Practical Guide to Operating AO-40 - Clarifications & FAQs
Created 26 November 2001
Updated 27 November 2001
Following some feedback from several readers, I've compiled the following clarifications on the original article appearing in RadCom in November 2001 and December 2001 issues. Some of the descriptions here are rather more theoretical than practical, but they can help explain the thought process behind some of the design criteria.
The descriptions here are part of a continuing learning process: what you see here are the successful results of lots of other bent (now scrap) metal and failed experiments!
I have a North-facing garden. Can I work AO-40?
Can you explain the matching section in more detail?
Can I protect the matching section against the weather with sealant?
Is the ‘through-plating' really necessary?
The helix is rather fragile. How can I make it more rigid?
How critical is the position of the Helix relative to the dish?
The URL for AMSAT-DL isn't working. How can I find out the schedule?
I have a North-facing garden. Can I work AO-40?
AO-40 orbits the earth around the equator, and so it will always appear to the South from the UK. When it's above the horizon, in the UK it spends about 70% of the time above 20 degrees elevation, and about 30% of the time above 30 degrees. During these higher elevations, conditions are generally far better in any case, with the spacecraft's antennas pointing more towards the Earth (i.e., a lower squint angle).
If you only have a view to the Southeast or East-South-East, you will have no problem working Africa, Asia, Australia and Europe from the UK.
Alternatively if you only have a view to the Southwest or West-South-West, you will have no problem working Africa, North America, South America and Europe from the UK.
Beyond this, you may be pleasantly surprised what you can achieve even if your garden is North facing. A bit of trigonometry is all you need to work out the satellite visibility. As many readers will know, I operate regularly from my tiny 8' by 10' basement garden in the front of my flat, much to the interest of both the passers by and the neighbours.
How to calculate the lowest elevation you can work AO-40 in a given direction.
Remember that it's perfectly possible to simply point the antennas out of your window or balcony too.
Can you explain the matching section in more detail?
The matching section is the first quarter turn (not a quarter wavelength) of the helix. It is really a tapered microstrip transmission line, and not meant to radiate.
The theory goes like this: an 8mm wide piece of thin conductor, or ‘shim', mounted 1.6mm over a ground plane (the reflector in this case) with air dielectric has a 50 ohm impedance, and is frequency independent. A quarter turn later, it is 8mm higher (each turn is 32mm from its adjacent turn), giving a total distance from the ground plane of 9.6mm. The 8mm wide shim has an impedance of about 140
ohms at 9.6mm, which is the characteristic impedance of an axial mode helix.
For want of a better description, it is fabricated like a helter-skelter, soldered to the underside of the copper wire.
For an accurate match, the 1.6mm distance (use some standard 1.6mm PCB as a gauge) and the 8mm shim width are the most critical dimensions, followed by the 9.6mm
distance which you can easily be +/- 1mm or so out without making much practical
difference to the matching.
The dimensions of the helix after the first quarter turn matching section are less critical.
The reality is that you can pick up AO-40 without any matching section at all. However the aim of the design was to present a reproducible design which would present an accurate match without having to resort to any test equipment or adjustment of any kind. I have built about a dozen helices based on this design, and all have presented VSWRs between 1.1:1 and 1.2:1. I have had a report from a Belgian reader who attached the helix to a network analyser and achieved a return loss of 26dB, equivalent to a 1.1:1 VSWR.
There's some more pictures of the feed and matching section here.
Can I protect the matching section against the weather with sealant?
After spending all that effort carefully making the matching section based on an air dielectric, it's a shame to break it with any sealant, which will have a significantly different dielectric constant, and therefore present a big impedance bump.
Spraying a couple of very thin coats of PCB lacquer (or colourless spray paint from your local DIY store) onto the surface will prevent copper corrosion.
While you're at it, a couple of very thin coats to the matching section itself
and copper helix will be fine too.
If you feel it's necessary, I've heard reports of using a 1.5 or 2 litre plastic fizzy drinks bottle, cutting off one half, in a bid to prevent too much water getting in. The aim here is to prevent a deluge of water getting in, but not to make it watertight. Indeed, it's better to allow the antenna to breath, so having a hole in the base of any water sheltering arrangement will stop condensation building up.
Is the ‘through-plating' really necessary?
The helix will receive AO-40 without any through plating. In the quest for perfection, this will improve the matching. It only takes a moment to add. Again, after spending all that time making the match, you might as well make a proper job of it!
There's no penalty for using double-sided PCB either. If plated through, I'm sure SRBP PCB would work too.
Alternatively, if you want to use solid aluminium for the
reflector, for example, that will work too. I had concerns when building the
feed about how difficult it might be to cut aluminium to the required size, and
so I resorted to PCB material.
What's the PCB spacer for?
The spacer is designed to make the Teflon insulator at the base of the N type solder pip level with the copper reflector surface. Generally the distance from the panel mount to the Teflon base is 3.2mm
- that's the width of the 1.6mm reflector itself and the 1.6mm PCB spacer.
Wrapping the spacer in the copper foil and pushing the foil through the holes to through-plate enhances the matching.
The helix is rather fragile. How can I make it more rigid?
The correct way to do this is to place a small diameter rod (l/10 maximum), either conducting or non-conducting, directly through the centre of the helix and use one or two small standoff insulators to make it rigid.
One suggestion by David G0MRF is to use some threaded rod (available from warehouse style DIY stores), or alternatively a very long screw, which allow easy attachment to the reflector and any standoffs. See
http://www.g0mrf.freeserve.co.uk/ao-40.htm and
my quadruple helix for ideas on centrally located booms.
Don't leave the helix on a former of any kind: either the loss of the former or its dielectric effect (or both) will affect the helix's performance. The design assumes an air dielectric.
Another method to avoid is using an insulating rod offset from the centre of the helix. This will have an asymmetrical dielectric affect on the helix, acting as a lens so that the main beam is tens of degrees off axis. As the
waves are bunched closer together on the side of the helix closest to the dielectric, the effect on the performance is very difficult to predict, but empirically I can assure you it's not recommended.
The effect is like the dielectric effect in coax cable giving a velocity factor
which bunches the waves closer together within the coax compared to being in
free space.
How critical is the position of the Helix relative to the dish?
At 2.4GHz, it's not particularly critical, certainly to within two or three centimetres. If we were attempting 10GHz or 24GHz, things become rather more crucial.
I have a 90cm (or 1m or 1.2m) offset-fed dish. Can I use this instead of the 60cm dish? How many turns should I use on the helix?
Up to a certain point, the bigger the dish, the louder you'll hear the satellite. Like the
60cm dish, simply remove the existing LNB from its fixture and screw on the helix feed. You may need to do a bit of modification to the fixture so that the helix can be positioned accurately like that demonstrated on the
60cm dishes I've used. Larger dishes have smaller beamwidths, so you'll need to point the dish more often and more accurately, which can make locating the satellite more difficult.
In general for offset-fed dishes, the limitations of the design dictate that about 5.25 or 6.25 turns are optimum for all of them. Prime fed dishes (where the feed is in the middle of the dish) don't have such design limitations and so you'll need to calculate the number of turns required. If you'd like to check your particular dish's optimum number helix feed turns, I refer you to
these calculations.
The URL for AMSAT-DL isn't working. How
can I find out the schedule?
Try the mirror site instead: http://www.dj1km.de/amsat-dl/journal/adlj-p3d.htm
I'm having difficulty finding the Transystem AIDC 3733 downconverter in the
UK. How can I purchase one?
I am importing a number of these units, already modified and tested, and with the N-adapter. The cost of the modified downconverter and N adapter will be in the region of £80 plus VAT and P&P, and no, I won't be making any money on it (the cost price
for both downconverter and N adapter is US$110 plus shipping from the US). Please contact me if you're interested.
Update: These all sold out in a week, but if there's sufficient interest I'll import some more.