Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

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

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
Author Topic: QRP EFHW Antenna  (Read 20691 times)

Posts: 15

« on: February 04, 2010, 08:33:47 AM »

Alan Applegate, K0BG, has written a piece on the Eham Main page titled "Mythical Antennas", slamming End-Fed Dipoles, OCF Dipoles, G5RV's and Quads. Perhaps an premature attempt at subtle humor (Ala Larson E. Rapp),  this article certainly could discourage anyone less than a antenna engineer from ever writing about their experiences using any of these or any other antennas.

I am writing this post as a QRP operator using a PAR end-fed halfwave (EFHW) multiband on 40, 20 and 10. Maybe in the purest sense it is not a "Dipole" but if it is a halfwave long and behaves the same as a dipole then I see no harm in calling it a dipole.  This EFHW has been the most consistent performing antenna I have used. I have used it in the Inverted Vee, vertical and sloping configuration.   I have compared it to verticals and doublets and it consistently delivers the contacts. The best configuration seems to be the sloper.   This is NOT a testimonial for the PAR antenna but rather a testimonial for the ability of a EFHW to perform well in any configuration in the portable QRP environment..  

There is an issue with whether or not a counterpoise is required. I use a K1 and varying lengths of feedline. The K1 built-in tuner tunes all bands with no problem and the EFHW performs.   I now have a AIM 4170 that is BlueTooth capable and can be operated completely remote. I propose to test the EFHW to see what effects the presence or lack of a counterpoise will have on feedpoint impedance.   Also I am interested to see what changes of feedpoint impedance is caused by differing configurations.  

I hope to perform these tests within the next two months. While I have a component test engineer background I would like to hear from other QRP antenna experimenters about any protocol issues and how to avoid testing mistakes...



Posts: 1058

« Reply #1 on: February 04, 2010, 02:54:24 PM »

Alan is a pretty smart cookie, and those are the kind of people it's hard to argue with. One of the good things about end-fed antennas are that sometimes they are more convenient to erect than the traditional dipole. For example, if you're a camper or hiker, you only need one tree to get one up and running, and you can eliminate the weight and bulk of the coax feedline.

My interest would not necessarily be the impedance change with a counterpoise, but with the efficiency of the antenna with various lengths of counterpoise. In other words, is a counterpoise really necessary, and if one is, how does antenna efficiency change with its length? Tnx.

Posts: 15066

« Reply #2 on: February 04, 2010, 03:23:57 PM »

First, a di-anything (like dipole) has two poles or conductors. The fact that an end-fed 1/2 wave can have the same pattern as a dipole doesn't make it a dipole.

The problem with end-fed antennas (half wave or any other length) is that it MUST be fed against a counterpoise of some sort in order to have a complete circuit. In the case of the PAR, the coax shield is the only thing that could be acting as the counterpoise thus it becomes a part of the antenna. That means that antenna performance is going to be quite dependent on how long the coax is and how it is routed. That may or may not be a problem in any particular installation. As long as that is understood then its fine. Problem is that some report it as some sort of "magic" antenna that doesn't require a counterpoise - which it is not. I've even had some claim that the antenna is "voltage fed" and requires no current flow and thus no counterpoise. P = I*E so if I is zero then P is zero, no matter how much E is present. How does an antenna radiate signal with zero power applied?

A center-fed dipole on the other hand has much less (mayby none) current flowing on the outside of the shield so it is not so dependent on what you do with the coax feed line.


Bob  AA4PB
Garrisonville, VA

Posts: 550


« Reply #3 on: February 04, 2010, 07:01:55 PM »

I'll confess up front that I'm not an expert, but I do disagree with this
statement of Alan's:

> If you just stop and think about the phrase end-fed dipole, the idea
> just falls apart. If it is end-fed, then it can't a dipole!

The term "dipole" (as I have understood it), refers to the number of
poles that the antenna has (it's field strength radiation pattern), not
the number of conductors.

So (I believe) your end fed antenna is indeed a dipole on the
wavelength at which it is a half wavelength long, because at that
wavelength, it has two lobes broadside to the antenna. On 20
meters, it will still have two lobes, although they will be sharper.
At 10 meters, though, it will have four lobes, so it is no longer
correct to call it a dipole at 10 meters. But that is also true for
a two conductor 40 meter center fed dipole on 10 meters!

Concerning the counterpoise, I've played with an EFHW antenna
for 40m in the field, and I spread out a set of some radials. While
listening to a signal, I repeatedly connected and disconnected the
connection to the radials. I couldn't hear any difference, although
it's entirely possible that I wouldn't be able to hear even a 3 db
difference (half an S unit).

Scott W5ESE

Posts: 8918


« Reply #4 on: February 04, 2010, 07:47:32 PM »

I now have a AIM 4170 that is BlueTooth capable and can be operated completely remote. I propose to test the EFHW to see what effects the presence or lack of a counterpoise will have on feedpoint impedance.   Also I am interested to see what changes of feedpoint impedance is caused by differing configurations.  

I hope to perform these tests within the next two months. While I have a component test engineer background I would like to hear from other QRP antenna experimenters about any protocol issues and how to avoid testing mistakes...


A requirement for a "counterpoise" is kind of missing part of the picture.  This is one of those things where plain English and physics don't quite work out neatly.

You have a feedpoint.  You're going to put an antenna analyzer box at that feedpoint.  Current comes out of the box.  Half a cycle later, the current goes back in.  Where did those electrons come from?  Where do they pile up and deplete?  At RF, the answer is that they pile up and deplete from the surface of the box, and anything connected to the surface of the box.

There's no way around that.  Without that, as has been said, there would be no current.  No current = no power.  The current *has to come from somewhere.* 

Now, the nice thing about an EFHW is that the current needed at the feedpoint is low for a given radiated power, because the feed impedance is high.   But you need some current.

If all the EFHW is connected to is a small instrument case, it will cause charge to pile up and deplete from that box, and that charge will attract and repel charges in the earth, causing them to pile up and deplete underneath the base of the antenna.  In this way, a displacement current flows through the capacitive coupling of the case to earth. 

I think that tiny capacitance is probably going to add enough capacitive reactance to make your EFHW impedance out of whack, but you might even be able to retune it to work that way.  But it *still has a counterpoise.*  That counterpoise is the case of your instrument, and the resulting small capacitor to the surroundings. 

This is the important thing... ***the counterpoise is the analyzer's case***

There are a lot of situations in which the lack of good, low impedance connection to earth just doesn't much matter to the *efficiency* of an end fed half wave.  Even a hundred ohm ground rod connection or a 300-j300 resistive plus capacitive coax shield to ground connection can be tuned out and doesn't add too much loss.     

When you use your EFHW in the field, even with a vanishing length of coax, your radio case is your counterpoise.  At QRP levels this is *totally a valid thing to do,* and that's something that people tend to neglect when they give overly broad advice.  A few tens of milliamperes of RF current flowing over the radio and on the earth nearby when you're out in the field is not at all a problem.  Probably never will be, unless you manage to set up a particularly pathological situation, like the preferred current path is on your CW keying line or microphone cord.

So when someone insists "an end fed half wave needs a counterpoise" without qualification, they aren't really considering the situation in which the radio case and coax *are a valid counterpoise.*  There are lots of reasons to use a stouter, low impedance counterpoise, but QRP field work, IMO, isn't one of them.

I think it will be useful for folks to see your result, so I think it's a good idea.  But remember to write up in your results, instead of "no counterpoise," the exact length of coax between matching box and radio, how high your radio is off the ground, how it's connected to the battery, etc.  If you just use the floating AIM box, tell people how high it is from the ground.

There's not really a "no counterpoise" situation that you can set up.  There never is.   And it doesn't take much counterpoise to get excellent efficiency from an end fed half wave. 

Good luck with your measurements.



Monkey/silicon cyborg, beeping at rocks since 1995.

Posts: 18521

« Reply #5 on: February 04, 2010, 09:52:10 PM »

One other thing missing from your proposed test is a reference antenna to
compare it to.  A half wave dipole at 25' or so, fed with RG-58 coax would
be a reasonable choice, since it is a practical alternative in many cases and
is well analyzed and understood as far as its performance (at least when there
is insignificant common mode current.)

The input impedance of an antenna isn't the most important attribute - the
ability to radiate power (especially in the desired direction) is.  We can
always add a tuner to match the impedance when needed.  So while the
impedance measurements with different lengths of counterpoise are of
interest, the radiated power compared to a standard is what determines
the usefulness.  You can use a signal from a local station for comparison,
or do a receive comparison with a number of signals received off the air.
The actual difference between the two antennas is expected to vary depending
on location and propagation path, so you may need to take readings on a
number of stations.  (You may find that one is better for DX and the other
for local work, for example.)

While you are at it, can you add a current probe and measure the return
current in the counterpoise?  Or the common mode current on the coax
and how it varies with length?

One other test that you can run is to use the same wire installed in various
configurations and see how much the impedance changes.  For example,
I would typically  make my own matchbox for an end-fed half wave with
a coil and  trimmer capacitor.  So far, it seems that once I tune it up in
a relatively nominal installation, the tuner doesn't need further adjustment
in the field (unless I have a really weird installation.)  The same is true of
dipoles - I've used the same wires many times over the years and never
bothered with a tuner after adjusting them the first time in a local park.

The reason for this test is that it is often easier and simpler to build a
matchbox for a specific wire and tune it once rather than having to retune
it every time I set it up.  (I'm big on simplicity.)  A fixed coil and a trimmer
capacitor that can be touched up if needed (but usually not) can be made
very small and light for QRP.  If the match is still good when the wire is
installed in various ways, it becomes more attractive to build such a tuner.

And I agree with Dan - you still have a "counterpoise" even if it is just the
case of the test equipment.  In that case you'll probably find that the antenna
performs very differently on the analyzer than it does when connected to
the K1 through a length of coax.

On one of the other threads we were discussing ways to compare antennas.
My suggestion was to run the antenna under test and a reference antenna to
a common receiver.  You can put an attenuator in one lead and adjust it
until the signals from both antennas are equal:  the amount of attenuation
required tells you the relative difference in the signals from the two antennas.
(This was designed for testing NVIS antennas, so it is less prone to confusion
due to different propagation modes, polarization, etc.)  You can put a digital
meter on the receiver's AGC line to measure the two signals, as long as
propagation is relatively steady compared to the speed of switching back and
forth between them.  This can be automated with a simple circuit to give a
meter readout of which antenna is stronger, making it easy to adjust the
attenuation to get them to match.  This gives a more accurate reading of
the relative performance than S9 + 25dB vs S9 + 30dB when the meter is
only marked at +20 and +40, and you're listening to a SSB signal.

Posts: 711

« Reply #6 on: February 05, 2010, 12:47:14 PM »

I read Alan's piece and found it quite interesting. I think I have a different take on what he wrote. In my way of thinking, he wasn't trying to discourage anyone from using any kind of antenna. Rather, he was just pointing out that many antennas today are somewhat misrepresented in their marketing and advertising with claims made that simply don't hold up or are just inaccurate. But that doesn't mean they don't work.

All kinds of less-than-optimal ham antennas work every day and produce many good contacts. But despite having skewed patterns, high losses, poor take-off angles, poor grounds, excessive feedline radiation and host of other less-than-optimal situations, they still radiate RF and people still make lots of contacts. But Alan wanted readers to know that the marketing and advertising folks are putting a lot of spin on what they put in their ads.

But then -- what product doesn't? Advertising has always been about creating desire -- desire to buy something -- and the fact that the product might be presented in the very best possible light -- even to the point of inaccurate exaggeration -- doesn't cause commerce to come to a grinding halt.

Heck, I do rather well on 40-meter SSB from my car with a 7-foot tall vertical whip antenna. There is no way under the sun that it's remotely efficient -- that would pretty much defy the laws of physics -- but I still make lots of contacts and often get good signal reports for a mobile station. And that just proves that RF is quite robust and loves to travel far and wide even though its take-off point -- the antenna -- is often not the best. But that doesn't mean it won't work because it will -- just like some of the silly homemade antennas I stuck out apartment and condo windows for years before buying a home.

73, N4KZ

Posts: 15066

« Reply #7 on: February 05, 2010, 01:10:12 PM »

My only point was that EVERY efhw antenna HAS a counterpoise. You may not recognize it but it IS there. If there were absolutly no counterpoise then the antenna would not radiate any signal. I define a countpoise as anything that conducts the return current - not necessaily a bunch of radial wires. It can be the coax shield, the radio case, and even the body holding the radio (as in the case of an HT).

A properly defined counterpoise (as in radials) usually means less variables and a better controlled antenna performance. That's not to say that an efhw connected to a QRP rig won't work. Try laying one or two 1/4 wave wires on the ground connected to the radio case. Usually you will see an increase in performance. Try adding 4 1/4 wave radials to your 2M HT. I'll bet you get better performance with distant repeaters.

Even Elecraft recommends that you lay out at least one radial when using their radios and tuners with end fed wires.

My only problem with PAR is when they claim it is a "no counterpoise antenna". In my opinion they should explain how it works and tell people that how they deploy it and the feed line will impact its performance.

Same issue with mfgs who sell a single element antenna with no traps or tuning and say it is "an all band antenna". When you read the fine print you discover that it requires the use of a tuner. Is it really an all band antenna?


Bob  AA4PB
Garrisonville, VA

Posts: 106

« Reply #8 on: February 19, 2010, 04:09:43 PM »

==So when someone insists "an end fed half wave needs a counterpoise" without qualification, they aren't really considering the situation in which the radio case and coax *are a valid counterpoise.*  There are lots of reasons to use a stouter, low impedance counterpoise, but QRP field work, IMO, isn't one of them.==


I can back your statements word-by-word. I'm also a QRP'er and an EFHWA user, and from the many experiments that I did I have come to exactly the same conclusions. In the reality, especially when /P, the rig is likely to be seated on the ground or very close to it. The rig case, the wiring, the coax (if any), the body of the operator himself, are more than enough to provide all the counterpoise that is needed with an EFHWA at QRP levels, especially on 20m and up. In fact, adding too much counterpoise in addition to the above "implied" one in some cases may actually make things slightly worse!


Posts: 15066

« Reply #9 on: February 19, 2010, 07:22:03 PM »

The dB loss in a less than ideal counterpoise (radio sitting on ground) is the same whether you are running 5W or 500W. Of course the 5W isn't enough to cause all the RFI issues, RF burns, etc so you can get away with it. It might be worthwhile to add a couple of counterpoise wires to see what difference it makes on Rx and Tx.

Remember, the difference in signal strength on the receive end when you loose half of 5W is the same as it is when you loose half of 500W. 3dB is 3dB no matter how you cut it.


Bob  AA4PB
Garrisonville, VA

Posts: 1824

« Reply #10 on: February 22, 2010, 08:25:46 PM »

I also have two of the EF40/20/10 as well as EF17,
EF40 and EF6.  I had built EFHW antennas homebrew
before buying to evaluate and understand the antenna.

FYI:the EF20/20/10 is a loaded antenna on 40 with a
total length of 40ft, on 20 the choke makes it a half
wave and on 10m it's a end fed fullwave.

They work as stated (at least per Dale at PAR, there
are other vendors) and the feed point at the end is
a major plus for some installations.

I found the article a bit too brusk and not totally
clear as to what what being complained about.

As to the end fed half wave antenna, it works.  There
are several good articles on the net explaining how
and why they work.  The short form is when you feed
a resonant halfwave at the end you see a very high
resistance non reactive load and ANYTHING at the
other side of the matching network will serve as a
counterpoise as the current is very low.  So the
coax shield is the return or a 3 foot wire (at 20M)
if that wasn't available.  If the antenna is correct
in it's length and the matchbox is right the
difference for coax as return or adding a counterpoise
is not measurable in the real world unless one is
working in the sub  1dB range.  Of course that may
vary with height, orientation and close field
parameters (houses and trees).

There is a yabut, if the antenna element is not
resonant it is not a EFHW it's a random wire that
is nearly a half wave and all bets are off.  No
maybe about it.  In that condition (non-resonant)
there can be substantial RF on the shield depending
on how far off things are. Also if there is RF on
the shield handling the coax disturbs the SWR so
making the radiator resonant is important.

If you need a antenna tuner to feed a coax fed EFHW
you are really doing it wrong.  It is not needed
as the feed point matching network does this already.

As to comparison, I did that with the EF6 (6M)
antenna and compared to a 6M dipole it's exactly the
same at 200M away from the reference signal point from
the same mounting (horizontal). The site was a large
area field to minimize other objects influence.
Compared to a vertically oriented dipole with the
feed dressed perpendicular for 1/2 wavelength no
difference measurable (less than 1DB). However it was
far easier to hang the EF6 than support the vertical
dipole and the coax.  The results were easily
reproduced and repeatable and 6M was used as its
smaller and I have known gear for that frequency

Also never call it a dipole, there is only one element
the radiator.


Posts: 93


« Reply #11 on: March 09, 2010, 11:45:51 AM »

Hi Jim,

It is easy to be convinced that no antenna will work! I can confirm that you'll probably be happy just feeding the end fed half wave antenna without an elaborate 'counterpoise' (I did not say NO 'counterpoise').  The PAR antennas simple use a high reactance transformer I believe.  I like to use a simple link coupled circuit with a parallel resonant secondary.  Just make sure the antenna is resonate to minimize the return currents (and the need for a 'counterpoise').  I do this by finding the resonate point on the coupler first for the desired frequency using a load resistor (I like to use 4.7k Ohms).  Then I connect the antenna and adjust its length until the coupler again achieves resonance at the same point on the variable capacitor's dial.

I've used this type of setup for many years and have compared it to normal center fed with coax dipoles with no detectable difference.  My most recent setup uses a 2 W, 40 m rig with the coupler attached directly to the back of the rig.  Yes the rig and its associated wiring are the 'counterpoise' but since I've adjusted the system for minimum (I didn't say NO) current through the coupler the efficiency is high and everything is just as stable as when using a center fed arrangement.  This being the case, I have no reason to change no matter how often purist say the setup won't work ;-)

Some people like to carry their whole portable HF station in a small bag.  Others are content to pull it on a trailer.  To each there own but this a QRP forum.

Steve - AA5TB
Pages: [1]   Go Up
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!