top of page

A Quick Guide to FT8, JS8 and FT4 Digital Modes

Digital Mode Transceiver Optimization for FT8, PSK31, JS8

Transmitter Linearity — The Key to Clean Digital Signal​

Digital modes like FT8 and PSK31 rely on precise signal shaping. Unlike voice modes, where minor distortion might go unnoticed, digital transmissions demand clean, predictable gain from audio input to RF output. Any distortion can smear your signal, create unwanted sidebands, and reduce decoding accuracy — not just for you, but for everyone on the band.
 

Why Linearity Matters

  • Digital modes encode data using exact amplitude and phase relationships.

  • Distortion widens your signal, making it harder to decode and more likely to interfere with nearby QSOs.

  • Clean signals respect the band — especially important in crowded segments.

​​

​

IMD — A Real-Time Window into Signal Quality

Intermodulation Distortion (IMD) is a key indicator of how clean your transmitter really is. Apps like FlDigi don’t just estimate it — they display it live, giving you instant feedback on whether your signal is clean or starting to distort.

What IMD Is
•     IMD happens when multiple tones mix inside a non-linear transmitter stage.
•     They’re unwanted mixing artifacts that show up as extra lines beside your signal on the waterfall.

Why It Matters
•     IMD widens your signal and degrades decoding — even if your own decode looks fine.
•     In packed bands, poor IMD affects others more than you.
 

Caveats — IMD readings can be skewed by:

  • Receiver limitations

  • RF noise

  • Front-end overload

     

Audio Level Calibration — Finding the Sweet Spot

Clean transmission starts with proper audio input level. Whether you're using a USB sound card or isolated interface, the signal fed into your transceiver’s DATA or accessory port must be carefully adjusted.
 

What to Avoid

  • Overdrive: Causes clipping, splatter, and high IMD.

  • Underdrive: Leads to weak signals and poor decode reliability.

  • ALC reliance: Many rigs apply ALC after distortion has occurred — so it’s not a safety net.
     

What to Aim For

  • A clean, narrow trace on the waterfall

  • No visible sidebands or excess bandwidth

  • Stable decode performance across the band

  • Use mid-range settings on transmit gain or audio level controls — avoid maxing out or running too low
     

Understanding the Curve: Where Linearity Lives

Here’s a simplified amplifier response curve, with audio input on the X-axis — representing the full transmit chain from audio drive, through the modulator, to RF drive and final amplification:
 

  • Green zone (linear): Output rises predictably with input. This is where digital modes thrive — clean, narrow, and decodable.

  • Yellow zone (non-linear): Gain compresses, distortion creeps in, and spectral purity drops. Some rigs apply some soft ALC here — but distortion may already be underway before ALC visibly reacts.

  • Orange zone (saturated): Amplifier runs out of headroom. Loud, but dirty. 

Ham radio Transceiver Simplified audio input vs power output graph

The ALC: A Helpful Guardrail or Hidden Culprit?

ALC (Automatic Level Control) reduces gain when input exceeds a threshold — but in many rigs, it kicks in after distortion begins.
 

Some transceivers apply ALC gradually, with soft onset near the top of the linear region. This can mask early IMD, making the signal appear clean while distortion quietly builds.
 

In digital modes, where signal integrity is critical, relying on ALC can hide poor calibration and lead to unpredictable output. The chart below shows how ALC engages — and why staying below the threshold is essential for clean transmission.

Ham Radio Transceiver ALC onset graph

Practical Recommendation: Stay in the Linear Zone

To avoid distortion and reduce stress on your transceiver’s output stages — especially with high duty cycle modes like PSK31 or JS8Call — set RF output to no more than 50% of rated max (e.g., 50W on a 100W rig).
 

When adjusting your digital mode audio input (from a PC or interface), aim for no more than 80% of the configured power during transmission — ~40W in this example. This keeps you well within the linear zone and, crucially, clear of harder ALC onset, minimizing splatter and maximizing decode reliability.

​

Set Digital Input Level or DIG Gain to around 50% in the rig’s menu. This places the input stage in its optimal range, reducing overdrive risk while preserving clean modulation.
 

Use your PC or interface’s output level control to fine-tune transmit power. Most interfaces deliver standard line-level audio — avoid excessive gain. Let the transceiver handle the amplification within its linear envelope

​

Practical Example — Yaesu FT-817 Output Validation
 

To illustrate practical calibration, I tested the FT-817 on 40m using a 1 kHz sine wave via the rear DATA port. The rig was powered by a 13.8V supply and set to high power mode (5W).

​

Although nominally rated for 5W, this unit stabilized at ~3.8W after warm-up — typical for the FT-817. Notably, cold starts can push output above 7W, risking poor linearity and stressing the finals.
 

Key Observations (noted in the chart below)

  • <1W: Under-sensitive and non-linear

  • 1–2.5W: Predictable gain region — the sweet spot for digital modes. On this unit, ~18mV RMS input yielded ~2W output.

  • ~1.5W: ALC begins to engage softly

  • >2.5W: ALC action increases, linearity degrades

  • >3.5W: Full ALC compression and heavy saturation

​​​

FT-817 Audio input vs Output Power

FT-817 Transmit Performance: Real-World PSK31 IMD Behavior

To assess the FT-817’s linearity under practical conditions, I conducted live PSK31 transmissions at stepped power levels — 1W, 2W, and 3W — using FLdigi to generate actual signals. The rig was set to its nominal 5W high-power mode, with receive monitoring via an Icom IC-718.

​

Rather than relying on synthetic two-tone tests, this approach reflects real-world digital mode behavior — where waveform purity, spectral footprint, and decoding reliability truly matter.

​

At each power level, the IC-718’s waterfall display was used to visually inspect signal width and intermodulation distortion (IMD) artefacts.

​

The following table summarizes typical received IMD levels and their implications for modest station setups, factoring in transmitter behavior, audio drive, and receiver fidelity

The FT-817 exhibits atypical ALC characteristics. Even when configured for 5W output, the ALC indicator begins activating around 1.5W. Its initial effect is relatively soft, with minimal impact on IMD at lower drive levels — suggesting early ALC engagement doesn’t immediately degrade waveform integrity. However, caution is warranted as power increases.
 

In practice, the FT-817 delivers acceptable PSK31 performance up to ~2W RF output. Beyond this point, signal widening and elevated IMD indicate a departure from optimal linearity — especially in crowded band conditions where spectral cleanliness matters.
 

Below are receive-side waterfall captures from the IC-718, each annotated with measured IMD. For reference, –28 dB IMD reflects cleaner transmission than –20 dB, and is preferred for reliable decoding and minimal adjacent-channel interference.

 

1W -- Rx IMD of -28dB. Slight sign of intermodulation products on the waterfall but good IMD​

FT-817 PSK31 signal trace at 1W output

2W -- Rx IMD of -23dB. Very little (if any) change in visible intermodulation products but IMD getting worse

FT-817 PSK31 signal trace at 2W output

3W -- Rx IMD of -17dB. Poor intermodulation products on the waterfall and IMD value unacceptable

FT-817 PSK31 signal trace at 3W output

Linearity Sensitivity Across FT8, FT4, and JS8Call​

Having explored PSK31’s sensitivity to non-linearity and IMD, it’s important to consider how other digital modes respond. While all benefit from clean transmission, their tolerance to distortion varies by modulation type, bandwidth, and decoding strategy.

​

FT8 and FT4: Moderately Sensitive

FT8 and FT4 use 8-FSK modulation, transmitting one of eight discrete tones per symbol. These tones are spaced 6.25 Hz (FT8) or 12.5 Hz (FT4) apart, with tight timing and narrow bandwidth.
 

Excessive IMD or splatter can blur tone separation, causing decoding errors — especially in crowded band segments. However, both modes are robust against modest distortion thanks to forward error correction and time-frequency correlation.
 

Signals with IMD around –20 dB may still decode cleanly in low-QRM conditions, provided SNR is favourable and adjacent signals are well-separated.
 

Most FT8/FT4 software (e.g., WSJT-X) lacks built-in IMD metrics, reflecting their relative tolerance. Operators concerned about linearity can still assess signal cleanliness using waterfall displays or spectrum analysis tools.


Verdict: Linearity matters, but FT8/FT4 can tolerate moderate IMD if signals are isolated and SNR is strong.

​

JS8Call: More Sensitive

JS8Call builds on FT8’s FSK modulation but adds longer messages, variable timing, and continuous decoding. It supports free-text messaging, acknowledgments, and relays — making decoding accuracy more dependent on sustained signal integrity.

​

IMD-induced tone smearing or drift can disrupt longer exchanges, especially in weak-signal or multi-hop conditions. While JS8 often operates in quieter sub-bands, poor linearity still degrades performance for both sender and receiver.

​

Like FT8, JS8Call lacks built-in IMD metrics. Operators aiming for clean transmission should monitor their signal using external receivers or waterfall tools — especially when running higher power or troubleshooting ALC behavior.

​

Verdict: JS8Call is more sensitive to linearity than FT8/FT4 due to its extended message structure and reliance on clean tone transitions.

​

​

Summary: Clean Transmission Is a Shared Responsibility

Digital modes like PSK31, FT8, and JS8Call all benefit from clean transmitter behavior, but their tolerance to non-linearity varies:

  • PSK31 demands high linearity for reliable decoding

  • FT8/FT4 tolerate modest IMD due to robust error correction

  • JS8Call sits closer to PSK31 in its need for spectral purity

 

Even when software lacks built-in IMD metrics, external monitoring — via waterfall displays, spectrum snapshots, or receiver-side IMD readings — remains essential for validation.

​

bottom of page