Four boards, one job: turn a faint optical signal into a stable, nanometre-accurate measurement. This note follows the QPD amplifier, 20-bit ADC, piezo controller and low-noise power supply as they form the sensing-and-feedback loop of an optical-tweezers microrheometer — then maps out where each board is used elsewhere.
The application
In an optical-tweezers microrheometer, a micron-sized bead held in a laser focus reports on its surroundings by how it moves. Reading that motion fast and cleanly is an electronics problem: the light leaving the bead must be turned into position, digitised without losing resolution, and — when needed — actively stabilised.
Each board owns one link in that chain. Together they set the instrument's noise floor and bandwidth, which in turn decide the smallest displacement — and therefore the smallest force and the weakest viscoelastic signal — the system can resolve.
A position signal flows from QPD to ADC to piezo controller and back. Change the ADC resolution and the readout coarsens into stair-steps; raise the power-supply noise and the floor climbs; close the feedback loop and the controller pins the true position to the setpoint. This is the whole argument for high-resolution, low-noise boards, made visible.
Drop the ADC to 8-bit and the readout turns into coarse stair-steps. Crank up the supply noise and the floor rises even at 20-bit. Close the loop and the piezo controller flattens the true position toward the setpoint — its residual is limited by exactly those two boards.
Position-detection model after Gittes & Schmidt [1] and Neuman & Block [2]; converter resolution / ENOB after Kester [5]; closed-loop nanopositioning after Moheimani & Fleming [4].
Why the electronics decide the result
20 bits over the QPD's linear range means a quantisation step far below the bead's thermal motion — you digitise physics, not the converter's staircase.
Clean, regulated rails keep amplifier and detector noise at the theoretical floor. Ripple on the supply shows up directly as position jitter.
250 kHz detection plus fast piezo feedback lets the system both observe kHz fluctuations and actively hold or scan position in real time.
What are they used for
Beyond optical tweezers, these are general-purpose precision building blocks. Here is where each one earns its place.
A quadrant photodiode with a fast, low-noise transimpedance amplifier converts a focused spot's landing position into sum and difference voltages at up to 250 kHz.
A 20-bit, 100 ksps converter with an ultra-low-noise front end and programmable gain turns tiny analog signals into clean digital data without throwing away dynamic range.
Multi-channel piezo drive with real-time feedback and positioning algorithms delivers sub-nanometre actuation for closed-loop stabilisation and scanning.
Regulated, dual-output rails with very low output noise and over-current protection give sensitive analog and detector stages the quiet supply they need.
Same chain, other instruments
AFM/STM tip tracking and closed-loop raster scanning.
Position-sensing feedback to hold a beam on target.
Sub-nm sample placement for lithography and metrology.
Active alignment of couplers, cavities and interferometers.
Strain gauges, photodiodes and bridges needing 20-bit reach.
Fast wavefront read-out driving deformable-mirror actuators.
Star-ground the analog section and shield QPD leads — a 20-bit ADC will faithfully record ground loops you didn't intend to measure.
Run detector and driver rails from independent low-noise outputs so piezo switching transients don't reach the front end.
Calibrate QPD volts-to-nm and trap stiffness with power-spectrum / equipartition methods before trusting absolute numbers [2,3].
Match controller gain and bandwidth to the mechanical resonance — too much gain turns a stabiliser into an oscillator [4].
F. Gittes, C. F. Schmidt, “Interference model for back-focal-plane displacement detection in optical tweezers,” Optics Letters 23(1), 7–9 (1998).
doi.org/10.1364/OL.23.000007K. C. Neuman, S. M. Block, “Optical trapping,” Review of Scientific Instruments 75(9), 2787–2809 (2004).
doi.org/10.1063/1.1785844K. Berg-Sørensen, H. Flyvbjerg, “Power spectrum analysis for optical tweezers,” Review of Scientific Instruments 75(3), 594–612 (2004).
doi.org/10.1063/1.1645654S. O. R. Moheimani, A. J. Fleming, Piezoelectric Transducers for Vibration Control and Damping, Springer (2006); see also S. Devasia, E. Eleftheriou, S. O. R. Moheimani, “A survey of control issues in nanopositioning,” IEEE Trans. Control Syst. Technol. 15(5), 802–823 (2007).
doi.org/10.1109/TCST.2007.903345W. Kester (ed.), “Understand SINAD, ENOB, SNR, THD, THD + N, and SFDR,” Analog Devices Tutorial MT-003; and Data Conversion Handbook, Newnes (2005).
www.analog.com/media/en/training-seminars/tutorials/MT-003.pdfP. Horowitz, W. Hill, The Art of Electronics, 3rd ed., Cambridge University Press (2015).
artofelectronics.net/This note is educational. The simulation uses a simplified, representative model of the signal chain; board specifications are indicative and should be confirmed against the datasheet for your configuration.