Pixel Scaling Trends
Explore how key image sensor performance metrics scale as pixel pitch shrinks, with reference data from commercial sensors.
Pixel Scaling Trends
Explore how key sensor metrics scale with pixel pitch. Compare theoretical scaling laws against published sensor data.
Technology Node:
Full Well Capacity (ke⁻)
Max SNR (dB)
Nyquist Frequency (lp/mm)
Diffraction QE (%)
FWC
10000 e⁻
Max SNR
40.0 dB
Nyquist
500 lp/mm
Diff. QE
28.3%
Dark Current @60°C
5.0 e⁻/s
Physics
Pixel miniaturization involves fundamental trade-offs governed by geometry and diffraction:
Scaling Laws
- Full Well Capacity: FWC is proportional to pitch squared (area), so halving the pitch reduces FWC by ~4x, directly impacting dynamic range.
- Signal-to-Noise Ratio: For photon-shot-noise-limited conditions, SNR scales with pitch (proportional to the square root of the signal).
- Nyquist frequency: f_Nyquist = 1/(2 x pitch) increases with smaller pixels, improving spatial resolution up to the diffraction limit.
- Diffraction-limited QE: As the pixel aperture approaches the Airy disk diameter (~2.44 x lambda x F/#), diffraction causes light to spill into neighboring pixels, reducing effective QE.
Practical Implications
Modern smartphone sensors use pixels from 0.56 um to 1.0 um. At these sizes, microlens optimization, deep-trench isolation, and computational photography become essential to maintain acceptable image quality despite the fundamental scaling penalties.
INFO
Reference data points are approximate values from published sensor specifications and serve as a guide for understanding trends across technology generations.