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: Prev 1 [2] 3 Next   Go Down
  Print  
Author Topic: 706Mk2G and Auto-Coupler in car - length of coax?  (Read 3008 times)
WX7G
Member

Posts: 6131




Ignore
« Reply #15 on: March 29, 2010, 11:56:21 AM »

If you are determined to measure the efficiency there are two ways I have done it for short monopoles:

1) A known reference antenna
2) A Base impedance and VSWR bandwidth measurement

Both methods will give the same efficiency number. I find that (2) is easier to do. For (1) you need to construct the antenna. For your antenna the reference antenna might be a quarterwave vertical with 30 radials. We can assume that the GND loss resistance is 5 to 15 ohms and peg the efficiency of the reference antenna as 71 to 88%. This is a range of 1 dB. A more precise measure of the efficiency can be had by measuring the base impedance.

Then you compare the mobile installation against the reference antenna using 'precision' attenuators and a receiver. Standard 1, 3, and 6 dB attenuators is all you need.

To do method (2) you need to measure the base impedance of your antenna at resonance. We know the radiation resistance by modeling the antenna and car. Take the radiation resistance and divide by the measured resistance. There is the efficiency. As a check we take the VSWR bandwidth (between 3:1 points) and plug this into the model. Add loss resistance unitl the model VSWR bandwidth looks like the real deal. This should agree with the first part of method (2).

A thermal measurement can be used to determine the loading coil loss. This can be very accurate using a DC substitution method. But we are left with no direct way to thermally measure GND loss.

And there is the "Wheeler Cap" method but for that you need to find a screen room to drive your car into.
Logged
AA4PB
Member

Posts: 12892




Ignore
« Reply #16 on: March 29, 2010, 12:54:50 PM »

"What you will find is that the base loaded 9' whip on your Mazda will be radiating around 10% of the power applied."

Doesn't that depend on what band we are talking? Are you trying to say that the efficiency of a 102-inch whip is only 10%, even on 10 Meters? How about 15 Meters or 20 Meters?

Even if the low efficiency you claimed was true on all bands, I still wouldn't waste an additional 3dB (half the power again) in a piece of coax when it can be avoided.
 
Logged
K0BG
Member

Posts: 9878


WWW

Ignore
« Reply #17 on: March 29, 2010, 07:26:09 PM »

Wouldn't it be nice if you could assume a specific measurement against a standard antenna? Well, the short answer is, you can't. Any given power level, at any given angle, for any given vertical, is highly dependent on the ground losses present. Unfortunately, you can not measure ground loss directly. You might infer it, or you might assume it, but you can't say with any certainty what it is. So comparative field measurements are not definitive, albeit relative (?).

Efficiency is another debatable figure. Since you can't measure ground losses directly (and modeling programs can't calculate them either), and the fact you can't measure field strength at multiple angles (from 5° upwards of 50°) easily, then calculated (or assumed) efficiencies aren't even close!

The key to maximizing efficiency is to minimize the losses we can't assume, but ones we know are present. The obvious one is ground loss, and mounting height consistent with maximizing the metal mass directly under the antenna is one key. Raising the current node within the length of the antenna is also important. This means, center loading, capacitive top loading, and maximizing overall length are also key ingredients.

Anything else, is assuming facts, and not ones which are evident.
Logged

WX7G
Member

Posts: 6131




Ignore
« Reply #18 on: March 31, 2010, 07:17:46 AM »

K0BG,
field strength measurements at multiple take-off-angles are not needed to indirectly measure the radiation efficiency of a mobile whip against a reference vertical. The elevation radiation patterns are very similar and only one angle will suffice. If you mean angles in Azimuth you are correct. Multiple measurements (rotate the vehicle) are needed.

Yes the radiated field is ground dependent for the mobile antenna itself. So we locate the vehicle over a representative ground, perhaps a parking lot. The reference 90 degree vertical has many radials and the efficiency of it falls within 1 dB of calculated. It can therefore be considered to be ground independent. The far-field ground falls out of the error equation as it effects both antennas indentically.
Logged
WX7G
Member

Posts: 6131




Ignore
« Reply #19 on: March 31, 2010, 11:59:33 AM »

AA4PB, the thread has been about 40 meters
Logged
AA4PB
Member

Posts: 12892




Ignore
« Reply #20 on: March 31, 2010, 02:44:47 PM »

Why would anyone invest in an autotuner to work only 40M?
Logged
K0BG
Member

Posts: 9878


WWW

Ignore
« Reply #21 on: March 31, 2010, 06:10:06 PM »

David, I wish it were that simple. The angle of maximum radiated power doesn't change much, but few of us have the equipment to measure field strength at several wave lengths away, and at an angle of about 27°. Measuring the pattern at lower angles, or at close-in distances, is fraught with problems.

The energy at the lower angles, especially that part below 15° is highly dependent on ground losses. You can change actually increase ground losses, and get a better near field reading, while the far field goes to pot.

It's difficult to model, if for no other reason than folks underestimate ground losses in a mobile. While the accepted figures of 2 to 10 ohms (10 through 80), the truth is they may be several times this, and you have no real way to measure them. This is yet another reason why antenna shootouts are suspect at best.
Logged

CHRISDX
Member

Posts: 244




Ignore
« Reply #22 on: April 01, 2010, 07:38:27 PM »

UPDATE:

First, I coordinated with a ham friend ten miles away. We tested 4 bands with both the 56" whip and the 102" whip.  Results are as follows:

10M S4 S7+
15M S2 S7
20M S0 S2
40M s0 S2


then today I swapped the 10' of coiled up coax feedline to a 8" single wire as in the OP.  The results with the 102" whip were a LOSS of 1 to 2 S units on all four bands.
Logged
WX7G
Member

Posts: 6131




Ignore
« Reply #23 on: April 02, 2010, 07:33:36 AM »

Let's see how your data compares to the predictions. The prediction is that the 56" whip will have 1/4 the radiation resistance of the 102" whip. Since the ground and coupler losses dominate the feedpoint resistance we expect a 6 dB difference. And due to the greatly increased coupler losses incurred by feeding the non-resonate whip through coax we expect further losses. The first set of data supports the prediction that the 102" whip will radiate more than the 56" whip.

The second prediction is that a short length of wire between the coupler and the whip will perform much better than a short (or long) length of coax. Note that the stated condition was that the GND connection between the coupler and the base of the whip be a "straight shot." What is meant by this is a low resistance and low inductance path. The second set of data is contrary to the prediction of lower performance with coax.

This leads me to think that there is not a "straight shot" GND from the tuner to the whip. An experiment to conduct is the tie the tuner GND to the whip base area with a wire. A screw near the base of the whip will do. The coax has now been replaced with a short section of parallel wire line - similar to the earlier recommendation of twin lead or ladder line.
Logged
CHRISDX
Member

Posts: 244




Ignore
« Reply #24 on: April 02, 2010, 08:15:38 AM »

Dave,

Thanks for the info.

I already had the whip mount grounded that way. A wire goes from the whip mount to the tuner ground to the radio chassis ground, to the car chassis/batt neg  (this car's battery is right there in the trunk).

-Chris
Logged
WX7G
Member

Posts: 6131




Ignore
« Reply #25 on: April 02, 2010, 08:35:16 AM »

Chris, interesting. We either have an installation that is different than what we are thinking, or we have a measurement error.

If the comparison between coax and wire was performed at the same time the measured data would carry more weight. That is, install coax and take measurements, change to wire and repeat.

However, if the coax/wire measurements were taken on different days - with the car having been moved and returned - or with the monitoring station having had any changes (changing bands or setting up the receiver differently) the data is suspect.

I perform similar EMC and RF measurements as part of my job. I have a thing I call 'data freshness' that does not allow relative measurement data from other than the same time and place to be compared. It's poor form for engineering decisions to be based on bad data. Bad data = bad decisions. Might as well flip a coin.
Logged
WX7G
Member

Posts: 6131




Ignore
« Reply #26 on: April 02, 2010, 08:43:20 AM »

Alan, it seems that most folks look at mobile antenna shootouts the wrong way. I know that you know what's happening but many do not.

A mobile shootout compares two antenna SYSTEMS. Example: If at a shootout my Smart Car with a Hamstick outperforms a Chevy van with a Tarheel 200A that is what has happened. My system beat his system.

But we cannot then say that a Hamstick is better than a Tarheel 200A.

If people were to view mobile antenna shootouts this way - the correct way - the results would not be used as marketing data for mobile antennas. The results would not be a reflection of the antenna but a reflection of the skill of the ham who built the installation (system).
Logged
WX7G
Member

Posts: 6131




Ignore
« Reply #27 on: April 02, 2010, 10:01:08 AM »

Chris, what makes your wire/coax data stand out as being in error is the direction of the data.

We expect that the ideal situation on all bands is a short length of wire between the tuner and the whip. We also expect that even a short length of coax will be greatly inferior and never superior. But your data says that the coax, and a long length at that, is significantly superior to a short length of wire. We must infer:

I1) The data is in error
I2) The 10' of coax is magically adding 12 dB of signal
I3) a 1' length of wire is magically adding 12dB of loss compared to a 10' length of coax and the additional tuner losses caused by it

We gain confidence in experimental data by using a mathematically based prediction. Either can be wrong. E is experiment and P is prediction. We can have:

E right
P right

E right
P wrong

E wrong
P right

E wrong
P wrong

We work until we get the first senario. We then know how to model it and we have good experimental data.

The last senario is a trickly one. There can be the case where I and P are wrong but the numbers agree. We then think we know how to model the system and that our measurement is right.

Your measurement of wire/coax fits the second or the third senario with things pointing strongly towards the third.

It is quite rare for someone to walk up to a new measurement - one they have not mastered - and output good data right away. I usually budget 40 hours to gaining confidence in a measurement.
Logged
K0BG
Member

Posts: 9878


WWW

Ignore
« Reply #28 on: April 02, 2010, 10:25:34 AM »

Dave, if you looked at my web site under Antenna Shootouts, you'd know I agree with you. In fact, there are so many different parameters which have to be considered, you can not make any definitive comparison between any two mobiles. The exercise is just that, and is about as accurate as your common S meter.
Logged

WX7G
Member

Posts: 6131




Ignore
« Reply #29 on: April 02, 2010, 12:26:18 PM »

Alan,

yes I've seen your take on shootouts. I've read your mobile site several times.

If the shootouts would invest in an MFJ-762 ($89) step attenuator and use an AC volt meter connected to the receiver audio output they could measure the relative gain of mobile antennas to within 1dB.
Logged
Pages: Prev 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!