Long distance 23cm aircraft scatter

The result: 590km QSO on 23 cms.

The recipe: IC-9700, 150W, 28el yagi with 1.5m boom, Airscout.

Once again, when I started on 1296MHz I would never have believed this was possible. A 590km microwave contact on a flat band with an antenna just 1.5m long? No chance. And before I started on 1296MHz I doubted if I would work anyone, even in my own square. On 21 August I worked a new one four squares down and three across.

I am pretty sure that this is my longest aircraft scatter contact on 23cm. Sure, I can get further during tropo lifts, but conditions were normal that day. To make things harder, my normal 36 element 3-metre boom antenna has been dismantled while I deal with corrosion damage. So I had gone back to the 28 element WIMO given to me by Neil G4DBN - my first ever 23cm antenna.

The WIMO 28 element antenna between 70cm and 4m yagis at GM4FVM

Click to enlarge images if necessary. 

Frankly, I was not confident about reverting to the WIMO antenna. The antenna has 20 elements along the boom and eight a right-angles to the boom as a reflector. It works OK, but being half the boom length of the 36 element brings a 2dB reduction in gain, despite the reflector. However, in the 23cm UKAC on 19 August I worked six stations and that restored my faith in it to some extent.

At 11:17 on 19 August I saw EI3KD calling on 70cm FT8. My beams were still pointing in that direction as I had earlier worked EI4ACB. EI3KD and I then worked on 70cm. Mark sent a message on Tx5 that he was going to 23cm. I followed up to 23cm and I could see strong traces of Mark's signal via aircraft scatter but the Doppler shift was too strong to decode the signal on FT8. I contacted Mark via KST Chat and said that I would like to try Q65 but at that stage I did not have the time to try. I did say though that I would like to try when we had time.  

And so it was on 21 August that I saw Mark again on the Microwave section of KST Chat and I asked him if we could try Q65. I think I blundered slightly by suggesting Q65-30C as the whole thing  might have gone better on a 15 second period. Anyway, off we went and within 30 seconds of calling EI3KD I decoded Mark calling me.

I was pretty sure that we could complete because of the path as shown on Airscout. There are plenty of planes passing over the Irish Sea along the path of our contact. The problem is that not many are at the right point along the path, shown by the short pink section in Airscout.

Screenshot of Airscout along the GM4FVM/EI3KD path on 24 August 2025

Only the mid section of the path west of the Isle on Man can be useful for reflections. On the screenshot above (taken today) I can easily 19 planes have just crossed or will cross the path, but only one is heading for the pink section and likely to provide us with the reflections we need. So this might take some time and need a few planes.

As Aircraft Scatter paths get longer, the pink part of the path gets shorter, plus other obstructions such as hills get in the way.

Screenshot of Airscout along the GM4FVM/EI3KD path on 24 August 2025

Airscout also shows this, and the vertical diagram above shows the small pink area which planes will have to be in to help. The vertical scale is the altitude of the plane. Anything landing in Ronaldsway airport in the Isle of Man is not going to be high enough! The bottom diagram above shows obstructions on the ground along the path (on a different vertical scale!), with the Irish Sea a long gap in the middle, but the Wicklow Hills and the Pennines at each end.

Another problem is going to be Doppler shift created by the movement of the aircraft across the signal path. Doppler effects rise as the frequency goes up. Quite a few 2m and 70cm contacts on FT8 are put down to Tropo propagation when in fact they are assisted by aircraft scatter. However, the Doppler is often too strong to allow FT8 to work at 23cm. Q65 is ideal in these circumstances as it compensates well for Doppler. I did increase "Max Drift"to 30. Exactly what the "Max Drift" setting does I have no idea as the WSJT-X Guide has not yet been updated to cover it. Suffice to say that in a previous aircraft scatter test it seemed to help so I increased it again this time.

Net result of all this - only a few planes at the right height would complete the contact and Q65 will need to cope with the Doppler.

Long pause after the first decode

Airscout produced hope with small planes and possibles which altered course. Lots of hope but no results. But then, after 19 minutes two planes in opposite directions were due to pass over the right area almost together. Result:-

Aircraft scatter QSO complete - even to the 73s.

The best decode at my end was -13dB which did look rather faint and Doppler shifted. The 73 afterwards was very difficult to see, but Q65 decoded it OK:-

Final decodes in Q65

So although nearly 600km is almost right at the end of the radius of aircraft scatter QSOs, it did prove practical even with a less that brilliant set-up. I now have 41 squares worked on 23cm. Who would have thought it?

Even if you find all this daunting I would still urge anyone to try 23cm aircraft scatter. There are lots of IC-9700s out that which, like mine initially, have never strayed on to 23cm DX. All you need is a decent antenna and coax. Or, in my case, a simple antenna and some coax I used to use on 432MHz. It is perfectly possible to make good aircraft scatter SSB QSOs during the contests using simple equipment. At other times many QSOs are arranged using KST microwave chat. Airscout helps, but you can just rely on the DX stations to tell you when a plane is approaching the path. Let them use Airscout!

Nothing here implies any greatness from GM4FVM or his equipment, which was largely donated by others. GM4JJJ (sadly SK), G4DBN, and G8SFA each helped me no end. Most local amateurs will help if you ask. 

There is still more for me to do on 23cm. It is not the weird place I used to think it was.

Perhaps later I will explain a bit more about Airscout. 

73 Jim

GM4FVM