Synchronized four-object frequency measurement protocol

Two devices do not automatically cover four objects

Four objects require four concurrent readout paths for each observable that must be captured simultaneously. Two instruments can be sufficient only if each supports two genuinely concurrent, suitable measurement channels. Four single-point optical readout paths feeding a compatible simultaneous-sampling DAQ are another candidate architecture. A scanning instrument visiting four points sequentially is not equivalent.

The number of laser beams in a 3D instrument must not be confused with the number of independently observed objects. The chosen hardware’s topology, bandwidth and timing specifications determine what is actually measured in parallel.

Proposed roles, not a replacement for the current LOC-1 draft

Use documented sphere dimensions, wall thickness, material and mounting. Matching reduces avoidable variability; it does not make objects exactly identical. Rotate roles and channel assignments in scheduled calibration blocks so object, channel and position biases can be separated.

Synchronization has three layers

1. Distribute a common compatible frequency reference, with lock status recorded.
2. Share or deterministically align sample clocks so records do not drift apart.
3. Use a hardware trigger and aligned acquisition windows; verify trigger and analog-path delay with a common injected signal.

NI explicitly notes propagation skew when a sample clock is shared between devices. Cable delay, analog filtering and sensor latency belong in the timing budget. A 10 MHz reference can stabilize compatible instruments, but is not by itself a common start trigger or identical analog response.

Finite windows, not an instantaneous frequency

Frequency is estimated from an observation interval. Align window starts, lengths and processing, not just timestamps. For sampled real waveforms the sampling rate must exceed twice the highest retained signal frequency, with an appropriate anti-alias filter and practical margin. Raw 500 kHz content therefore requires more than 1 MS/s per channel; demodulated outputs have different requirements. An FFT’s bin spacing is 1/T, but bin spacing is not a complete frequency-uncertainty estimate.

Trial sequence proposed for review

1. Define the tracked mechanical mode, excitation method and frequency estimator.
2. Co-locate objects for baseline and common-signal timing checks.
3. Calibrate thermal sensitivity, repeatability and channel delays.
4. Randomize blinded A/B and sham placements; allow a measured settling period.
5. Capture all channels over identical windows and log environmental conditions.
6. Freeze exclusions, smallest effect of interest and confirmatory analysis before unblinding.
7. Publish raw data, calibration and null results; replicate candidate signals independently.

Public design challenge

Would four objects improve fault diagnosis enough to justify added cost and calibration? Describe a multi-sensor project you have built: how did you verify timing, what failed, and what would you change?

Continue the design review

• Synchronized measurement research hub
• Does Earth’s motion affect resonance measurements?
• Frequency measurement error budget: timing, drift and false positives
• What a miniature clock network teaches SPHERE about synchronized measurements
• Earth rotation and resonator tests: what SPHERE can learn
• Hardware for synchronized resonance measurements
• Community review: how would you improve the experiment?
• Current draft LOC-1 protocol