Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 3 Next   Go Down
  Print  
Author Topic: HF mobile ant position  (Read 7388 times)
WX7G
Member

Posts: 6035




Ignore
« on: December 11, 2010, 04:48:58 PM »

The best kept secret in HF mobile antenna performance is the mounting position of the "antenna." The reason I put antenna in parentheses is because what we call the antenna is often really an ACED, or Antenna Current Excitation Device. The ACED actually excites antenna current onto the automobile itself, which then becomes a part - and sometimes the dominant part - of the actual antenna.

A good example of this is my compact car. I wanted good 7 MHz performance and settled on two options: A 5' tall Yaesu ATAS-120A screwdriver mounted on the roof, or a 10' Tarheel HP-100A mounted on a trailer hitch. The first option is small, unobtrusive, and costs $370 with a Diamond 400C mount. The second is large, is quite visible, and costs $700 with controller and trailer hitch mount. Both antennas top out at 10' above the pavement.

A NEC-2 model revealed that mounted as described the ATAS-120A would have a radiation efficiency of 9%. And mounted as described the Tarheel would also have a radiation efficiency of 9%. This is revealed by a NEC-2 model where the "ground" loss is 10 ohms and the coil Q of the ATAS-120A is only 90 while the HP-100A is a high 300. Due to the different mounting configurations the radiation resistance of the ATAS is 3.5 ohms while the Tarheel is 1.4 ohms.

Why is there the same efficiency from the smallest (and lowest Q) screwdriver antenna and from a much larger (and high-Q) screwdriver antenna? Antenna mounting is the answer. With the ATAS-120A mounted on the vehicle roof the ATAS and the vehicle form a 10' long antenna. With the Tarheel mounted near the pavement much of it's magnetic field is canceled by an opposite magnetic field from the vehicle. The opposite magnetic field is due to antenna displacement current causing vehicle conduction current in a downward direction (as the antenna has conduction current in the upward direction).

Placing the Tarheel parallel to the vehicle negates that portion of the Tarheel; the lower 5' of the Tarheel becomes ineffective and we are left with a 5' antenna.

So, the 10' Tarheel becomes a 5' antenna while the 5' ATAS becomes a 10' antenna.

The radiation efficiency of the actual ATAS-120A installation was indirectly measured by a VSWR bandwidth method. This measurement shows a radiation efficiency, at 7 MHz, of 11%.

Conclusions: Bigger isn't always better when it comes to mobile "antennas." And higher coil Q isn't always better. System performance can depend more on where the antenna is mounted than what type (length, coil Q) of antenna it is.

Tip: Avoid mounting the antenna parallel to the vehicle.


« Last Edit: December 12, 2010, 04:53:18 AM by DAVE CUTHBERT » Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #1 on: December 12, 2010, 06:52:53 AM »

Your logic is flawed.
Logged

AD4U
Member

Posts: 2164




Ignore
« Reply #2 on: December 13, 2010, 08:22:59 AM »

I am not qualified to comment on your theory or your logic.  I am in SC and you are in UT some 2000 miles apart.  As I previously stated I run an original Texas Bug Catcher on a F-250 Ford diesel truck and I am very happy with it.   However I do not think the Bug Catcher is 11% effiicent on 40 meters.  Even if it were, I do not know how to accurately measure it.

I wish we could park side by side - the only real test that matters in the real world - my full size original Texas Bug Catcher VS your ACED antenna "system" each running 100 watts and do as many tests as practical on any and all bands.

Dick  AD4U
Logged
WX7G
Member

Posts: 6035




Ignore
« Reply #3 on: December 13, 2010, 08:43:01 AM »

Dick,

every measure of radiation efficiency is an indirect measurement. A field strength measurement is an indirect measurement. For amateur measurements (where an antenna test site with known characteristics is not known) an A-B comparison can be used. A full size monopole is often used. I have used this method to test antennas such as the Isotron and the PF antenna.

The most direct way to measure radiation efficiency of a mobile antenna is with an input impedance (Rin) measurement combined with a NEC model of the vehicle and antenna placed over "perfect" ground. This is used to determine the radiation resistance (Rr). Radiation efficiency = Rr/Rin. The formula used to calculate the radiation resistance of a monopole cannot be used because it does not account for the vehicle. Place the "antenna" on top of the vehicle and the vehicle adds to the antenna length. Place the "antenna" next to the vehicle and the vehicle subtracts from the antenna length. Neither has exactly a one-to-one correspondence and so a simulation must be used for a precise calculation of radiation resistance.

Another method - and this is a good check of the base impedance measurement - is to measure the VSWR bandwidth. This is useful for antennas that have a base matching reactance. With a base matching reactance this method is not perfect but gives useful data. The VSWR points must be selected to be close to what the the antenna would present without the base matching reactance. Loss resistance is then added to a NEC model until it exhibts the same VSWR bandwidth.

These are the two methods I've used for several physically short antennas I have designed for commercial use. Signal strength measurements against a reference antenna (quarter wavelength monopole) were also used. All three methods were in good agreement.

I'll take field strength measurements of my mobile antenna vs a 20' vertical having 64 radials. That reference antenna has a radiation efficiency of 50% (found by a base impedance measurement).


« Last Edit: December 13, 2010, 09:23:13 AM by DAVE CUTHBERT » Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #4 on: December 13, 2010, 11:58:08 AM »

David, I just don't agree with your logic.

First, if I were to use your original hypotheses, then a base-station vertical mounted on the ground would have an effective height of about 8,000 miles! That's nonsense.

Next, You say field strength is an indirect measurement. It is, but only if you do it incorrectly. If you were correct, then folks like WB2WIK, who spent a life-time measuring antenna field strengths, would have wasted all his career.

Then you state:

Quote
The most direct way to measure radiation efficiency of a mobile antenna is with an input impedance (Rin) measurement combined with a NEC model of the vehicle and antenna placed over "perfect" ground. This is used to determine the radiation resistance (Rr). Radiation efficiency = Rr/Rin.


This isn't logical either. When you measure the input resistance, you measure everything; Rr, Rg, Rc, etc. They are in series with one another, and by factoring one out so to speak, you don't account for changes in the rest.

Probably the most inane comment is the one about measuring an antenna's bandwidth to determine if efficiency. I can easily build a model with great bandwidth, and poor efficiency, or even great efficiency. And, I can build a model with lousy bandwidth, and have either poor or good bandwidth. The input impedance notwithstanding in any case!

And again, considering the size of the vehicle into an antenna's efficiency calculation, is fraught with problems from the get go.

Logged

AD5X
Member

Posts: 1428




Ignore
« Reply #5 on: December 13, 2010, 04:32:42 PM »

Alan - I think he meant to say that he uses EZNEC to calculate the radiation resistance by modeling the antenna over lossless ground.  Assuming the model is correct, you can then calculate the efficiency since the losses would be the total input resistance minus the radiation resistance.

Phil - AD5X
Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #6 on: December 13, 2010, 05:51:07 PM »

Well, Phil, it isn't as simple as that.

EZNEC will calculate anything you put into it. The problem is, however, that all of the losses are in series with one another, except for some coupling losses between the antenna and the body of the vehicle in question (those are typically small in comparison to the other losses). Since they are in series, changing one, affects the others, and you cannot measure that change with an impedance measurement. Nor can you assume what you measured is a positive effect. A good example of this, is the installation of a cap hat. Any place you put it above the coil, will indeed increase the capacitance equally. However, if it is placed very close to the coil, the added capacitance reduces the coil's Q. Placed correctly, the radiation resistance increases. Both increase the input impedance, but one is by additional loss, the other by an increase in radiation resistance. A fact you cannot measure, or calculate by EZENC (separate as it were).

Dr. Belrose, VE2CV, did point out a way to use EZNEC, and the actual input impedance of an antenna, to arrive at an efficiency level within reason, accuracy wise. That procedure was not in any way close to what David suggested.

As Tom Rauch, W8JI, has stated many times, you cannot assume any measurement you can't measure directly. In other words, there is no way to separate the components making up the input impedance by calculation (factoring out),and for darn sure, you can't assume them which he apparently has.

As I stated firstly, his logic is faulty.
Logged

AD5X
Member

Posts: 1428




Ignore
« Reply #7 on: December 14, 2010, 04:41:23 AM »

Alan - I'm having a hard time agreeing with you on this.  Model an antenna on EZNEC.  Base loaded, center loaded, and add a capacity hat if you want and locate it wherever you want.  The inductor should be modeled as lossless of course.  EZNEC will give you the radiation resistance.  Now measure the actual resonant antenna input.  Subtract the radiation resistance from EZNEC, and the remainder will be all other losses.  This will show the effect of incorrectly placing the capacity hat.  With knowledge of the radiation resistance (if you believe in your EZNEC model), you can determine efficiency.  Where is the error here?

Phil - AD5X
Logged
KG4RUL
Member

Posts: 2721


WWW

Ignore
« Reply #8 on: December 14, 2010, 06:20:11 AM »

Keep in mind that EZNEC is modeling software.  MANY assumptions are made in each analysis that can lead you down the primrose path.  Use EZNEC as a guide but, expect to make changes in a real-world situation.
Logged
AD5X
Member

Posts: 1428




Ignore
« Reply #9 on: December 14, 2010, 09:29:05 AM »

Keep in mind that EZNEC is modeling software.  MANY assumptions are made in each analysis that can lead you down the primrose path.  Use EZNEC as a guide but, expect to make changes in a real-world situation.

In this case, however, the discussion is about modeling a simple antenna over perfect ground with lossless components in order to determine radiation resistance.  I suspect that the EZNEC results are probably pretty good in this case.

Phil - AD5X
Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #10 on: December 14, 2010, 10:04:04 AM »

Well, unfortunately, Dennis is correct. Modeling programs are tools, not panaceas.

In this case, the losses are in series. If you change one, you effect the others, and there is no way around it. In other words, the radiation resistance will change as a result of changing the ground losses. Fact is, I used to believe just like you Phil. That Rr is an exact number based on the overall length, diameter, etc. of the radiating element. But after reading the white papers done by both Rudy, N6LF, and Dr. Belrose, VE2CV, et. al., I've had to retrench my beliefs.

Even if you use the data collected (true or untrue), you can't apply it to the real world, because you have no way of accurately measuring the ground loss. And speaking of which, I have always believed that the ground losses increase with decreasing frequency. That is generally true. However, because there are standing waves between the body of the vehicle and the surface under it which cause the ground loss factor in the first place, you can't assume the losses are linear. In fact they are not. So, the methodology, or logic if you please, is flawed.
Logged

WX7G
Member

Posts: 6035




Ignore
« Reply #11 on: December 14, 2010, 10:07:27 AM »

Phil you're right. For the antennas we are discussing, where the loss resistance is apportioned is not critcal. For example, let's look at a 7.5' center loaded 7 MHz vertical over perfect GND. We can add loss resistance to make the 3:1 VSWR bandwidth 290 kHz: With the loss resistance at the base it's 28 ohms and the radiation efficiency is 8.9%. With the loss resistance at the loading coil it's 25 ohms and the radiation efficiency is 9.8%. The difference in gain is only 0.4 dB.

So, the VSWR bandwidth method of determing radiation efficiency is robust; it is forgiving of incorrectly proportioning the loss resistance. Now our possible error is perhaps 0.2 dB in the example above. One does not need to know how much is ground loss and how much is inductor loss to arrive at an answer that is within a fraction of a decibel of the correct efficiency value.

And we can fairly accurately apportion the loss resistance. A program such as the K6STI coil program tells us the loss resistance of our inductor. Next we add base loss resistance until the modeled and measured VSWR bandwidth are the same.

The vehicle enters into this because it becomes part of the radiating structure. An HF mobile antenna is not quite a vertical. And it's not quite a ground plane. It's closer to a vertical dipole. One proof of this is to model a mobile antenna (vehicle and "antenna") with and without ground. The resonant frequency of the system does not change much; it is somewhat ground independant.

When the mobile system is configured as an "antenna" on top of the vehicle - at an instant in time - there is antenna current upward in the "antenna" and upward in the vehicle. The magnetic fields add, the system current-area is maximized and we have maximum radiation for a given antenna current.

When the mobile system is configured as an "antenna" on the bumper - at an instant in time -there is antenna current upward in the "antenna" yet downward in the vehicle. The magnetic fields partially cancel, the system current-area is minimized and we have minimum radiation for a given antenna current.

When one steps back and looks at a vehicle antenna system it's really quite simple. Placement of the "antenna" to maximize current-area is the most fundamental aspect of mobile antenna system design. If one does not understand an installation can be hit or miss. Hit it right and it works well. Hit is wrong and you can have an expensive installation with inferior performance.

Present day electrical engineering is based on modeling. That is, being able to predict how a circuit will function before it's built. Cut-and-try or lab-only measurements went out of style 100 years ago. Mathematical (paper and pencil) modeling gave way to computer modeling 40 years ago. I design circuits in SPICE, they are built and they work as designed. There is no cut-and-try. Complex integrated circuits are designed entirely in SPICE. Design, build, and they work the first time. The same thing applies to antenna design. And that is much, much simpler than SPICE circuit design. How complicated can a piece of wire be? And if one does want to accurately model real ground they need only use NEC-4.
« Last Edit: December 14, 2010, 10:15:18 AM by DAVE CUTHBERT » Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #12 on: December 14, 2010, 10:16:43 AM »

Apparently, Dave, you didn't read what I said previously. If you really put some thought into it, you'll realize you can't use the VSWR as a measure of efficiency. But... you're going to have to think outside the box for once.
Logged

NK5G
Member

Posts: 102




Ignore
« Reply #13 on: December 14, 2010, 10:59:11 AM »

Your logic is flawed.

Sigh
« Last Edit: December 14, 2010, 11:00:44 AM by David Jackson » Logged
K0BG
Member

Posts: 9860


WWW

Ignore
« Reply #14 on: December 14, 2010, 12:32:09 PM »

What are the numbers? I'd like to look them up.
Logged

Pages: [1] 2 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!