Q65 Mode in WSJT-X: Weak-Signal Digital for EME and Scatter

Most people meet WSJT-X through FT8, a mode designed for crowded HF bands. Q65 mode belongs to the other side of the program: it was built for signals that are not only weak but also unstable, the kind you get from moonbounce, rain scatter, ionospheric scatter and trans-equatorial paths. This guide explains what the WSJT-X User Guide says about Q65, how its sequence lengths and submodes fit together, why a priori decoding gives it an edge, and how to configure and operate it. All technical details below come from the WSJT-X 3.0 documentation, so check the current guide if you use a newer release.

What Q65 Is Designed For

The WSJT-X User Guide describes Q65 as a mode designed for fast-fading signals: EME, tropospheric scatter, rain scatter, ionospheric scatter, trans-equatorial propagation (TEP) and similar paths. A fast-fading signal rises and falls in strength within a single transmission, which defeats decoders that assume a steady carrier. Q65 uses a strong error-correcting code and a wide choice of tone spacings so that the decoder can still recover the message when the signal is smeared in frequency or time.

The guide also states that on VHF and higher bands, QSOs with the weak-signal modes are possible at signal levels 10 to 15 dB below those required for CW. Q65 sits in that family, which is why it is widely used for 6 m, 2 m and microwave moonbounce. If you are new to the idea of bouncing signals off the Moon, read our EME moonbounce beginner guide first, then return here for the settings.

Sequence Lengths and the Q65 Submode Names

Every Q65 submode name has two parts: a number for the length of the transmit/receive sequence in seconds and a letter for the tone spacing. According to the protocol specification in the guide, Q65 offers T/R sequence lengths of 15, 30, 60, 120 and 300 seconds, and submodes A to E have tone spacings of 1, 2, 4, 8 and 16 times the symbol rate. So Q65-60A is a 60-second sequence with the narrowest tone spacing, and Q65-120C is a 120-second sequence with tone spacing four times the symbol rate.

The occupied bandwidth is 65 times the tone spacing. The guide gives the extremes: 19 Hz for Q65-300A and a maximum of 1733 Hz for Q65-15C, Q65-30D and Q65-60E. From the 1733 Hz figure for Q65-60E you can work out the rest of the 60-second family, because each step up in letter doubles the spacing. The table shows that arithmetic.

The right-hand column is a guideline, not a rule from the documentation. The guide only says that Q65 offers a range of tone spacings for different propagation conditions. In practice, the activity group for your band and path decides which submode everybody uses, and the sensible approach is to follow that convention rather than invent your own.

How the Q65 Message Is Built

Q65 uses the same 77-bit message structure as FT8, FT4 and FST4, so you can use the same standard exchange of callsigns, grid locator, report and 73. The error correction is a specially designed (65,15) block code with six-bit symbols. Two symbols are punctured, meaning they are computed but not transmitted, which leaves an effective (63,13) code: 13 information symbols carried by 63 channel symbols. The punctured symbols hold a 12-bit CRC computed from the 13 information symbols, and the CRC keeps the false-decode rate very low.

A 22-symbol pseudo-random sequence spread through the transmission is sent as tone 0 and used for synchronization. Adding the two parts gives 63 + 22 = 85 channel symbols in every Q65 transmission. You do not need these numbers to operate the mode, but they explain why Q65 tolerates drift and fading: the synchronization is spread through the whole sequence instead of sitting at the start.

Why A Priori Decoding Matters in Q65

The Q65 decoder uses a priori (AP) information, which means it makes use of what it already knows about the message. At the start of a QSO it knows your own callsign and the word CQ. As the QSO progresses, the available AP information grows to include the callsign of the station you are working and perhaps its four-digit grid locator. The decoder uses whatever is available at that moment.

The guide notes that AP decoding raises decoder sensitivity by up to 4 dB at the cost of a slightly higher rate of false decodes, and that for Q65 it is always enabled. Decoded lines show the type of AP information with a code of the form qP, from q0 (no assumed information) through q1 (CQ), q2 (MyCall), q3 (MyCall and DxCall) and q4 (MyCall, DxCall and a final RRR, RR73 or 73) to q5. If the decode was averaged over several periods, a further digit shows the number of messages, so q34 means a Q65 decode using AP type 3 from four messages, and an asterisk means ten or more.

Seeing a q-code on a line is useful because it tells you how much trust to place in it. A decode that appears only because the program was already expecting your callsign is not the same as a decode of a CQ from a stranger. WSJT-X also marks lower-confidence decodes with a question mark and does not forward them to PSK Reporter, which keeps occasional false decodes out of the spots. Our PSK Reporter guide shows how those reports are used.

Where Q65 Is Used

EME on 6 m and up

The guide illustrates Q65 with a Q65-60A sequence received by EME on 6 meters, with several stations decoded over the Earth-Moon-Earth path. It also describes the single-tone short messages that some operators use for Q65 EME contacts on the microwave bands. Checking the box labeled Sh generates these messages, and selecting Tx6 produces a single tone at 1000 Hz to help the other station find you; switching Tx6 to 1250 Hz tells them you are ready to receive messages. These short-form messages are not decoded automatically and auto-sequencing will not respond to them, so you interpret them yourself.

If you want background on the path itself, the Wikipedia article on Earth-Moon-Earth communication explains why echoes are so weak and why the Moon’s motion causes Doppler shift and fading. That fading is exactly the problem the wider Q65 tone spacings were made for, and it is why moonbounce stations on higher bands choose wider submodes than stations on 6 m.

Ionospheric scatter on 6 m

Another example in the guide is a 6 m ionospheric-scatter signal over a 1155 km path, received using submode Q65-30A and barely audible most of the time. That example shows the practical appeal of the mode: a contact that would be nearly impossible by ear or with CW becomes routine when the computer can integrate the signal. If you are curious about the band, our 6 meter band guide covers the openings in more detail.

Tropo, rain scatter and TEP

Tropospheric ducting, rain scatter and trans-equatorial propagation all produce signals that fade and flutter. Our guides to tropospheric ducting and transequatorial propagation describe these paths. The Q65 mode is simply the digital tool that suits them, since the longer sequences let the decoder collect more signal energy than FT8 can in 15 seconds.

Setting Up Q65 in WSJT-X

Setup follows the same pattern as other WSJT-X modes. Choose Q65 on the Mode menu, then select the sequence length and submode your activity group uses. WSJT-X 3.0 adds buttons for quick access to six popular Q65 submodes and to Echo mode, and the guide notes that the experimental 15A, 15B and 15C submodes can be decoded with EME delay. For EME work, the 3.0 release also makes it easier to pick dial frequencies within an EME sub-band and to offset your transmit frequency by a whole number of megahertz from the receive frequency to match different regulatory allocations.

Before your first Q65 QSO, check these points:

• Clock accuracy: the guide requires a means of synchronizing the computer clock to UTC within one second, and the longer the sequence, the more important a steady clock is for averaging. Our FT8 time sync guide applies equally here.
• Radio control: use CAT control through Hamlib so that WSJT-X can set the dial frequency and handle split operation.
• Audio level: keep the transmit audio low enough that the radio does not show ALC action, exactly as you would for any digital mode.
• Repeat transmissions: WSJT-X offers a general option, MSK144/Q65: Tx until 73 is received, which keeps sending until it receives a 73 when copy is marginal. The guide says a maximum of ten attempts is made before it times out.

MAP65 and QMAP for Serious EME Stations

For serious moonbounce, WSJT-X installs a companion program called QMAP. The guide describes it as a program that uses a single polarization channel and only the Q65 mode, covering a 90 kHz sub-band at a time and working together with WSJT-X to provide automatic Doppler compensation and other EME features, especially on 1296 MHz and higher bands. Its older relative MAP65 handles JT65 as well and can use two orthogonal polarization channels for automatic polarization-matched reception. Both are installed with WSJT-X, although MAP65 only on Windows. Most beginners will never need them, but it is useful to know that the Q65 ecosystem extends beyond the main window.

Q65 Compared With FT8 and FST4

FT8 uses 15-second sequences and is built for busy HF bands where speed matters more than the last decibel. FST4 is designed primarily for LF and MF, with sequence lengths up to 1800 seconds. Q65 occupies the middle ground for VHF and above: sequence lengths from 15 to 300 seconds and tone spacings chosen to match how much the signal spreads. If you want a broader overview of other modes, our article on digital modes beyond FT8 puts them side by side.

Reading the Decoded Lines

A Q65 decode line looks like any other WSJT-X line, with the UTC time, signal-to-noise ratio in dB, time offset DT in seconds and audio frequency in Hz, followed by the message. The mode character for Q65 is a colon, and the end-of-line information carries the q-code described above. Signal reports are expressed as signal-to-noise ratio in a standard 2500 Hz reference bandwidth, so a report of -22 means the signal power was 22 dB below the noise power in that bandwidth, a level far below what you can hear. When you work a station on a long sequence, expect a QSO to take many minutes, and use the waiting time to check your frequency, your clock and your antenna pointing rather than to retune the radio.

Common Mistakes

• Choosing a submode on your own instead of following the convention for the band and path, so that nobody hears you.
• Forgetting that sequences are timed to UTC: stations transmit in alternating periods, and starting in the wrong one means you transmit when your partner does.
• Reading too much into a q-coded decode. Treat a decode that depends heavily on AP information as likely, not certain, until the next period confirms it.
• Running too much power or overdriving the audio. Weak-signal modes reward clean signals, not loud ones.
• Expecting results from a poor receive system. Q65 can pull a signal out of the noise, but a low-noise receive setup still decides how far you can go.

The Bottom Line

Q65 is the WSJT-X mode for signals that fade, flutter and drift: EME, scatter, ionospheric paths and TEP. Learn the submode naming (period in seconds plus a tone-spacing letter), follow the convention used by activity on your band, keep your clock and audio levels in order, and let the AP decoder help you without trusting it blindly. Start with a sequence length and spacing that matches published activity, listen for a few periods to see what the decoded lines look like, and then make your first call.

Frequently Asked Questions

What does the name Q65-60A mean?

The number is the T/R sequence length in seconds and the letter is the tone spacing. Submodes A to E use tone spacings of 1, 2, 4, 8 and 16 times the symbol rate, so Q65-60A is a 60-second sequence with the narrowest spacing.

Which sequence lengths does Q65 offer?

The WSJT-X documentation lists 15, 30, 60, 120 and 300 seconds.

Is Q65 only for EME?

No. The guide lists EME together with tropospheric scatter, rain scatter, ionospheric scatter and trans-equatorial propagation, as well as other fast-fading signals on VHF and higher bands.

Does Q65 use AP decoding?

Yes. A priori decoding is always enabled for Q65 and can raise sensitivity by up to 4 dB, with a slightly higher chance of false decodes. Decoded lines carry a q-code that shows what information the decoder assumed.

Can I use the same message format as FT8?

Yes. Q65 uses the same 77-bit message structure as FT8, FT4 and FST4, so the standard exchange of callsigns, grid, report and 73 works.

What is QMAP?

QMAP is a companion program installed with WSJT-X that decodes Q65 over a 90 kHz sub-band using a single polarization channel and works with WSJT-X for Doppler compensation. It is aimed at EME, especially on 1296 MHz and above.