How does height above ground change the pattern?

Ground reflection shapes the elevation pattern. A horizontal antenna at half a wavelength height has its main lobe at about 30 degrees above the horizon, and raising it lowers the take-off angle while adding extra lobes.

How accurate is the simulator?

It uses a simplified wire simulation of the same kind as NEC, with an idealised ground and no mast, feedline or nearby objects. It is good for comparing shapes and polarisation, but verify real designs with a full NEC tool and a measurement.

GazLabs › Homebrew Antennas › Pattern simulator

Antenna Radiation Pattern Simulator

Q: Does the hentenna radiate horizontally or vertically polarised?

Mounted upright it radiates almost entirely horizontally polarised signal. Turned 90 degrees on its side the same loop radiates almost entirely vertically polarised signal, and at 45 degrees the signal is a mix of both.

See the predicted radiation pattern of a homebrew antenna and whether it behaves as a horizontal or a vertical antenna, then tilt, rotate or re-feed it to watch the polarisation change.

Orientation comparison

Link to these exact settings

Collinear on a mast: why the SWR goes high

Most VHF and UHF collinears, such as the stacked 5/8-wave types, are fed at the bottom and have no radials. The outside of the coax and the mast below take the place of the missing ground plane, so they become part of the antenna. A mast that is a different length each time, or a cable that runs down it, then changes the SWR. This check shows that effect for an antenna on a mast at the height you choose, using the frequency and ground set in the simulator above.

Frequency (MHz)

Stacked 5/8-wave sections

Height of the antenna base (m)

Mast material

Mast length below the antenna (m)

Choke impedance (Ω)

Calculating…

How to read this: the antenna is assumed to be matched perfectly when the coax is choked, so the SWR shown is only the change caused by the mast and cable. The model is a simplified thin-wire stack with idealised phasing coils, so treat the shape of the curve and the benefit of a choke as the lesson, not the exact SWR of any particular product.

What to do about it

Choke the feed. Fit a ferrite choke or a few turns of coax wound into a coil just below the antenna so the shield cannot carry current down the mast. The choke designer works out the turns for your frequency and cable.
Mount it at the top. Put the antenna above the end of the mast, and where the mast must continue past it, use a side arm that holds the base at least half a wavelength clear (about 1 m on 2 m, 35 cm on 70 cm).
Keep the cable off the mast. Take it away at right angles for a metre or so before it drops down, instead of taping it along the metalwork.
Check the connectors and cable. Water in a roof-mounted N or PL259 connector, or a cable with damaged braid, looks exactly like an antenna problem.
Tune on the mast. If you must change the mast, re-measure with an analyser at the working height, because the reading on the bench does not carry over.

SWR and reflected power at this frequency

Uses the frequency entered above. Enter the SWR you measure at that frequency (1:1 is a perfect match) and the transmitter power.

Frequency (MHz)

SWR at the antenna (x:1)

Transmitter power (W)

This assumes a lossless feedline. For cable loss and a full reference table use the SWR calculator.

Reading the plots

Elevation pattern: a side view. The horizon runs left to right and the zenith is straight up. The tall lobe shows the angle above the horizon at which most of the power leaves the antenna, called the take-off angle.
Azimuth pattern: a plan view from above at the chosen take-off angle. A circle is omnidirectional, a figure of eight is bidirectional and a single lobe is directional.
Blue and orange traces: the horizontally and vertically polarised parts of the signal. The green trace is their sum, which is what you would measure with a perfectly matched receiving antenna.

Why orientation changes polarisation

Polarisation follows the direction of the current in the antenna. Current flowing along horizontal wires gives horizontal polarisation, and current flowing along vertical wires gives vertical polarisation. A flat-top dipole is therefore horizontal and a vertical dipole is vertical, with a 45° slant giving a mix of both. A loop is set by its feed point: feeding the middle of a horizontal side drives currents in the top and bottom wires, which radiate horizontally, while feeding a vertical side makes the vertical wires do the work.

The hentenna works the same way. Upright, the horizontal wires carry the radiating current and the vertical wires largely cancel each other, so it behaves as a horizontal antenna. Rotate it 90° on its side and the same currents now flow vertically. In both cases the loop still radiates broadside to its plane, so the beam direction does not change.

How it works and how far to trust it

The simulator solves for the currents on a simple wire model of the antenna using a method-of-moments calculation, the same family of method as the NEC programs, then adds the reflection from the ground using Fresnel coefficients for the ground type you choose. It runs entirely in your browser and nothing is sent anywhere.

Limits: the ground is an idealised flat plane, and there is no mast, feedline, balun, nearby building or tree in the model. The J-pole is treated as a vertical half-wave section, the end-fed half-wave ignores its transformer and counterpoise, the off-centre-fed dipole is modelled as a single wire cut for the frequency you enter, and the stacked collinears are driven in phase with equal power and have no phasing network or mast in the pattern plot, the small magnetic loop is lossless in the model, so its real gain is lower by the efficiency shown in its calculator. Use the results to compare shapes and polarisation, then check a real design with a full NEC tool such as 4nec2 or xnec2c and with measurements.

As a sanity check the engine reproduces textbook values: a free-space half-wave dipole gives about 2.15 dBi, a full-wave loop about 3.1 dBi, and a horizontal dipole half a wavelength above perfect ground peaks at roughly 30° elevation. For the hentenna it predicts about 5 dBi (around 3 dBd) in free space, which is lower than the 5 dBd often quoted for the design, so treat quoted gain figures with some caution until you have measured one.

Pattern simulator FAQ

Does the hentenna radiate horizontally or vertically polarised?

Upright, it is almost entirely horizontal. On its side it is almost entirely vertical, and tilted to 45° it is a mix. Try the presets above.

Why does a dipole have a different pattern at different heights?

The ground reflects part of the signal, and that reflection adds to or cancels the direct signal at different elevation angles. Raising a horizontal antenna lowers the main take-off angle and adds more lobes.

Why is a vertical antenna weak at the horizon over real ground?

Real ground absorbs and reflects low-angle vertically polarised signals, so far-field gain at the horizon falls towards zero even though the antenna is omnidirectional. A better ground, or more height, helps.

Which polarisation should I choose?

Match the stations you want to work: horizontal for SSB, CW and weak-signal modes on VHF and UHF, vertical for FM and repeaters. See the polarisation guide for compromise installs.

Back to the homebrew antenna guide and advisor, the calculators or the hentenna calculator.