Deviation Creep and Adjacent-Channel Splatter
A technician services a 25 W VHF FM marine radiotelephone operating on 156.800 MHz in 25 kHz channels. To "make it sound louder" he turns up the microphone gain trimmer, and peak deviation rises from 5 kHz to about 9 kHz on voice peaks. The audio passband is 300–3000 Hz and the deviation limiter is supposed to hold peaks at 5 kHz. Nearby stations now report distorted, pinched audio and interference to the adjacent channel, and the ship's own receiver seems noisier on weak signals. Output power is unchanged at 25 W and the antenna system measures 1.3:1 at the operating frequency.
1. Use Carson's rule to compute the necessary bandwidth before and after the adjustment, and write the corresponding emission designators.
Read the worked answer
Carson's rule: BN = 2(Δf + fm), with fm the highest modulating frequency, 3 kHz.
- Correct setting: 2(5 + 3) = 16 kHz → emission designator 16K0F3E (16.0 kHz necessary bandwidth, F = frequency modulation, 3 = single channel containing analogue telephony, E = telephony).
- After the adjustment: 2(9 + 3) = 24 kHz → 24K0F3E.
Judgement: a 25 kHz channel does not give 25 kHz of usable bandwidth. With 16 kHz occupied, roughly 4.5 kHz of guard remains on each side; at 24 kHz the occupied spectrum runs almost to both channel edges, and the sidebands beyond the 99 % points put real energy into the adjacent channels. The design point is 16K0F3E for a reason - the 5 kHz deviation was chosen so that full deviation plus full audio fits the channel, not to limit loudness.
2. The technician proposes dropping the carrier to 5 W to fix the interference. Evaluate this against the actual mechanism.
Read the worked answer
It does not fix it, and it costs range. In FM, occupied bandwidth is set by deviation and modulating frequency, not by carrier power. Reducing power from 25 W to 5 W lowers the amplitude of every sideband equally and shrinks the interference radius somewhat, but the emission still spills into adjacent channels at whatever range remains - and the ship's own reception drops by 7 dB.
It also confuses two different failure modes:
- Splatter / adjacent-channel interference - caused by excessive deviation, a clipped or overdriven audio stage, or a limiter that has been bypassed or moved out of the loop. Wrong here.
- Front-end overload / receiver desense in nearby sets - caused by high field strength. That is the case where reducing power actually helps.
The diagnosis here is confirmed by: interference appears only when this station transmits on voice peaks, the audio sounds pinched (limiter clamping hard), and it began with the mic-gain change.
Correct action: restore the microphone gain to bring peak deviation back to ≤5 kHz, then verify that the deviation limiter actually clamps - a limiter that was never re-adjusted after the gain change is the root cause.
3. List the measurements you would make to prove the set is compliant, and what each one tells you.
Read the worked answer
- Peak deviation with a deviation meter, driven by a 1 kHz tone at the normal microphone level and then at +20 dB: must not exceed 5 kHz, and must stay at 5 kHz as drive increases (proves the limiter works). If deviation keeps climbing, the limiter or its sensing point is faulty.
- Deviation vs audio frequency, 300–3000 Hz: confirms the pre-emphasis and limiter are flat in the intended passband and that nothing is boosting the low end.
- Occupied bandwidth / spectrum mask (99 % power bandwidth, or the mask from the equipment spec): should show ~16 kHz, not 24 kHz, and the adjacent-channel power should fall away steeply.
- Modulation sensitivity at the mic connector, to set the gain correctly rather than by ear.
- Carrier power and frequency error, to confirm 25 W and that the carrier is on 156.800 MHz - changing deviation should not have moved either, but you verify.
- On-air check with a distant station: audio should sound clean and the adjacent-channel complaint should stop.