The WR650 waveguide is an ideal choice to use with the SETI Horn of Plenty Hydrogen Horn Radio Telescope.
I have noticed that 150mm x 50mm aluminium box section is easily obtainable in the UK. A 400mm long section costs less than £20 at the current time, and hence is massively below the cost of a commercially sold WR650 aluminium rectangular waveguide, which can be up to £500 in price. Steel oil tins are about the same dimensions as a WR650 waveguide, and are cheap/free. As a result they have been used by many amateurs in the past to make homemade WR650 waveguides.
I therefore decided to carry out a theoretical analysis of the pros and cons of choosing one over the other to construct such a waveguide. Most obviously, the oil is the correct dimensions and the aluminium box section is a bit smaller than ideal. However, the aluminium is far more robust, does not rust, and the aluminium has far better conductivity than the steel of the bin. So, in practice, what difference does this make?
For hydrogen-line work at 1420.405 MHz, a 150 mm x 50 mm rectangular conductor is actually quite a reasonable waveguide size.
The dominant TE10 mode cutoff frequency is:
fc = c / (2a)
where a is the wider internal dimension.
For a 150 mm wide guide:
- a = 0.15 m
- fc ~= 1.0 GHz
Since 1420 MHz is above cutoff by about 42%, propagation is comfortable and the guide should work well in TE10 mode.
150 mm x 50 mm aluminium box section
Advantages:
- Good electrical conductivity.
- Predictable dimensions.
- Mechanically rigid.
- Easy to drill for probes and mounting hardware.
- Low RF loss at 1420 MHz.
The next mode (TE20) cuts off at approximately:
fc(TE20) = c / a ~= 2.0 GHz
so at 1420 MHz only the dominant TE10 mode propagates, which is desirable.
Rectangular oil tin
A typical oil tin might be roughly 160 mm x 80 mm internally (varies by manufacturer).
Advantages:
- Cheap.
- Already forms a closed metal cavity.
- Often adequate for feed experiments.
Disadvantages:
- Thin steel has higher RF loss than aluminium.
- Dimensions vary and corners may be rounded.
- Seams can introduce small losses.
- Less mechanically stable.
At 1420 MHz the difference in loss between a short oil-tin feed and a short aluminium-box feed is usually small compared with other losses in the receiving system, but aluminium is generally the better engineering choice.
For a hydrogen-line feed
If you are building a waveguide feed for a dish:
- The 150 mm x 50 mm aluminium box section is likely the superior choice.
- It provides a well-defined TE10 mode.
- A probe feed can be placed approximately one-quarter guide wavelength from the back wall.
- The 50 mm height is sufficient to support TE10 while keeping higher-order modes suppressed.
Expected performance
For a waveguide length of only a few hundred millimetres:
- Aluminium box section: very low insertion loss (a few hundredths of a dB).
- Oil tin: slightly higher loss, but probably not enough to noticeably affect hydrogen detection unless your system is already near its sensitivity limit.
For radio astronomy, where stability and repeatability matter, I would choose the aluminium box section every time. It is closer to a purpose-built waveguide and should give more predictable illumination of the dish and easier optimisation of the probe position.
One thing to check before purchasing is the internal dimensions. Many 150 mm x 50 mm aluminium box sections have wall thicknesses of 2-5 mm, giving internal dimensions closer to 140-146 mm by 40-46 mm. Even so, the TE10 cutoff remains around 1.03-1.07 GHz, which is still comfortably below 1420 MHz.
If the quoted size is the external size of the box section, the internal dimensions are what matter for waveguide performance.
Example:
- 150 x 50 mm external, 2 mm wall -> 146 x 46 mm internal
- 150 x 50 mm external, 3 mm wall -> 144 x 44 mm internal
- 150 x 50 mm external, 5 mm wall -> 140 x 40 mm internal
The TE10 cutoff frequency depends only on the internal width, a:
fc = c / (2a)
Approximate values:
| Internal width | TE10 cutoff |
|---|---|
| 150 mm | 1.000 GHz |
| 146 mm | 1.027 GHz |
| 144 mm | 1.042 GHz |
| 140 mm | 1.071 GHz |
At 1420.405 MHz:
| Internal width | f/fc |
|---|---|
| 150 mm | 1.42 |
| 146 mm | 1.38 |
| 144 mm | 1.36 |
| 140 mm | 1.33 |
All of these are comfortably above cutoff and will work well for a hydrogen-line feed.
The main practical effects of reducing the internal width from 150 mm to 140-146 mm are:
- Slightly shorter guide wavelength.
- Probe position moves a few millimetres.
- Slightly different feed illumination pattern.
- Negligible change in sensitivity.
For radio astronomy, the difference between a true 150 x 50 mm internal guide and a typical 150 x 50 mm external aluminium box section is likely to be well under 0.1 dB. You would not expect a noticeable difference in hydrogen-line detection.
In fact, many builders deliberately choose widths around 140-150 mm for 1420 MHz because they place the operating frequency at roughly 1.3-1.4 times cutoff, which is a good region for stable TE10 operation and simple probe matching.
So, from a practical hydrogen-line perspective:
- 150 x 50 mm internal: excellent.
- 146 x 46 mm internal: essentially the same.
- 144 x 44 mm internal: essentially the same.
- 140 x 40 mm internal: still excellent.
The exact probe location and probe length will have a much greater effect on performance than the few millimetres difference in waveguide dimensions.