a1pa2	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	radius
GW	4	15	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	0	-feed-0.70711*a1ma2pa3-b	0.70711*a1pa2pa3	radius
GW	5	35	x	g1	z1	x	l1+g1	z1	radius
GW	6	37	x	g2	z2	x	l2+g2	z2	radius
GW	11	7	0	-n_top_l	n_top_z	0	n_b_ystart	n_b_z	hat_rad
GW	14	15	0	n_b_ystart	n_b_z	0	n_b_yend	n_b_z	hat_rad
GX	20	010
GW	10	49	0	n_top_l	n_top_z	0	-n_top_l	n_top_z	hat_rad
GW	15	101	n_back	-n_refl_size	n_refl_z	n_back	n_refl_size	n_refl_z	radius
GX	40	001
GW	100	17	0	-feed	0	0	feed	0	radius*.8675
GW	16	113	n_back	-n_refl_size1	0	n_back	n_refl_size1	0	radius
GE	0
LD	5	0	0	0	24900000
GN	-1
EK
EX	0	100	9	0	1			
FR	0	0	0	0	800	0
RP	0	60	73	1001	0	0	3	5		
EN

Numeric results:

Ch.	Freq	RawGain	R-in 	X-in 	SWR	NetGain
	171	9.69	428.61	-240.2	2.1	9.1
	174	9.61	415.59	-123.24	1.61	9.37
7	177	9.5	417.9	-57.03	1.44	9.35
	180	9.38	423.34	-18.55	1.42	9.25
8	183	9.27	427.45	4.69	1.43	9.13
	186	9.17	429.21	20.15	1.44	9.03
9	189	9.08	429.2	32.33	1.45	8.93
	192	9	428.61	43.71	1.46	8.85
10	195	8.92	428.7	55.54	1.47	8.76
	198	8.86	430.59	68.28	1.5	8.68
11	201	8.8	435.29	81.99	1.54	8.6
	204	8.75	443.7	96.48	1.6	8.51
12	207	8.7	456.79	111.34	1.67	8.42
	210	8.66	475.69	125.91	1.76	8.32
13	213	8.63	501.76	139.13	1.86	8.22
	216	8.6	536.72	149.19	1.98	8.1
	219	8.57	582.52	152.98	2.12	7.97
						
	467	13.15	204.81	-65.08	1.59	12.92
	470	13.24	222.9	-46.48	1.41	13.11
14	473	13.29	241.94	-34.18	1.28	13.22
	476	13.31	261.25	-26.23	1.18	13.28
15	479	13.33	279.84	-22.58	1.11	13.32
	482	13.33	296.69	-22.82	1.08	13.32
16	485	13.32	310.97	-26.21	1.1	13.31
	488	13.31	322.14	-31.78	1.13	13.29
17	491	13.29	330.03	-38.46	1.17	13.26
	494	13.26	334.83	-45.28	1.2	13.22
18	497	13.24	337	-51.48	1.22	13.2
	500	13.21	337.1	-56.54	1.24	13.16
19	503	13.18	335.75	-60.18	1.25	13.13
	506	13.16	333.51	-62.32	1.25	13.11
20	509	13.14	330.85	-63.03	1.25	13.09
	512	13.12	328.13	-62.44	1.24	13.07
21	515	13.11	325.63	-60.74	1.23	13.06
	518	13.1	323.53	-58.13	1.22	13.06
22	521	13.1	321.96	-54.8	1.21	13.06
	524	13.1	320.99	-50.92	1.19	13.07
23	527	13.11	320.66	-46.65	1.18	13.08
	530	13.11	320.98	-42.13	1.16	13.09
24	533	13.12	321.95	-37.48	1.15	13.1
	536	13.14	323.55	-32.79	1.14	13.12
25	539	13.15	325.76	-28.16	1.13	13.13
	542	13.16	328.54	-23.66	1.13	13.14
26	545	13.18	331.88	-19.37	1.13	13.16
	548	13.19	335.73	-15.35	1.13	13.17
27	551	13.2	340.04	-11.64	1.14	13.18
	554	13.22	344.78	-8.3	1.15	13.2
28	557	13.23	349.91	-5.36	1.17	13.2
	560	13.24	355.37	-2.87	1.18	13.21
29	563	13.25	361.11	-0.85	1.2	13.21
	566	13.25	367.09	0.68	1.22	13.21
30	569	13.26	373.26	1.71	1.24	13.21
	572	13.26	379.58	2.23	1.27	13.2
31	575	13.26	386	2.24	1.29	13.19
	578	13.26	392.49	1.74	1.31	13.18
32	581	13.25	399.03	0.73	1.33	13.16
	584	13.25	405.59	-0.79	1.35	13.15
33	587	13.24	412.17	-2.83	1.37	13.13
	590	13.24	418.75	-5.4	1.4	13.12
34	593	13.23	425.32	-8.55	1.42	13.1
	596	13.22	431.86	-12.31	1.44	13.08
35	599	13.22	438.34	-16.74	1.47	13.06
	602	13.22	444.69	-21.93	1.49	13.05
36	605	13.22	450.82	-27.93	1.51	13.04
	608	13.22	456.61	-34.81	1.54	13.02
37	611	13.24	461.88	-42.6	1.56	13.03
	614	13.25	466.48	-51.28	1.59	13.02
38	617	13.27	470.21	-60.81	1.61	13.03
	620	13.3	472.91	-71.08	1.63	13.04
39	623	13.33	474.39	-81.96	1.66	13.06
	626	13.36	474.56	-93.27	1.68	13.07
40	629	13.4	473.3	-104.83	1.7	13.1
	632	13.43	470.56	-116.42	1.72	13.11
41	635	13.47	466.32	-127.85	1.74	13.14
	638	13.5	460.58	-138.89	1.76	13.16
42	641	13.54	453.38	-149.33	1.77	13.19
	644	13.58	444.8	-158.99	1.79	13.22
43	647	13.61	434.93	-167.66	1.8	13.24
	650	13.65	423.9	-175.19	1.81	13.27
44	653	13.69	411.84	-181.43	1.82	13.31
	656	13.72	398.93	-186.24	1.82	13.33
45	659	13.76	385.32	-189.53	1.83	13.37
	662	13.79	371.21	-191.23	1.83	13.4
46	665	13.82	356.76	-191.27	1.82	13.43
	668	13.85	342.17	-189.64	1.82	13.47
47	671	13.87	327.6	-186.33	1.81	13.49
	674	13.89	313.23	-181.33	1.79	13.52
48	677	13.9	299.21	-174.67	1.78	13.55
	680	13.91	285.71	-166.39	1.76	13.57
49	683	13.91	272.85	-156.52	1.73	13.59
	686	13.9	260.79	-145.12	1.7	13.6
50	689	13.87	249.64	-132.23	1.67	13.59
	692	13.82	239.54	-117.91	1.63	13.56
51	695	13.75	230.6	-102.21	1.59	13.52
	698	13.66	222.95	-85.18	1.55	13.45
52	701	13.54	216.72	-66.88	1.52	13.35
	704	13.38	212.04	-47.37	1.48	13.21
	707	13.19	209.05	-26.71	1.46	13.04
	710	12.95	207.91	-4.96	1.44	12.8

Top hat dimensions:

 
           -----------a-----------           
           |                     |             a = 0.4149m
           b                     |             b = 0.0533m
           |                     |             c = 0.122m
-----c-----                       -----------

The top hat is mounted at 12mm distance form the active element's legs.

And the narod reflectors dimensions:

           ------------L1------------

                                          L0 = 0.9518m
          -------------L0-------------    L1 = 0.849m
                                          L1_z = 0.4594m
                                          X = 0.303m
           ------------L1------------

X here is the distance between the active element and the narod reflectors plane

[300ohm]

The top hat is mounted at 12mm distance form the active element's legs.

Just so everyone is clear about that, thats 12mm center of wire to center of wire distance. That applies to all NEC spacing measurements.

[nikiml]

A note to the DBGH builders who try to stack two of those.

It seems that the narod reflectors can be moved closer with no much gain loss,
as long as the top and bottom narods are also are aslo moved away from the middle ( the optimizer knows exactly by how much).

Thus one could may be try to have 5 narod reflectors for the whole DBGH...

[nikiml]

With just the middle reflector. It has lower VHF gain and a small dip at ch 17-21. But the SWR is ok.

CM model segmentation is derived from autosegmentation 22 at 800Mhz	
CM for VHF AGT of 1 set the radius multiplier of the souce wire to 1
CM Wire	15	95	n_back1	-n_refl_size	n_refl_z	n_back1	n_refl_size	n_refl_z	radius				
CE
SY radius=0.003175
SY hat_rad=0.003175	'0.0010265
SY l1=0.300
SY g1=0.018
SY z1=0.128
SY l2=0.312
SY g2=0.045
SY z2=0.444
SY x=-0.084
SY a1=0.196
SY a2=0.202
SY a3=0.171
SY b=0.122
SY feed=0.075
SY a1ma2=a1-a2
SY a1pa2=a1+a2
SY a1ma2pa3=a1ma2+a3
SY a1pa2pa3=a1pa2+a3
SY n_top_l=0.183474
SY n_b_l=0.129366
SY n_z_sp=0.012
SY n_h=0.051536
SY n_b_z=a1pa2pa3*.7071+n_z_sp+hat_rad-radius
SY n_top_z=n_b_z+n_h
SY n_b_ystart=-n_top_l	'-feed-0.70711*a1ma2pa3-n_b_ysp
SY n_b_yend=n_b_ystart-n_b_l
SY n_back=-0.317
SY n_back1=x
SY n_refl_size=0.4146
SY n_refl_z=0.726
SY n_refl_size1=0.442318
GW	1	23	0	-feed	0	0	-feed-0.70711*a1	0.70711*a1	radius
GW	2	25	0	-feed-0.70711*a1	0.70711*a1	0	-feed-0.70711*a1ma2	0.70711*a1pa2	radius
GW	3	21	0	-feed-0.70711*a1ma2	0.70711*a1pa2	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	radius
GW	4	15	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	0	-feed-0.70711*a1ma2pa3-b	0.70711*a1pa2pa3	radius
GW	5	35	x	g1	z1	x	l1+g1	z1	radius
GW	6	37	x	g2	z2	x	l2+g2	z2	radius
GW	11	7	0	-n_top_l	n_top_z	0	n_b_ystart	n_b_z	hat_rad
GW	14	15	0	n_b_ystart	n_b_z	0	n_b_yend	n_b_z	hat_rad
GX	20	010
GW	10	43	0	n_top_l	n_top_z	0	-n_top_l	n_top_z	hat_rad
GX	40	001
GW	100	17	0	-feed	0	0	feed	0	radius*.8675
GW	16	105	n_back1	-n_refl_size1	0	n_back1	n_refl_size1	0	radius
GE	0
LD	5	0	0	0	24900000
GN	-1
EK
EX	0	100	9	0	1			
FR	0	0	0	0	800	0
RP	0	60	73	1001	0	0	3	5		
EN

Numeric gain results:

Ch.	Freq 	RawGain	R-in	X-in	SWR	NetGain
	171	5.31	458.77	61.19	1.58	5.09
7	177	5.89	473.08	12.08	1.58	5.67
8	183	6.27	409.56	-33.25	1.38	6.16
9	189	6.51	334.91	-15.11	1.13	6.49
10	195	6.67	290.98	37.68	1.14	6.65
11	201	6.78	277.93	102.23	1.43	6.64
12	207	6.85	292.55	171.94	1.77	6.5
13	213	6.89	339.51	245.61	2.14	6.28
	219	6.9	436.62	317.89	2.52	6.01
						
						
	467	12.38	212.36	-5.77	1.41	12.25
14	473	12.38	257.52	29.49	1.2	12.34
15	479	12.24	307.96	48.56	1.18	12.21
16	485	12	358.84	48.96	1.26	11.94
17	491	11.77	401.6	30.3	1.36	11.67
18	497	11.66	427.16	-0.43	1.42	11.53
19	503	11.69	434.26	-31.35	1.46	11.53
20	509	11.81	429.11	-54.77	1.47	11.65
21	515	11.95	418.52	-69.16	1.47	11.79
22	521	12.09	406.84	-75.91	1.45	11.94
23	527	12.22	396.34	-76.95	1.43	12.08
24	533	12.34	388.09	-73.96	1.4	12.22
25	539	12.45	382.58	-68.27	1.37	12.34
26	545	12.55	380.02	-60.92	1.35	12.45
27	551	12.64	380.49	-52.77	1.33	12.55
28	557	12.72	383.97	-44.59	1.32	12.64
29	563	12.8	390.39	-37.1	1.33	12.71
30	569	12.87	399.6	-31.04	1.35	12.77
31	575	12.93	411.31	-27.17	1.38	12.82
32	581	12.98	425.01	-26.26	1.43	12.84
33	587	13.03	439.96	-29.03	1.48	12.87
34	593	13.06	455.08	-36.02	1.53	12.86
35	599	13.08	469.03	-47.44	1.59	12.85
36	605	13.08	480.28	-62.95	1.65	12.81
37	611	13.07	487.46	-81.49	1.7	12.77
38	617	13.04	489.8	-101.41	1.74	12.71
39	623	12.99	487.55	-120.9	1.78	12.64
40	629	12.93	481.86	-138.85	1.81	12.55
41	635	12.88	473.98	-155.34	1.84	12.48
42	641	12.86	464.16	-171.29	1.87	12.44
43	647	12.88	451.38	-187.14	1.9	12.44
44	653	12.94	434.29	-202.03	1.93	12.48
45	659	13.03	412.39	-214.03	1.96	12.55
46	665	13.13	386.4	-221.13	1.98	12.63
47	671	13.23	357.9	-221.97	1.99	12.73
48	677	13.32	328.69	-215.94	1.97	12.83
49	683	13.38	300.46	-203.1	1.94	12.91
50	689	13.41	274.62	-183.85	1.89	12.98
51	695	13.39	252.25	-158.74	1.81	13.01
52	701	13.29	234.27	-128.34	1.71	12.98
53	707	13.09	221.47	-93.16	1.6	12.85
	713	12.75	214.67	-53.65	1.48	12.58

top hat dimensions:

 
           -----------a-----------           
           |                     |             a = 0.3669m
           b                     |             b = 0.0515m
           |                     |             c = 0.1294m
-----c-----                       -----------

the middle reflector length is .8846m and it is in the plane of the regular reflectors.

[Xauto]

Thank you nikiml for the simple GH4 with simple Top hat. This should work well for me.

[nikiml]

Having not built it, all I can say is:
I hope so!
Post feedback.

[300ohm]

and it is in the plane of the regular reflectors.

That does make it unique. The NAROD reflectors and regular UHF reflectors are only 3.3" behind the driven element. You do lose some gain, but still its a easy to build mid range vhf-hi/uhf antenna.

The FCC intial specs were for 6 db vhf-hi and 10 db uhf for the intended area coverage.

[nikiml]

This one is probably not for build. It is just to give ideas where the narod reflectors could possibly go:

CM model segmentation is derived from autosegmentation 22 at 800Mhz	
CM for VHF AGT of 1 set the radius multiplier of the souce wire to 1
CM 
CE
SY radius=0.003175
SY hat_rad=0.003175	'0.0010265
SY l1=0.300
SY g1=0.018
SY z1=0.128
SY l2=0.312
SY g2=0.045
SY z2=0.444
SY x=-0.084
SY a1=0.196
SY a2=0.202
SY a3=0.171
SY b=0.122
SY feed=0.075
SY a1ma2=a1-a2
SY a1pa2=a1+a2
SY a1ma2pa3=a1ma2+a3
SY a1pa2pa3=a1pa2+a3
SY n_top_l=0.181961
SY n_b_l=0.120969
SY n_z_sp=0.012
SY n_h=0.050178
SY n_b_z=a1pa2pa3*.7071+n_z_sp+hat_rad-radius
SY n_top_z=n_b_z+n_h
SY n_b_ystart=-n_top_l	'-feed-0.70711*a1ma2pa3-n_b_ysp
SY n_b_yend=n_b_ystart-n_b_l
SY n_back1=x
SY n_refl_size=0.410447
SY n_refl_z=1.044516
SY n_refl_size1=0.415772
GW	1	23	0	-feed	0	0	-feed-0.70711*a1	0.70711*a1	radius
GW	2	25	0	-feed-0.70711*a1	0.70711*a1	0	-feed-0.70711*a1ma2	0.70711*a1pa2	radius
GW	3	21	0	-feed-0.70711*a1ma2	0.70711*a1pa2	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	radius
GW	4	15	0	-feed-0.70711*a1ma2pa3	0.70711*a1pa2pa3	0	-feed-0.70711*a1ma2pa3-b	0.70711*a1pa2pa3	radius
GW	5	35	x	g1	z1	x	l1+g1	z1	radius
GW	6	37	x	g2	z2	x	l2+g2	z2	radius
GW	11	7	0	-n_top_l	n_top_z	0	n_b_ystart	n_b_z	hat_rad
GW	14	15	0	n_b_ystart	n_b_z	0	n_b_yend	n_b_z	hat_rad
GX	20	010
GW	10	43	0	n_top_l	n_top_z	0	-n_top_l	n_top_z	hat_rad
GW	15	97	n_back1	-n_refl_size	n_refl_z	n_back1	n_refl_size	n_refl_z	radius
GX	40	001
GW	100	17	0	-feed	0	0	feed	0	radius*.8675
GW	16	99	n_back1	-n_refl_size1	0	n_back1	n_refl_size1	0	radius
GE	0
LD	5	0	0	0	24900000
GN	-1
EK
EX	0	100	9	0	1			
FR	0	0	0	0	800	0
RP	0	60	73	1001	0	0	3	5		
EN

with gain results:

Ch.		7	8	9	10	11	12	13				14	15	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31	32	33	34	35	36	37	38	39	40	41	42	43	44	45	46	47	48	49	50	51	52	53	
Freq	171	177	183	189	195	201	207	213	219		467	473	479	485	491	497	503	509	515	521	527	533	539	545	551	557	563	569	575	581	587	593	599	605	611	617	623	629	635	641	647	653	659	665	671	677	683	689	695	701	707	713
RawGain	7.9	8.78	8.78	8.62	8.48	8.36	8.27	8.2	8.15		13.14	13.25	13.3	13.31	13.29	13.24	13.13	12.97	12.75	12.54	12.4	12.34	12.36	12.41	12.49	12.57	12.65	12.72	12.79	12.85	12.91	12.95	13	13.04	13.07	13.09	13.11	13.11	13.09	13.06	13.03	13.01	13.01	13.06	13.13	13.23	13.31	13.37	13.39	13.33	13.17	12.88
SWR	2.57	1.32	1.41	1.41	1.24	1.15	1.34	1.63	1.95		1.58	1.29	1.12	1.07	1.12	1.19	1.24	1.3	1.35	1.4	1.43	1.43	1.42	1.4	1.38	1.37	1.37	1.38	1.41	1.45	1.49	1.55	1.6	1.66	1.72	1.77	1.81	1.85	1.87	1.89	1.9	1.91	1.91	1.92	1.92	1.91	1.89	1.84	1.78	1.68	1.58	1.46

[nikiml]

I tried convergency test on the original Jed GH8 at freq of 590Mhz and it did not converge.

I am starting to think that maybe the AGT does not stay at 1 when 4nec2 does the convergence test.

[Autofils]

Re post 59:

Originally Posted by Nikiml

I tried convergency test on the original Jed GH8 at freq of 590Mhz and it did not converge. I am starting to think that maybe the AGT does not stay at 1 when 4nec2 does the convergence test.

Nikiml.

j3d's nec files have a high segment count with a segment per 1/2 wave >>10, so that doesn't leave a lot of room for increased segmentation before you exceed the segmentation rules for len/rad parameter. He does this because he believes adjusting the wire radius of the Vsource to fine tune the AGT breaks nec sgmenation guidelines, but .... that's another topic for later discussion.

Don't be overly concerned about the AGT test. Since it is calculated at a specific frequency, it will change at other frequencies.
The 4nec2 help gives this guideline:

Average-Gain-Test
To check the reliability of your model, you can run an ‘average gain test(AGT). 
To run such a test, use the ‘Generate (F7)’ command, select ‘Far field pattern’ and check ‘Aver-Gain test’. Then click the ‘Generate’ button.

When the ‘expert’ button is selected and using the lowest option-box, 
one should select the ‘1 – average gain’ option to perform this same test. (The other option-boxes could be left to 1:Ver/Hor/Tot field, 0: no norm, 0: power-gain). 

To get usefull test results 4nec2 automatically (temporary) removes all 
wire- and other resistive loading and (if specified) changes the gound-conditions to perfect-ground when doing the required calculations.  

The test-results are presented on the main (F2) form. 
The following gradation can be made for the Average-Gain-Test results:

> 0.95   and <    1.05      Model is likely to be accurate
> 0.90   and <    1.10      Model is usable for most purposes.
> 0.80   and <    1.20      Model may be useful, but can be improved.
< 0.80   or    >    1.20      Model is questionable and should be refined.

When the resulting value is below 0.8 or above 1.2 the value is displayed in red color. The db value between brackets can be used to correct the far-field pattern gain.

To see the conv-test in a better light, start the model with segmentation based on the minimum guideline of 10 segs per 1/2 wave. Remember to disable your auto-segmentation option before running the conv_test.

The bandwidth of your UHF is Ch14-51 (470 to 698) mHz. The mid-freq is 584mHz. Set your frequency in the freq tab of the NEC Editor (new) to 584mHz. You will see that 4nec2 reports that freq and it's wavelength in the main window. The 1/2 wave for 584mHz is half of that value or 513/2 = 256.5mm or 0.2565m.

At 10 segs per 1/2 wave, your seg-len (segment length is 25.65mm)
The number of segs for the Vsource (tag 100) is it's (length / seg-len)
In meters that's 0.15/0.02565 = 5.84 segs, which you round off to 5, because the Vsource must have an odd number of segments to ensure the Vsource is exactly at the center of Tag 100.

Repeat the same process for all other tags, except round up the number of segs. (these can be even or odd rounded up integers) and that is your segmentation for 10segs per 1/2 wave at the mid UHF freq of your design.

Now run conv-test and post your results of Gain vs total segments. Then we can discuss the degree of convergence.