EME basics

Last Updated on May 15, 2026 by John Berry

Radio amateurs have been bouncing signals off the moon, and hence having QSOs across the world via the moon, for the past seventy years. It’s termed moonbounce or EME (Earth-Moon-Earth communications).

The attraction of moonbounce is obvious. Just to be able to say, “I bounce signals off the Moon”, has kudos. It’s got to at least get the response, “Really, how interesting”, if accompanied with a bit of a blank look. It’s technically complex and very niche.

Bouncing signals off the moon

Then once you’ve done it, you can move on to work more stations on more bands before going off to seek another obscure interest.

Here’s my analysis of the issues and about my journey.

In the first sixty years of this pastime, the stations at either end of the link needed to be substantial – low noise amplifiers, stacked and bayed long Yagis or big dishes, and kilowatts of transmit power. The reason is easy to see. The System Value minus total loss, for modest stations at each end, is zero or negative for CW/Morse at 144MHz. In the 1950s and 60s, CW was the weak signal mode of the day. Received signals were on, or in, the noise. Simply, it would only just work. And ‘only just’ meant limited enjoyment. In the early days it took big stations, big budgets, and long development times to implement a station.

Recent technology improvements

About ten years ago radio amateurs developed narrow band data transmission and clever coding methods. These improved the System Value from the (low) CW state of art. These new transmission systems have been developed now such that, as an example, the JT65 data method developed by Joe Taylor K1JT, allows detection of signals at around -28dB below the 2.5kHz noise level. This is enough to overcome the various losses that previously complicated success.

The 2.4kHz bandwidth noise level is the typical reference – indeed the only reference. In this bandwidth, an SSB voice signal would need around 10dB signal-to-noise ratio to be intelligible. JT65 allows successful decoding at around 38dB below the 10dB receiver threshold (28dB below the noise floor) – a huge improvement in System Value. It’s so good that it approaches the Shannon limit.

Q65, JT65B or other narrow band data modes now make QSOs via the moon easily possible between modestly constructed stations.

Which band?

Typically radio amateurs set up initially on one of three bands: 144MHz, 432MHz, or 1296MHz. There are a huge number of reasons why individuals choose a particular band. I’ve tried here to build a decision matrix to explain some of the reasoning. This comes from my own ideas modified by comments on an early version of the matrix on an EME Facebook forum. The main comments were that I’d underestimated the cost!

For a novice starting out, it all depends on your objectives. If it is to do EME as an occasional experience, modest equipment can work if you are happy to only work big stations. If you want to win contests, huge complexity and cost is involved.  I’ve chosen a point in the middle – to be able to reliably QSO with a similarly scoped station. My decision matrix assumes this objective on each band and then scores each characteristic for that band. The usual RAG scoring applies – red is bad, green is good etc. Simply, despite 1296MHz having the highest cost (at £8,000/$US10,500) and greatest equipment complexity, it is the most popular: QSO certainty and lowest environmental noise trump all other factors. QSO certainty is in itself partly an outcome, but also reflects the growing 23cm EME community.

I’ve assumed that the novice here already has an Icom IC9700 or similar transceiver. If not, add another £2k to the bill. 1296MHz prices will rise further when administrations adopt the RA/WRC Recommendation (and ECC Decision) specifying the use of at least 3m dishes. A 1296MHz implementation could soon touch £10k.

‘Regulatory risk’ is the likelihood that the use of the band is constrained some time in the future (so wasting investment). There is some indication (in ITU opinions) that the 1296MHz band will be further constrained. And recently, the FCC, the American spectrum administrator, has licensed satellites in 432MHz. We’ll have to wait to see there if green will turn red.

Other bands are attractive too, from 13cm, to 3cm and up. But technical complexity increases with frequency. I have no aspiration to develop GHz bands capability in the near future.

Dimensioning a basic EME station

I mentioned above my objective to assemble a station such that I can normally QSO with similarly dimensioned stations. That basic EME station is shown below for 432MHz and described elsewhere in a path budget.

My basic EME station comprises 4×14 element Yagis, a low noise amplifier at the array, and 500W RF power at the array. Anything lesser and the distant station must contribute more (power and gain). With this basic EME station at each end of the link, both operators will realise signal strengths of around -22dB using the Q65-60B digital mode.

Basic EME stations can be dimensioned for 144MHz and 1296MHz by calculating the path budget in each case.

Innacuracies

This page has been a quick overview. Sadly, there are a lot of inaccuracies online in ham web sites and a lot of biased opinion. The result is that it’s difficult to understand why bouncing signals off the moon works, and hence how to best configure a station. When I asked about the path budget on one Facebook forum, well-meaning hams there advised that I should ‘not over think it – just assemble some kit and try it’. But it’s expensive and time consuming and something that most people will want some certainty in before getting started.

This site discusses all the propagation mechanisms for EME. I aim to define the ‘why’ behind the stories. It’s not overthinking. It’s explaining.

To succeed in EME Moonbounce, you must overcome a 250 dB path loss. You need to account for distance, ionospheric polarisation rotation, and Sun and Milky Way noise. Precision in your engineering approach is the difference between a contact and silence.

Moon pic by Mike Petrucci on Unsplash.