8400 S-meter readings strange behavior

The best way to test this is with a signal generator. And, you have landed in that unique problem. If you have 2 meters, which one is correct. And, S Meters readings are a great area for discussion on how do you actually measure the strength of a signal in the Receiver passband. It is hugely different between a superhet radio, hybrid radio and a direct sampling radio. Generally, direct sampling radios have it correct.

I can't speak for how Icom calibrates their S meters, but I can guarantee that your Flex was correctly calibrated when it left the factory.

FlexRadio calibrates its S‑meter (signal strength meter) according to a combination of industry standards and direct measurement within the digital signal chain of the radio.

I just hooked my newly calibrated IFR-1200S and set it to -73dBm.

1. Following the IARU Standard

• FlexRadio aligns with the IARU Region 1 S‑meter standard for HF receivers:
  • S9 = 50 µV at the antenna input (terminated into 50 Ω)
  • Each S‑unit ≈ 6 dB difference in signal strength
• Signals above S9 are indicated in dB over S9 (dB over S9+xx).

2. Digital Measurement in the SDR

Unlike traditional analog radios, FlexRadio’s S‑meter is derived directly from the digital signal chain:

1. The RF signal is converted to digital early in the receive path (ADC).
2. The radio measures RMS voltage or power in dBm across the selected bandwidth.
3. This value is then mapped to S‑units using the IARU scale.

• Example Mapping:
  • S1 ≈ ‑121 dBm
  • S9 ≈ ‑73 dBm
  • S9+20 dB ≈ ‑53 dBm

3. Factory Calibration

• During production, FlexRadio radios are factory calibrated:
  1. A known RF signal is injected (e.g., 50 µV for S9).
  2. The receiver’s measurement is compared and adjusted via software.
  3. Calibration ensures consistent readings across bands and preamp settings.

4. Dynamic and Accurate Measurement

• Because the S‑meter is computed from the FFT bins of the digital IF, it is:
  • Frequency-accurate (per slice receiver)
  • Independent of AGC (it measures the true signal level, not audio level)
  • Consistent across the entire passband

If you a stickler for accuracy, I would trust the FlexRadio measurement as being as close to a bench standard as you can get. Your Icom may be correct as well, but Icom may be using their own interpretation of what is correct.

BTW, the amazing thing about your FlexRadio is that that S Meter reading does not change regardless of the preamp setting. And, why should it?

73