Choosing FT8 Frequency & Interval on 6 Meters

As the summer sporadic E season winds down there are several items to reflect upon. I'll have more to say about the season in my usual end of season wrap up article, later in August. There is one topic I'd like to address now. Although it can apply to any band where FT8 is used, there are behaviours that are unique to 6 meters due to the unique character of DX propagation.

This is a typical view of a mostly domestic (within NA) opening, taken several weeks ago. There are several strong signals in the US (via sporadic E), strong local stations and a smattering of weak DX (Europe if I recall correctly). European signals are mostly on 00/30 intervals and NA signals are on 15/45, except those working stations within NA. 

It's a mix of DXers (like me) and those that are happy to work anyone when 6 is open. There are a few poorly modulated signals which you can probably identify without me pointing them out. There was at least one robot that I am certain of. I blacklist the most obnoxious robots to keep the message pane "clean".

Notice the relative absence of signals at low and high audio frequencies. That's typical when activity is moderate, as in this case. Stations tend to cluster at the centre of the 3 kHz pass band, and quite a few will nail themselves to exactly 1500 Hz. Two of those are visible, one strong and one weak.

But why? Modern rigs have amazingly flat receive and transmit response curves so there is no reason to favour the centre. Further, by using the recommended split feature (real or "fake it") in WSJT-X and its close relatives the transmit audio is kept close the centre of the transmitter's pass band regardless of what frequency is selected. Certainly there are old or minimalist rigs used in SSB mode, including microphone equalization or fixed bandwidth shaping, but from the observed behaviour you'd think half the stations use those. That's highly unlikely. Old habits die hard.

When activity is high in a good sporadic E opening you'll see the opposite behaviour: crowding the edges. The idea is that since most hams are observed to favour the centre you are less likely to collide with others, including those you don't copy, by going high (above 2500 Hz) or low (below 500 Hz). Some will operate over 3000 Hz or below 200 Hz which may not decode in many station configurations. I set my receive bandwidth to 3300 Hz but not everyone does or can do that. I have a hard lower cut off set at 200 Hz.

This brings us to the ancient idea known as the tragedy of the commons. Put simply, one person seeks an advantage in a situation by choosing to behave counter to the crowd. They fail to realize that they are not unique or smarter than others. Many others will have the same thought and they will all choose the same contrary behaviour. The result should be obvious. They may persist even when they can see that others are doing the same and therefore negating the benefit of their "contrariness". Humans are funny that way.

WSJT-X can decode overlapping FT8 signals but not with assured high probability. The closer they match in frequency and timing the more likely it is that the weaker signal will not be decoded. However, especially with sporadic E propagation, you may not know if that distant DX station sees overlapping signals on your frequency. Maybe you can deduce it by how well or poorly they decode your transmissions, or you can skip a transmit period to listen for interfering stations, ones that may be weak to you but strong at the DX side of the circuit.

I actually do move to the edges in many openings. However, as the opening widens and more stations appear on the waterfall I start to move around quite a lot to find a reasonably clear frequency. I continuously search for an advantage if propagation favours another area within NA. I may stick to my frequency when I know that I can outgun interlopers since I have very capable 6 meter station (antenna and power) regardless of others that may camp out on or near my transmit frequency. 

Flexibility is the key. Never marry a frequency and never persist in behaviour that doesn't achieve acceptable results. Yet a surprisingly large number of committed 6 meter operators do just that. There are stations that can always be found on the exact same frequency throughout an opening on every day they're active. I never ask but I do wonder what convinces them that this negative behaviour is to their benefit. If another station happens to select that frequency they'll transmit there regardless. It's both rude and foolish.

I am equally flexible in my contest operations on HF. Running low in the band tends to increase rate. But if it's too crowded or another station moves too close to my run frequency I'll move. Struggling wastes time and doesn't get results. FT8 is no different. That is especially true on 6 meter sporadic E when activity and band occupation can cycle from low to high and back again within a minute or two. Flexibility will make you more successful at reaching your objectives.

Overlap is not necessarily evil. At right is an example of two kinds of overlap on the same frequency: same interval and opposite interval. Both have their pros and cons.

Some stations automatically transmit on your transmit frequency when they call you. WSJT-X has this option. Or they may do it manually when they perceive a benefit -- I do the same. The reasoning is that stations prefer a frequency that is clear on their receive intervals. That is, when I transmit 15/45 it is preferred that the frequency be quiet during 00/30 intervals to copy especially weak DX. Callers might otherwise have difficulty escaping the QRM elsewhere in the 3 kHz window. 

If you manually QSY to a station's frequency in order to work them, be sure to QSY afterward. Many forget and remain there to work other stations. In most cases it is a mistake, not an attempt to "steal" the frequency. Once they realize what they've done they QSY. Don't assume that other hams are malicious!

If two or more stations QSY to your transmit frequency there may be a decoding problem. Letting WSJT-X do the QSY is more problematic than doing it manually since, as noted above, the decoding algorithm can't easy separate signals too close in both time (clock accuracy) and frequency. Manual frequency selection, by being less accurate, ensures at least several Hz separation. Sloppy is good!

On the other hand, if another station is on my frequency but the opposite interval, that may be acceptable during busy openings. The practice in effect doubles the available bandwidth -- what those in the telecom business might recognize as TDM (time division multiplexing). However, in this case when callers QSY to your transmit frequency there can be bedlam since now we have both transmit and receive on the same frequency and interval instead of alternating transmissions. There is no perfect solution on a busy band. Again, be flexible.

The other type of overlap is visible near the top of the screenshot: a strong station has started transmitting on the same frequency and interval as a weaker station. At least that's what I hear, which can be very different to what they hear; the two stations might not copy each other at all. This is more common with sporadic E than what for typical propagation on the HF bands. 

Some stations, in a bid to find a clear frequency, will click on an empty spot on the waterfall without first checking whether the frequency is clear on their interval. That requires skipping a transmit slot to listen. Many hams are in a hurry and either don't care or don't realize what they're risking. Even when I'm running stations I will occasionally skip a transmit period to check whether some interloper has appeared, whether by their QSY or a change of propagation. It may be entirely unintentional, but can still be a problem.

When things get seriously busy on 50.313 MHz during a good DX opening, stations should move to the inter-continental window at 50.323 MHz. This doesn't occur as much as it should. A station will check 323 when 313 is busy, see little activity and move back to where everyone seems to be. It's self-defeating behaviour. It can be inconvenient to check 323, or worse if you decide to stay there for a while. You may be missing some juicy DX on 313 amidst the QRM!

This season I have occasionally CQ'd on 323 in a bid to increase activity by demonstrating activity. It seems to work, though not as much as I'd like. Like everyone else, I have to be where the stations I want to work are operating. I don't know if 6 meter DX'ing behaviour will ever change. 

On HF there is less formality about whether to transmit on 00/30 or 15/45 than there is on 6 meters. On 6 meters the convention is that you transmit on 15/45 when working stations to the east and 00/30 to the west, especially for inter-continental DX. When working Europe, North Americans transmit on 15/45 and the Europeans transmit on 00/30. In the other direction I switch to 00/30 for working the NA west coast, the Pacific and east Asia. The convention is widely followed and works pretty well. 

There is less consistency when working north-south. For example, from NA to South America, Africa and central/west Asia. As long as 6 meters is open to one region at a time there is not problem. Just do what others do, or switch your interval choice occasionally so that you don't overlook stations you'd like to work.

A problem can arise when 6 meters opens to more than one continent at the same time. You can be somewhat selective when you point your yagi but that may not be enough. This has happened a few times this July when we had concurrent openings to Europe and Japan.

What happens is the following. The band is open between NA and Europe before opening to Japan. A Japanese station hears an NA station and wants to work them. So they click on the call, such as mine, and transmit on 00/30 rather than 15/45 by the inter-continental convention. They have to! The NA station works them and then another JA calls. Others soon get into the same pattern as the opening spreads.

Before long all the east Asian stations are transmitting on 00/30. It like a chain reaction or the old cliche about tipping over dominoes: as one falls onto the next, it falls onto the next in line and soon all the dominoes are down. There is no way to reset during the opening. Sometimes the KH6 and ZL stations also enter the mix and find themselves having to operate on the opposite interval as well. When the band opens to Japan the next evening, after the opening to Europe has ended (or never happened), everyone resumes following the convention.

Conventions are helpful but not universally understood or followed. Even when you follow the convention it is more important to go with the flow. Like driving down a highway, it is safer and easier to match your speed to that of the prevailing traffic, whether higher or lower than you'd prefer, rather than rigidly adhering to the posted speed limit. 

Do what works in the moment and don't fret about it too much. The DX is hard enough to come by on 6 meters that you don't want to make it any harder.