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

donate to eham
   Home   Help Search  
Pages: Prev 1 [2] 3 4 5 6 7 8 Next   Go Down
  Print  
Author Topic: EFHW impedance  (Read 8696 times)
K4SAV
Member

Posts: 2385




Ignore
« Reply #15 on: October 14, 2017, 04:42:03 PM »

The OCFD gives a consistently better report by about 2 'S' units for most contacts.

There has been lots of discussion about end fed antennas over the years but I don't think I have ever heard anyone say they don't work.  That's too easy to disprove.  The fact that you say the end fed is down about 2 S units from an OCFD is a little surprising, but possible in some cases.  I suspect the reason for that will be somewhere in the details of installation or the testing.  It's difficult to compare a vertical to a horizontal antenna since the patterns are so different.  Generally when you do this kind of testing you find that one of the antennas is never always best, but the best antenna depends on the azimuth and elevation angle of the arriving signal.

Jerry, K4SAV
Logged
KD1I
Member

Posts: 396




Ignore
« Reply #16 on: October 14, 2017, 04:48:31 PM »

All very true statements, Jerry. I'm just passing along my experiences with these antennas at my current location. In a different location or configuration, I'd expect different results to a greater or lesser extent.
The difference in the two is, however, consistent, at least at my location.       Jim
Logged
K6BRN
Member

Posts: 450




Ignore
« Reply #17 on: October 14, 2017, 09:54:44 PM »

Gentlemen (or....not):

I'm going to have to side with Ron (KD1L) and Danny (E73M) on this one and suggest that Jerry and Fred go back and do a little more homework.  I really had a good laugh over "... end feeding a wire antenna is really not possible."  All physical evidence to the contrary.

By direct experience in both improvised and well thought out permanent installations I've done at serveral QTHs and while travelling, well designed EFHW antennas, like Danny's (MyAntennas) are incredibly easy to put up, are not fussy about installation details and work very well.  I'm not alone - go read the reviews. They are stellar.

Regarding modelling antennas - the accuracy of the model depends on the software used, the accuracy of the input and the skill of the user, all of which are generally pretty limited in the amateur radio community.  EZNEC, quoted again and again on these forums, is a deliberately limited version of a more capable package, OK for modelling antennas in free space or in proximity to a simplified ground plane.  Great for teaching concepts or acting as an INDICATOR of what might happen when parameters are adjusted, but terrible for accurate predictions in the vast majority of amateur radio installations in the hands of the average user.  Why?  160M-10M bands have LONG wavelengths and couple readily to everything nearby - gutters, imperfect ground conditions, fences, stucco backing, vehicles.... etc. (look up "Near-Field Effects")  This coupling very often distorts antenna patterns in complex ways that are very hard to precisely predict - the environmental input data is rarely very good.  Also, EZNEC makes some simplifying assumptions that result in "corner condition" predictions that are simply nonsense - under conditions that experienced users know to avoid.  All modelling software has limitations.  So please think twice before attacking someone, like Danny, who has actually produced a variety of working and useful antenna designs, because YOUR EZNEC model says his approach will cause a rift in space-time.  Instead, ask for advice.  We are a sharing/learning community at our best.

BTW, last time I looked, high ratio 49:1 matching transformers work better than 9:1 units in practical EFHW antennas.  Ask Danny why.  And the real challenge these antennas have is in power handling.  If a few of the self proclaimed antenna geniuses out there would turn their typing talents to finding better ways of cooling the ferrite cores of the matching network while still maintaining a good weather seal, they might actually create something useful.  As always, asking the RIGHT question FIRST is always critical to finding a USEFUL answer.

Brian - K6BRN

Logged
W1VT
Member

Posts: 2480




Ignore
« Reply #18 on: October 14, 2017, 10:05:33 PM »

If well designed EFHW antennas aren't fussy with regard to installation details, this suggests you may not need a complicated computer model to understand EFHW antennas.

Reply #3, which has a very simple model, predicts that a 49:1 impedance transformation should be better than a 9:1.

Zack W1VT
« Last Edit: October 14, 2017, 10:08:25 PM by W1VT » Logged
K4SAV
Member

Posts: 2385




Ignore
« Reply #19 on: October 14, 2017, 10:39:31 PM »

Gentlemen (or....not):

I'm going to have to side with Ron (KD1L) and Danny (E73M) on this one and suggest that Jerry and Fred go back and do a little more homework.  I really had a good laugh over "... end feeding a wire antenna is really not possible."  All physical evidence to the contrary.

I guess you didn't understand the difference between a theoretical end fed antenna and a practical end fed antenna.  You can't build a theoretical end fed antenna because the source would have to be infinitely small and the source would have only one connection to the antenna, and no counterpoise.  The feedpoint impedance of the antenna would also be infinite.  

I gave a description of a practical end fed antenna earlier, or at least what everyone calls an end fed antenna.  You probably know that can be built and does work.

Jerry, K4SAV
Logged
KB1GMX
Member

Posts: 1496




Ignore
« Reply #20 on: October 15, 2017, 09:33:35 AM »

End Fed Half Wave (EFHW) have the same height vs feed point impedance curves
as a dipole save for the relative impedance at resonance (R,j0)

An EFRA [end fed random antenna] is a wire of non resonant length and
typically has a far lower impedance and is also a reactive antenna. please do
not confuse with the EFHW.   However even the random/nonresonant wire
will have feed point resistance(and reactance) that varies with height.

The use of a 1 wavelength dipole (half wave elements) at resonance with
NEC can with varied height model give the feedpoint impedance change
with height with more than sufficient accuracy for practical uses.  Do not
forget to reresonante the antenna with varied height.  

Now for the punch line.  If your feed network matches 50 ohms to lets
say 3000 Ohms and the antenna 4500 ohms in reality the SWR is a modest
1.5:1.  Conversely its 2:1 range of impedance is 1500 to 6000 ohms.
That little detail explains what makes them less fussy appearing.

The other way to look at it is at xxx height it is perfect.  Your trees are
not that.   Use the likely value as you didn't predict ground conductivity
vs actual accurately, take into account actual height error due to sag,
trees in the near field, loading effects of the coax, and direction of the wind.

Allison
« Last Edit: October 15, 2017, 09:40:07 AM by KB1GMX » Logged
K6BRN
Member

Posts: 450




Ignore
« Reply #21 on: October 15, 2017, 08:35:07 PM »

Jerry:

So... I don't understand that you proposed an ideal antenna (that can never really exist) modelled by imperfect software, that predicts the antenna will not work?  Is that right?  OK.  We agree, I don't understand.

Or did you mean to say "I simply showed an example of an obvious way not to model and EFHW antenna to illustrate that a practical model needs to consider other elements..."  OK.  If so, I misunderstood.  Sorry.  Or maybe you were simply unclear in your earlier post.  Because you did later say EFHW antennas could work.  Very good.  Now we are on the same page.

BTW... I've experimented with Danny's EFHW antennas a lot.  They work fine without a wire counterpoise, without a ground and with a high-impedance common mode choke right at the feedpoint.  So the residual and likely imperfectly modelled LC in the matching network and coax coupling seems to be enough to allow the antenna to radiate reasonably well.  They MAY work a little better (hard to detect, even with PSKReporter signal reports) with six feet or so of coax attached to the antenna before the choke, or with a few feet of "counterpoise" wire attached to ground.  I've looked at one operator's experiments with a shortened EFHW antenna on 20M, about 5 feet off the ground, in which he looks at the near field electrostatic field strength vs. counterpoise length and concludes the counterpoise helps antenna efficiency dramatically.  In terms of far field performance, my results do not match his, as stated above.  Any difference seems relatively small.

Brian - K6BRN
Logged
K6BRN
Member

Posts: 450




Ignore
« Reply #22 on: October 15, 2017, 08:55:15 PM »

Allison:

While I understand what you are trying to say, the reality is that the EFHW antennas ARE less fussy TO USE, period, when equipped with a 49:1 transformer.

Which is probably why they are so popular.  Efficiency and pattern will vary with height, layout, etc.  But users are able to get on the air fast, with a variety of compromises, and useful results, with an antenna that is difficult to see (stealth) and can be carried in a backpack with room to spare, and no outboard antenna tuner (which means less to carry).  I've done this repeatedly when travelling and have a permanent installation as well.  Both experiences have been very good.

In comparison, I find the EFHW antennas work MUCH better than the Comet CHA-250B, which is another easy-up, no-tune antenna (no radials, essentially a gianr whip antenna) but is no where as easy to transport.

Still, other than Danny, no one is addressing the power dissipation and heating issues inherent in the matching transformers of EFHW antennas.  If you'd like to do some work that would actually be useful, find a way to cool the transformer cores so that they can handle much more power for a longer period, and still be weather sealed.  That's a really interesting problem, and one that really deserves attention.  Because it would make a major difference in how EFHW antennas can be used, and add a lot to their utility.

Brian - K6BRN

Logged
K4SAV
Member

Posts: 2385




Ignore
« Reply #23 on: October 15, 2017, 09:51:48 PM »

Brian, my first model reflected what most people think an end fed antenna looks like, and that resulted in an antenna that can't be built.  Obviously they have a misconception of what a real end fed antenna looks like and how it works.   I said before:
"A practical implementation of an end fed wire would have a counterpoise and a method of cancelling the reactance of the antenna."

If you don't have that counterpoise, you are back to the theoretical antenna than can't be built.  That doesn't mean you have to intentionally add a counterpoise, it means there will always be a counterpoise for any end fed antenna you build, whether you intentionally add it or not.  Many people have difficulty visualizing where that counterpoise is located and some people argue endlessly that it doesn't exist because they can't see it. Of course those same people can't analyze the antenna and get it to work without the counterpoise either.

The reactance at the end of a half wave end fed wire is generated by the counterpoise, which often is very short. If you don't cancel that reactance it becomes a nonresonant wire.  That doesn't mean it won't work.  It means the feedpoint impedance won't be what you may expect.

All these things can be analyzed with NEC but it requires that all the parameters be included in the model.  In order to include those parameters you need to know they exist and you will probably have to make some measurements on things like transformer capacitances and losses so that can be included in the model.  For those items it is usually best to measure those parameters.  In most cases the model can start with a calculated answer for losses but it will need to be measured if you want an accurate answer.

Jerry, K4SAV
Logged
K6BRN
Member

Posts: 450




Ignore
« Reply #24 on: October 16, 2017, 05:06:43 AM »

Jerry:

Well stated.  Thank you.

Brian - K6BRN
Logged
K9AXN
Member

Posts: 341


WWW

Ignore
« Reply #25 on: October 16, 2017, 03:09:40 PM »

Can anyone point me to a source showing the impedance of a resonate EFHW  based on height above ground for the ham bands.  If it's in wavelength above ground that would be great.  Just thought I would ask before modeling each one and changing height for each .1 wavelength to .5 wavelength.  I thinking someone must have done this already.
Thanks Fred

Good afternoon Fred,

Here’s a simple way calculate the feed point impedance of a resonant EFHW antenna.

Develop a model of a CFHW antenna using all of the physical properties i.e. wire size, altitude, the barn next door, and the wife’s clothes line.  Then take the resultant impedance and do the following.

Divide 600 by the HWCF calculated impedance) ---- then multiply the result by 600) --- you are home.  600/72 = 8.3 --- 8.3X600 = 5000 ohms. 

Good luck and a great day to you.

Regards Jim
Logged
WB6BYU
Member

Posts: 17043




Ignore
« Reply #26 on: October 16, 2017, 05:09:54 PM »

Which is effectively assuming the antenna wire has a characteristic impedance
of 600 ohms, and transforms impedance like a quarter wave transmission line.

If you want to be more precise you can adjust the 600 ohm impedance for
the actual impedance based on the wire diameter.
Logged
K9AXN
Member

Posts: 341


WWW

Ignore
« Reply #27 on: October 16, 2017, 06:42:32 PM »

Good evening Brian,

Which is effectively assuming the antenna wire has a characteristic impedance
of 600 ohms, and transforms impedance like a quarter wave transmission line.

NO, like a 1/2 WL wire.  It has deliberate losses due to radiation.  The EFHW antenna functions far differently than a CFHW wire.  The reflected energy from the far end is contradictory to the drive energy whereas the CFHW it is complementary.  It takes many cycles to spool up a CFHW antenna whereas the EFHW can spool up as fast as the drive source is capable.

Keep in mind that all current flowing in that antenna has entered through a port that has an insertion impedance of approximately 600 ohms.  When Fred created the CFHW model he entered the wire size which I suggested.  


If you want to be more precise you can adjust the 600 ohm impedance for
the actual impedance based on the wire diameter.

That is done when he defines the inputs for the CFHW model.



Have a super day and thanks for the post.

Regards Jim
« Last Edit: October 16, 2017, 06:45:03 PM by K9AXN » Logged
G8HQP
Member

Posts: 595




Ignore
« Reply #28 on: October 17, 2017, 04:00:49 AM »

Quote from: K9AXN
That is done when he defines the inputs for the CFHW model.
No. The radiation resistance (also the current-maximum-fed feedpoint impedance) of a half-wave dipole (or multiples thereof) varies very little with wire size. The end-fed impedance varies greatly with wire size.

Quote
It takes many cycles to spool up a CFHW antenna whereas the EFHW can spool up as fast as the drive source is capable.
Are you saying that a centre-fed halfwave has a much smaller bandwidth than an end-fed antenna of the same length?

Quote
Keep in mind that all current flowing in that antenna has entered through a port that has an insertion impedance of approximately 600 ohms.
?
Logged
K9AXN
Member

Posts: 341


WWW

Ignore
« Reply #29 on: October 17, 2017, 01:37:23 PM »

Quote from: K9AXN
That is done when he defines the inputs for the CFHW model.
No. The radiation resistance (also the current-maximum-fed feedpoint impedance) of a half-wave dipole (or multiples thereof) varies very little with wire size. The end-fed impedance varies greatly with wire size.

WHY?

Quote
It takes many cycles to spool up a CFHW antenna whereas the EFHW can spool up as fast as the drive source is capable.

Are you saying that a centre-fed halfwave has a much smaller bandwidth than an end-fed antenna of the same length?

No, not at all, they are the same and so to is the Q.  The CFHW antenna requires a significant number of cycles to reach max current because the current in the complementary reflected  waves must enter the antenna through a 600 ohm insertion impedance, not 72 ohms. 

Feed it with 1 KW.  The drive from the source first half cycle will be 270 volts and capable of 3.75A and because it sees a 600 ohm wire (No reflections yet) only .450A will enter the antenna.  The feed point at the beginning of the second half cycle sees the return of the first complementary reflection which will 270V minus radiated energy about 12%.  This will be combined with the drive for a net voltage of 540V @.90A. This goes for many cycles until the reflections equal 3.3A plus the drive 270V@.450A through the 600 ohm entry equal 3.75A.       
 

The insertion impedance of the EFHW is also 600 ohms.  Stop and think for a moment.  If your driving an end EFHW antenna the WORKING impedance will be approximately 5000 ohms.  Feed it with 1 KW.  It will require 2250V @ .450A.  The first cycle will see 2250V applied to the end of the wire which is 600 ohms --- 3.75A because no reflections have returned.  The drive source could be capable of 3.75A and if it is, 8437 watts will enter the antenna --- not good.  However the drive will normally have been designed for .45A @ 2250V in which case only .450A enter the antenna because the capabilities of the driver will limit the surge of entry to .450A.  The second  cycle will see the first reflection returning which is only 270 V because the driver impedance limited the voltage and current to 270V @.450A.  This reflection is contrary to the drive reducing the voltage by 270 to 1980.  This will continue until the reflection reaches 1980V. plus the drive reducing the net drive voltage from 2250V to 270@.450A.  2250-1980 = 270 and the current has been reduced from 3.75A to .450A 

This process is limited by the capabilities of the driver.  If it is capable of more current the iterations will be truncated whereas the CFHW antenna driver is limited by the 600 ohm insertion impedance.  The CFHW will take about 90 iterations to reach steady state.   
 

Quote
Keep in mind that all current flowing in that antenna has entered through a port that has an insertion impedance of approximately 600 ohms.  Also as I recall, it will be close to 600 ohms as long as the diameter of the wire is significantly smaller than the wave length.


If anything is impossible to understand --- I will do best I can to clarify it.

Thanks and a good day to you. 

Regards Jim
Logged
Pages: Prev 1 [2] 3 4 5 6 7 8 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!