Patch Feed Questions and Answers by James Miller G3RUH 2020 Jul 05 I receive many questions about patch feeds, dish antennas and related matters. This document summarises the replies. Source: http://www.jrmiller.online/products/patfaq.txt ___________________________________________________________________________ What is a Patch Antenna? ------------------------ A patch antenna is essentially a radiating element spaced fairly close to a ground plane. Often the embodiment looks like a circular band-aid has been stuck onto a flat sheet, hence the name 'patch'. The driven element can be electrically small, a fraction of a wavelength; a familiar example is the consumer-grade GPS antenna. Often the implementation uses printed circuit board techniques, usually with fibreglass dielectric. The driven element is sometimes circular, though square and linear shapes are used too. The driven element can also be full sized, about 1/2 wavelength in diameter, and air spaced from the ground plane. This is what we use in our S-band patch. How does a Patch Work? ---------------------- A patch is usually fed at the edge, or a little way in from the edge. If you look each side of the axis of symmetry, you can view it as two slot dipoles side by side. Alternatively, consider it as a resonant cavity with open sides that radiate. The relative phase of the currents in the two 'sides' can be controlled using a capacitor to give circular polarisation. What about the ON6UG/G3RUH Patch? --------------------------------- Our S-band design is full size which narrows the beamwidth to an angle very suitable for f/d 0.3-0.45 dish illumination. It has negligible sidelobes which keeps the efficiency high, and it doesn't 'look' over the rim of the dish at the hot and noisy ground. The dielectric is air, which has no losses and this also helps to make it a quiet feed. The bandwidth is about 200 MHz, so it rejects out-of-band signals that can desense low-noise converters with their generally poor input selectivity. This also keeps the noise down. Polarisation is almost perfectly circular. This patch feed, plus a deep, slightly underilluminated dish result in good gain, low sidelobes and very low noise, allowing the potential of contemporary ultra-quiet downconverters to be fully realised. http://www.jrmiller.online/products/patch.html Can it be used for Transmit? ---------------------------- Yes. The patch has been tested as a transmit antenna up to 150 watt - the limit of the PA used. Anyone sucessfully exceeding this, please let me know. [2019] Many are using it for uplinking to the Es'hail-2 (QO-100) geostationary satellite. How do you make the patch? -------------------------- The patch feed assembly contains 13 parts; seven of these are standard components like nuts and bolts; the remaining six are specially made items, such as the reflector plate. These are manufactured in bulk by a commercial engineering workshop using CNC machines. Thus the parts are all the same within tight tolerances, which ensures that each patch works identically. These precision parts are not cheap ... the reflector plate and rim, for example, costs about $40. Before assembly, parts are individually checked for accuracy, and are then degreased. Thereafter they are handled only with gloves. Each patch is jig assembled, checking key dimensions as work progresses. A soft brush and compressed air are used frequently to keep the product meticulously clean. After adjustment and testing the patch is stored in a sealed bag where it remains until you receive it. It takes about one hour to manufacture a patch, including time to receive the order, log it, do paperwork, send emails, pack, label and despatch the parcel. What is the Gain of the Patch? ------------------------------ The patch feed has a gain of 8.5 dBi and is circularly polarised. The gain has been determined in two ways. The first uses the measured polar diagram, from which the gain can be calculated to be 8.8 dBi. The second was a direct measurement using an antenna range. This was performed at the Amsat-UK Satellite Symposium 2001 held at the University of Surrey in Guildford, England. The gain was compared with that of a standard horn antenna whose performance is known from its dimensions. This gave 8.5 dBi for the patch, and is the adopted figure. At the same session, a large number of other antennas were also measured and gave reasonable results, so lending credibility to the reference antenna and to the measurement technique. Does the patch gain add to the dish gain? ----------------------------------------- No. The gain of the feed is not added to the gain of a dish. The purpose of the feed is to "light-up" the dish (in transmit mode). In other words provide RF energy over the total surface of the dish. In receive mode, the purpose of the feed is to receive all possible RF energy from the dish, without receiving spurious energy from beyond the dish edges. The gain of a dish is set by the area of the dish. The theoretical maximum for a large, perfectly lit circular aperture is given by G = 4*PI*area where the area is expressed in wavelengths. Illumination is never perfectly uniform though; dishes have finite size, sidelobes, some losses and so on and the actual dish gain is typically 45-65% of that maximum. Can you use the Patch on its own as an Antenna? ----------------------------------------------- Yes. The patch feed can be used as a regular 8.5 dBic antenna in its own right. In fact a RHCP version of the patch could receive error-free telemetry from the AO-40 (P3D) satellite at apogee over a large azimuth. Note that the standard patch polarisation is LHCP; when installed in a dish the polarisation will be reversed to RHCP, conventional for amateur radio satellites. Must I use the G3RUH 60cm dish? ------------------------------- Absolutely not!! The patch feed can be used with any size dish. The feed polar diagram suits dish f/d (focal_length/diameter) ratios in the region of 0.3 to 0.5 . You can also use it with f/d ratios larger than 0.5, but the dish is then overilluminated which reduces gain and increases spillover noise because the feed can see the thermally hot ground behind the dish. That does not mean that it will not work, simply that it works more optimally with a smaller f/d ratios. You can use the patch feed with ex-TV offset fed parabolic dishes. These typically have an f/d around 0.6, and the above caveats apply. I want the Patch with an SMA Connector -------------------------------------- An SMA connector is small and has different mounting arrangements than the N-type used. So the reflector plate design would have be to be changed. This is uneconomic for small quantities. Simply use an N to SMA adaptor. SMA connectors are not strong enough to support the weight of a downconverter under adverse weather conditions. Also most SMA connectors are not water resistant. For these reasons an SMA was not used in the final production. Don't the feed support legs shadow the dish and reduce gain? ------------------------------------------------------------ No. In the G3RUH dish+patch the legs are 1/20th wavelength in diameter at S-band and thus are effectively invisible to the patch. There are far greater sources of performance variation, such as propagation and spacecraft spin induced amplitude variations. Even if shading were an issue, it is well under 1% ! Paint your Patch? ----------------- There is no electrical requirement to paint the patch feed. However you might wish to do this for aesthetic reasons. Do not allow paint inside the patch, or on the transparent cover. Unscrew the M4 domed nut to remove the cover. Take care not to lose the washer, and don't overtighten when replacing. The dish can also be painted. * Do NOT use paint that is loaded with metal * Do NOT use paint that that can crack or collect dust. * DO use only a thin coating of paint Books ----- Patch antennas don't seem to figure much in books, even though there are many papers. Kraus "Antennas" 2nd edition ISBN 0-07-035422-7 discusses the patch in general terms, but for practical details you should review the IEEE Transactions on Antennas and Propagation. Related Topics ============== My antenna with feed and an downconverter hears nothing -------------------------------------------------------------------------- 1. Is the antenna working? Find out by doing the Sun noise test detailed in G/T stuff further down. If you cannot hear at least a 1 dB increase in noise when you point your creation at the Sun compared with overhead, your system is deaf. Most likely cause is the downconverter/cabling; then the feed itself. 2. Were you on frequency? Make up the little S-band marker beacon by G0MRF as published in many Amsat magazines during 2001. http://www.g0mrf.com/source2.htm 3. Was it pointed at the spacecraft? Big antennas have small beamwidths ... 4. Was the spacecraft turned on? See published schedule ... I have been given a dish ... what's its focal length? ------------------------------------------------------ You can calculate the focal length of a parabolic dish by measuring its depth and diameter. Let: D = diameter of dish d = depth of dish Then the focal length is given by: f = D^2/(16*d) The shape of the dish is described by the ratio of focal length to diameter, written "f/D". From the above equation: f/D = D/(16*d) Many dishes have a reinforcing rim which is not part of the parabolic section, but which makes the overall diameter larger than the electrical diameter. In these notes, "Diameter", "depth" and "focal length" refer to dimensions in the parabolic part. ------ Example: I have a dish 90cm in diameter and it's 161mm deep. Using the formulas: f = D^2/(16*d) = (900 x 900)/(16 x 161) = 314 mm [focal length] f/D = D/(16*d) = 900/(16 x 161) = 0.35 [f/D ratio] ----- Tip: A simple practical way to find the focus of any dish is to stick a few tiny mirrors on the surface, aim the dish at the Sun, and use a small piece of white paper to find where the reflections converge. You can also make small reflective squares from aluminium cooking foil. Scotch-tape (Sellotape) them to the dish. What's all this G/T stuff? -------------------------- G/T stands for 'Gain-to-Temperature ratio' and is concise measure of the actual performance of a receiving system including the environment, antenna and receiver. The more Gain the better (=more signal) and the less Temperature the better (=less noise). So, their ratio reflects both of these parameters in a 'bigger number = better' way. With a dish antenna, highest gain is achieved by using the proper shape for the dish, and a well designed feed to illuminate it. Lowest noise is achieved with a quiet environment such as is found looking up at the sky, a quiet downconverter properly installed, low feed spillover (so feed sidelobes do not see the ground) and low sidelobes from the dish (so it doesn't see ground and/or man-made terrestrial noise). Based on lower frequency experience where noise is always very high, it is often thought that gain and related signal strength are all that matter in microwave systems. This is incorrect; high signal AND low noise are what matter; not the one or the other, but both together. Amateur systems are often diminished not so much by reduced gain as by increased noise. More noise is /very/ easy to introduce through poor design, components and installation. How can I measure G/T of my System? ----------------------------------- You need a signal of known strength and a noise environment the same as you intend to use. Happily both of these exist, and are free. The Sun is your signal source, and your environment is already there - space, of course. You point the antenna at the sky and measure how much noise comes out of the receiver. Then point the antenna at the Sun and measure the noise once again. The ratio of these two noise powers is called the Sun noise increase, and is directly related to G/T. Call this power ratio Y, then, for antennas with a beamwidth exceeding the Sun's diameter: Y-1 G/T = --- I where at S-band I ~ 0.5 and depends on the noisiness of the Sun (solar flux), which varies from day to day. You can obtain its values now and for the recent past via the reference: ftp://ftp.swpc.noaa.gov/pub/lists/radio/45day_rad.txt These tables actually list the solar flux at 2.695 GHz (typically 120, but can vary a lot), which needs adjusting slightly to 2.4 GHz. Doing that one obtains: 235 G/T = (Y-1) ------------ SFU @ 2.7GHz For example, the G3RUH 60cm dish plus patch feed plus Kuhne MKU-24 downconverter on 2001 July 25 measured a Sun noise increase of 2.44 dB. Thus Y=10^(2.44/10) = 1.75. On that day the Solar flux for 2.7 GHz was 129, and so G/T = 1.37 (units are K^-1). Measuring power using the AF output of a receiver requires the AGC not to be activated. Since the S-meter is usually driven from the AGC line, when the meter flickers, AGC is working. Lower the RF gain or insert a small attenuator at the radio input to force the S-meter to stop reacting. Some radios, but not many, have an actual AGC on/off switch. On the other hand, a Sun noise increase of 2.4 dB is easily audible, and may be all you need to know. You must use a moving coil meter on its AC range and, because the pointer jumps about, some skill is needed to take a mean reading. Beware of taking measurements when the Sun is low in the sky, as the antenna will pick up ground noise which will give you optimistic readings. What can I do with G/T? ----------------------- G/T is a measure of system performance, and if you check it from time to time, you can tell whether your S-band system is working consistently, or whether there's a problem. So, continuing the above example, if a Sun noise increase is not 2.4 dB, give or take say 0.2 dB, on any particular day, one would suspect a problem. You can also estimate your system's noise temperature. For example, a G3RUH 60cm dish has a gain at S-band of about 21 dB, or 10^(21/10) = 126. If G/T=1.37, then T = 126/1.37 = 93K. This would be made up from the converter noise (45K), antenna noise (maybe 20K) and sky noise (maybe 30K). If you make some changes to your system, maybe a different converter, dish or feed, G/T will indicate whether you made an improvement or not. For detailed results of a Sun noise test with a 60cm dish+patch over a month or so, see: http://www.jrmiller.online/products/figs/sun.zip Original 2002 Mar 11 Updated 2018 Jan 28 - added power handling info. URLs 2018 Nov 22 - revised noaa link 2019 Apr 30 - added Es'hail-2 QO-100 2020 Jul 05 - URLs updated