Source of article:https://js1fvg.kabu.direct/?page_id=360
Author of the article: T.Kabu/JS1FVG
Translation method: DeepSeek + manual adjustment
Introduction
※ Initially, this was based on a previous document, but due to the extremely disorganized nature of the original and the subsequent discovery of numerous errors, I have decided to reorganize it completely using a correct methodology.
※ This refers to a half-wave antenna for end-to-end transmission.（EFHW）The first document aims to guide you through the entire process and ultimately achieve the final goal.Construction of the installation circuit board for a dedicated EFHW UNUN.。
After several decades, I rediscovered amateur radio and found that the popular FT8 mode is incredibly convenient – you can achieve DX communication (i.e., communicating with stations overseas) without even speaking!
However, to be honest, a mere 5-watt power output with a whip antenna simply wasn't enough. So I thought about using a full-sized dipole antenna – ideally one that was simple in design, such as a collinear antenna like the Zepp antenna. But all I had was a SAGA Electronics 7/21 MHz dual-band antenna, and I wanted to experiment with other frequency bands...
Thus, the recently popular "EFHW" (End Fed Half Wire Wave antenna) within the amateur radio community has caught my attention, and I have started to research and develop a construction plan.
EFHW's popularity stems primarily from its remarkable characteristic – it is said to achieve a near-unity standing wave ratio (SWR) across almost all frequency bands in the HF band.（SWR）It dropped to 1.x! But, when you actually try it, you'll find that some people have succeeded while others have failed, and there are significant differences in the various methods used. It's impossible to determine which method is correct.
However, during this process, I discovered a truly essential tool – or one that I couldn't resist buying: the "NanoVNA" vector network analyzer. This incredible device, which can accurately measure antenna characteristics, was simply unimaginable in the past, but now it can be purchased for around 200 yuan!
NanoVNA-H4
I didn't hesitate and placed the order immediately. The result was that it was incredibly easy to use! Even when adjusting the Saga electronic zip line antenna, I could clearly see the changes in SWR, which was already amazing. And don't even mention making my own coils or DIY antennas – this is definitely a must-have tool!
Now, let's move on to the main topic – discussing the core aspects of a true EFHW antenna (end-fed half-wave antenna).
As mentioned earlier, when discussing antennas in the HF band, many people immediately think of Zepp-style antennas. Of course, non-radial vertical whip antennas are also a good option – and with the availability of automatic antenna tuners today…（ATU）Coordinated electric telescopic whip antennas. In essence, these all fall under the category of "end-fed antennas."
However, the EFHW antenna that I am planning to build, which is commonly referred to as an "EFHW," differs fundamentally from the types mentioned above.
The reason for this lies in the fact that all of the antennas mentioned above essentially operate at a narrow resonant frequency – they meticulously tune the target frequency to that specific, narrow point. As an antenna becomes shorter (and therefore has less length), the SWR peak typically becomes sharper. However, the EFHW completely overturned this understanding: its wide SWR characteristics allow it to achieve extremely low VSWR values across the entire 1-30 MHz band – a truly remarkable antenna.
The SWR characteristics of the EFHW antenna using a Barron sample, after fine-tuning. A 5.1kΩ dummy load was connected. The yellow curve represents the SWR characteristic, as shown in the figure, where SWR is controlled at approximately 1.x within the range of 1.8MHz～28MHz.
Regarding the manufacturing process, as mentioned earlier, perhaps it's easy to make mistakes because everyone is working with a degree of uncertainty. Even if someone explicitly tells you "you should do this," and actually implements it, various factors often prevent achieving ideal SWR characteristics, ultimately leading to the awkward situation of "it must be a trick after all."
Even I, the person who wrote this article, initially thought with considerable doubt: "Is this really going to work?!" However, when I actually measured a SWR curve as flat as the one in the picture, I couldn't help but exclaim: "It actually works!!"
Therefore, because I myself had been deeply confused while searching for information about the EFHW antenna, in order to prevent others from repeating my mistakes and also as a personal reference, I decided to systematically document the true method of constructing an EFHW antenna.
Materials to prepare
The materials needed to make an EFHW are:
- Iron oxide magnetic core (FT140-43, FT240-43, or equivalent products);
- Copper wire (recommended polyurethane insulated wire)（UEW）And that's it. Just maintain insulation. Using insulating tape is also fine.)
- 47～200 pF capacitor (recommend using a high-voltage version with a voltage rating of several kV)
- Use a lead wire that is several meters to tens of meters long (as an antenna resonator, it is recommended to have a length at least equal to half the target frequency wavelength).
- Standard M-type connectors for connecting coaxial cables (recommended to use with female ends, and a combination of adapter + male end).
Furthermore, waterproof enclosures and other accessories for housing the "Baren" section also need to be prepared.
Furthermore, to measure and adjust the assembled EFHW antenna, the following equipment and tools are also required:
- NanoVNA (or a similar antenna analyzer – if you only have an SWR meter, adjustments will be very difficult)
In short, instead of debating whether or not to buy a NanoVNA, it would be more practical to simply purchase one and start building your EFHW antenna (laughs).
Additionally, in addition to the FT (ferrite toroidal) series, there are also FB (ferrite toroidal) systems and T (ring core) series options available. However, when actually winding the coils, the FT system provides an inductance value that is significantly higher than other series (the AL value of the FB series is good, but its size is too small, so it was not adopted).
Since the AL value (inductance coefficient) of the ferrite core directly determines its inductance characteristics, if you already have a ferrite core, it is strongly recommended to first determine its AL value and then select a model with parameters similar to those in the FT series.
| Model | AL value |
|---|
| T68-6 | 4.7 +/- 5 % |
| T200-2 | 12+/- 5% |
| FT140-43 | 885+/-20% |
| FT240-43 | 1075+/-20% |
For details, please see:https://toroids.info/
EFHW using the method described by Baron.
While various impedance matching methods are documented on overseas websites and by previous researchers, the industry-recognized gold standard is to use a primary 2-turn + secondary 14-turn configuration with a 1:7 turn ratio, resulting in an impedance transformation ratio of 1:49.
The initial turns of the primary and secondary coils are wound together in a bifilar manner. The secondary coil is then wound using the common practice of "W1JR winding," where it is wound to approximately half its length before being reversed. Finally, a capacitor is connected between the center tap of the primary coil and the center taps of both the primary and secondary coils – this creates an ideal EFHW balun. However, in reality, achieving this is almost impossible.
When actually building this EFHW Balun – or, more specifically, just the impedance matching section – depending on the ferrite core type, winding method, and capacitor capacity used, one of the following scenarios is likely: In the HF band, only a very low SWR can be achieved at frequencies below 20MHz or 25MHz, or the SWR across the entire band remains high.
If done casually, it often results in SWR that cannot be reduced, or extremely narrow usable bandwidth.
Initially, I tried using an FT240-43 magnetic ring with 0.4mm polyurethane insulated wire.（UEW）The coil was wound with a 2:14 turns ratio, and a total capacitance of 141 pF was added to the circuit. As a dummy load, I connected a 5.1 kΩ resistor on the secondary side and tested it using a NanoVNA. The results, as shown in the photo, only yielded acceptable SWR values in the HF low-frequency range.
Due to the large size and difficulty of use of the FT240-43, I decided to first test and verify using the FT140-43 magnetic ring. Compared to the FT240-43, the AL value of the FT140-43 is lower, so I needed to slightly increase the number of turns and adjust it to a 3:21 turn ratio (while maintaining a 1:7 impedance transformation ratio).
Due to its smaller size (FT140-43), it can significantly reduce the amount of cable required. If optimized for a specific frequency band, would this compact design be feasible?
As shown in the image, although the SWR near 21 MHz is slightly higher, the overall value remains below 2.0 – and thus, a fully functional EFHW balun was unexpectedly created, with performance even exceeding expectations.
After subsequent, arbitrary adjustments to the wire spacing, I was ultimately able to successfully fabricate an EFHW antenna (prototype) with extremely ideal SWR characteristics, as shown in the initial photograph.
However, this is just an example of "accidental success." Since the verification only used 0.4mm polyurethane copper wire as a temporary test material, it may be sufficient for QRP (low-power) operation, but it is clearly not reliable enough to achieve 50W or 100W power output for international communication. (Note: For high-power applications, you need to use thicker wires (such as 14-16 AWG) and ensure proper magnetic core cooling design, otherwise overheating may lead to performance degradation or even damage.)
Considering this possibility, my spare FT140-43 magnetic loop antenna came in handy. This time, I used a 0.6mm polyurethane copper wire and completely re-wound the EFHW balun according to a 3:21 turn ratio. However, testing revealed that while the winding method was identical, the SWR characteristics of the resulting balun were subtly different.
Each time a coil is wound, it produces different characteristic parameters. This low repeatability forces me to re-evaluate my adjustment methods.
First, try to wind the primary and secondary coils more tightly by using a bifilar winding technique on some of the turns.
When the SWR curve's valley range is lowered overall by adjusting, the high-frequency band (around 21 MHz) experiences a significant improvement in SWR, but the side effect is that the SWR above 30 MHz decreases.<The wideband characteristic of 1.5 GHz results in a reduced usable bandwidth.
By further increasing the winding density of the return path of the secondary coil and fine-tuning its position, the SWR in the 28 MHz to 29 MHz FM band was finally brought within an acceptable range.
These SWR values are quite ideal!
The key adjustment points are summarized below:
- Control of density in the primary/secondary bidirectional and interleaving section (red part)
- Control of spacing between the bi-directional parallel winding and the secondary winding after the return (blue section)
- Control of density for the secondary bypass after the return (green part)
- Control of spacing for the secondary loop after the return (yellow section)
- Capacitance (Increasing capacitance in Longnan will shift the good SWR range to lower frequencies) (Orange section)
1) Affecting the SWR characteristics near 21 MHz,
2)～4) Influencing the SWR performance near 28 MHz in conjunction with 1),
5) Increasing capacity will shift the overall resonant frequency towards lower frequencies, but also compress the bandwidth range for excellent SWR.
First, reduce the SWR around 21 MHz by tightening the spacing between 1) and loosening the spacing between 2). Then, adjust the spacing of 3) and 4) to find the optimal balance point that minimizes (widens) the SWR in the same frequency band.
Even though we have already created a high-performance EFHW antenna, adding a specially designed M-type connector will cause the carefully tuned characteristics to degrade.
After re-measuring the SWR characteristics after temporarily installing the M-type connector, it was found that although the previous fine adjustments had reduced the values, the overall SWR had significantly increased, and the usable bandwidth had further narrowed.
Initially, I mistakenly believed that the problem was caused by the coil shifting on the ferrite core. However, after keeping the 5D-V coaxial cable unchanged and retesting with the M-type connector removed, the SWR characteristics remained ideal.
Adding just a single M-type female connector will significantly alter the SWR characteristics, so it is not recommended to use an M-type female connector as the coaxial cable end for an EFHW balun. Although there were initial suspicions, subsequent investigations confirmed that this was indeed due to the inherent properties of the connector itself.
I recommend using a Type M adapter (for use with repeaters) in combination with standard Type M connectors.
This is why it's not recommended to directly install a M-type female connector on the Balun – it will significantly alter its SWR characteristics. While the converter is more expensive than a standard connector, considering that cheap M-type connectors can severely degrade carefully calibrated SWR characteristics, this investment is actually very worthwhile.
Ultimately, we successfully created a very ideal EFHW antenna using the specified components:
The following diagram shows the design for this EFHW balun, which uses an FT140-43 magnetic ring and has a primary of 3 turns and a secondary of 21 turns.
Simplified design diagram of EFHW Balun using FT140-43 magnetic loop ↑
However, this design only uses a 5kΩ resistor to simulate the impedance characteristics of the EFHW antenna's radiating element. In actual deployment, further adjustments are necessary based on the specific site environment. As is well known, factors such as the distance between the feed point and the conductor, the height of the antenna relative to the ground, etc., significantly affect the impedance characteristics. Therefore, the final optimization must be performed by matching the EFHW assumptions with the actual site conditions.
When the feed point position is lower, the antenna's actual impedance will decrease significantly (e.g., from a theoretical value of 5 kΩ to 2.5 kΩ). Therefore, it is recommended to use a 2.5 kΩ dummy load instead of a 5 kΩ load during testing for preliminary adjustment. (Note: This method may achieve surprisingly good SWR optimization results; it is strongly recommended to verify this experimentally.)
It is important to note that a lower feed point means the antenna resonator and ground are also at a lower height, so attention must be paid to the radiation angle and harmonic frequency. However, the advantage of EFHW antennas lies in not having to worry about resonance frequencies, so in practice, it is sufficient to focus on the radiation angle – of course, this only applies when the antenna resonator is horizontal…
Installation and adjustment of EFHW
First, the antenna resonators are set to a length of 20.4 meters, which corresponds to a half-wavelength of approximately 42.857 meters at a frequency of 7 MHz, taking into account a shortening factor of about 0.96. Then, two resonators of this length are prepared, so that they can operate at 3.5 MHz. The experiment will be conducted in an open area similar to a sports field.
When actually implementing, approximately 5 miles of support lines are pre-allocated at both the feed point and the antenna ends, resulting in a total length of about 50 meters. The EFHW antenna is deployed with a ground clearance of 3-5 meters to ensure that it is not affected by any other medium besides the ground, and its SWR characteristics are measured using a NanoVNA. No ground plane or balun is used.
First, only connect the 20.4m antenna, and monitor it in real-time using a NanoVNA. Simultaneously, repeatedly adjust and test by employing techniques such as end-of-antenna return loss measurement.
After multiple adjustments, the length was finally reduced to approximately 1 meter to 19.4 meters, and it achieved operational status in the 7/14/21/28 MHz frequency bands, working in conjunction with the antenna tuner.（ATU）It can be used easily by everyone.
Based on the test results above, another root with a 20.4m coil was shortened by 1m, resulting in a total length of 38.8m. The SWR characteristics were then measured using a NanoVNA.
The test results showed that in version 19.4, the 3.5 MHz frequency band, which was completely unusable, actually exhibited a significant decrease in its SWR value.
When using the FT140-43 magnetic loop balun with a 40m antenna, SWR (Standing Wave Ratio) testing in the 1-30MHz frequency range showed that while good matching was achieved across all amateur bands (3.5/7/10/14/18/21/24/28MHz), there was a slight overall tendency for the resonant frequency to be slightly higher than optimal. To address this, the excess 1m of antenna length cut during adjustments was used as a matching short circuit. By precisely trimming its length, the resonant frequency could be lowered, resulting in improved SWR characteristics.
Ultimately, it was confirmed that an SWR characteristic consistent with theoretical calculations could be achieved as long as the feed point height reached approximately 5 meters. However, for lower frequencies (such as those in the 3.5 MHz band), the radiation angle would inevitably become higher (this phenomenon is consistent with basic antenna principles: the ratio of the tower height to the wavelength directly determines the radiation angle; at a height of 5 meters, the value of λ/4 is approximately 7 MHz, but only λ/8 at 3.5 MHz, which causes the radiation beam to tilt upwards).
As is well known, when the length of an antenna exceeds 1/2 wavelength, its radiation pattern will no longer be omnidirectional. Therefore, if directional transmission is required, it is essential to pay particular attention to the range of the antenna – the direction in which the electromagnetic waves are most strongly radiated must remain consistent with the extension direction of the antenna.
After completing all the experiments, when I was using FT8 mode to call CQ on various bands, I suddenly received a response from an Italian radio station and successfully established communication. Subsequently, I not only managed to communicate with domestic stations but also with several overseas stations – this EFHW antenna, which can operate on multiple bands with just one wire, is truly amazing!
In reality, the experiment used AWG28-specification (approximately 0.08 square millimeters) very fine wires as oscillators. Although the SWR bandwidth for each frequency band became quite narrow, there were no problems when transmitting at 50W power. However, it was crucial to carefully control the tension; otherwise, the wires could easily break. (In fact, the initial breakage started from the connection points of the terminals.)
The plan is to: first, observe changes in SWR after upgrading the coaxial cable to at least AWG20 specification (approximately 0.5 mm²); second, use the two FT240-43 magnetic loop coils that are currently in stock for testing; and third, consider adding a coil in the 19.4m version to attempt to achieve similar support for 3.5MHz, as was done in the 38.8m version (although the top-loading configuration may cause voltage problems) – these experiments are ongoing.
This report not only involved antenna-related equipment, but also comprehensively utilized a variety of tools, from welding guns to parts boxes. The following is the report that sparked discussion (?!): EFHW antenna construction and testing results.
闽ICP备2021006864号-7
闽公网安备35020602002794号