Which way did it go?

Foundations of Amateur Radio

Propagation, the art of getting a radio signal from one side of the globe to the other, is a funny thing. As you might know, I've been experimenting with WSPR or Weak Signal Propagation Reporter and for about a year running a beacon on 10m. Out of the box my beacon uses 200 mW to make itself heard. I couldn't leave well enough alone and I reduced the output power. Currently a 10 dB attenuator is connected to the beacon, reducing output to a notional 20 mW. I say notional, since I haven't actually measured it, yet.

With so little power going out to my vertical antenna, a homebrew 40m helical whip, built by Walter VK6BCP (SK), and tuned to 10m with an SG-237, it's interesting to discover what's possible.

Last night my signal was heard in Denmark. Picked up by Jorgen OZ7IT, 13,612 km away. That report broke another personal best for me, achieving 680,600 kilometres per Watt. I was stoked!

I shared a screen-shot of my report with friends. One friend, Allen VK6XL, asked a very interesting question. "What makes you think it was short path?"

Before I go into exploring that question, I need to explain. If I was to fly from Perth to Sydney, the popular way to travel is across the Australian Bight, over Truro, north of Adelaide, clip the northern tip of Victoria, over the Blue Mountains to Sydney. The distance is about 3,284 km. This route is known as the great circle route, more specifically, the short great circle route.

It's not the only way to travel.

Instead of heading East out of Perth, if I head West, I'd fly out over the Indian Ocean, Africa, the Atlantic Ocean, the Americas, the Pacific Ocean and finally arrive at Sydney. That journey would also follow a great circle route, the long great circle route. It's about 37,000 km long. You might notice that I wasn't very specific with either the path or distance. There's a reason for that. None of the tools I've found actually provide that information, other than to point out that the entire circumference of the planet is about 40,000 km and that it's not uniform since Earth isn't a perfect sphere.

You might be asking yourself at this point why I'm spending so much energy worrying about taking the long way around and how that relates to my 20 mW WSPR beacon.

In amateur radio we refer to these two travel directions as the short-path and the long-path.

Radio signals travel along the curvature of Earth bouncing between the Ionosphere and the surface. How that works exactly is a whole different topic, but for the moment it's fine to imagine a radio signal skipping like a stone on water. As a stone skips a couple of things happen. If the angle at which it hits the water is just right, it will continue on its journey, get the angle wrong and you hear "plop". Every skip is slightly lower than the previous because the stone is losing a little bit of energy. Every time the stone touches the water it creates a splash that ripples out in a circle from the place where the rock hit. These ripples also get weaker as they increase in diameter. Consider what happens if you skip a rock across concrete or sand instead of water and if you really want to geek out, there's also wind resistance on the rock.

A complex equivalent dance affects a radio signal when it propagates between two stations. For success, enough radio energy needs to reach the receiver for it to be decoded. For our signal to make it to the other side of the globe it must bounce between the Ionosphere and Earth's surface. Every bounce gets it closer to the destination. Each time it loses a little bit of energy. This loss happens at the Ionosphere, at the surface and in between through the atmosphere.

To give you a sense of scale, my signal report from Jorgen in Denmark was -28 dB. It started here in Perth as 13 dB, so we lost 41 dB along the way. We're talking microwatts here. I'll note that I'm avoiding how this is exactly calculated, mainly because I'm still attempting to understand how a WSPR signal report actually works since it's based on a 2,5 kHz audio signal.

As I said, enough energy needs to make it to the receiver for any of this to work.

There's an assumption that less distance means less energy loss. It's logical. A shorter distance requires less hops and as each hop represents a specific loss, less hops means less loss.

But is that really true?

There's nothing stopping my beacon signal from taking a different route. Instead of travelling the short-path, it can just as easily head out in the opposite direction. Theoretically at least, my vertical antenna radiates equally in all directions. The long-path is mostly across water between Perth and Denmark. What if hops across the ocean were different than hops across a landmass? Turns out that they are in several ways. For example, there's less energy loss in a refraction across the ocean, how much less exactly is still being hotly debated. Much of the data is empirical at the moment.

It gets better.

What if I told you that the report was near to sunset? At that time there's a so-called grey line phenomenon related to how the sun stops exciting the Ionosphere and how different layers of the Ionosphere start merging. As a result the angles of refraction across the Ionosphere change and longer hops are possible.

What if the long-path took less energy to get to Denmark than the short-path did?

Would Jorgen's decoder care?

If that's the case, my signal didn't travel 13,612 km, it travelled twice that and I'd have well and truly cracked a million kilometres per Watt.

So, is there a way we could know for sure?

Well, yes and no.

For starters we'd need beacons that transmit at a very precise time. Then we'd need synchronised receivers to decode the signal. A signal travels 3,000 km in a millisecond, so we're going to need something more precise than the timing set by NTP or the Network Time Protocol used by your home computer. If we used GPS locked transmitters and receivers we'd be working in the order of 50 nanoseconds and be in the range of 15m accuracy.

That would allow us to calculate the physical distance a signal travelled, but that's not the whole story.

What happens if your signal travels all the way around the globe, or if some of it reflects back, so called back scatter, like the ripples from a stone coming back towards you, and that signal travelling back past you to the receiver? There's endless variation, since the planet isn't round with a flat surface nor is the Ionosphere.

So, do we know if my report was a long-path or a short-path? Not really. Based on the time of day, there's a good chance that it was a long-path report, but only if we actually measure the delay between send and receive will we have data to make a better assurance than "possibly" or "probably".

As I started, propagation is an art.

I'm Onno VK6FLAB