Tuesday, December 08, 2020

Omnivision's Sensors Tested with 1Vpp Supply Noise

2016 MSc Thesis "Impact of Power Supply Noise on Image Sensor Performance in Automotive Applications" by Shane Gilroy from Waterford Institute of Technology, Ireland, tests 1MP and VGA Omnivision sensors row noise when applying 1Vpp noise on the supply rail. Surprisingly, the sensors are able to work and output a reasonably good images for such a huge supply noise:


  1. the conclusion from thesis is not this. I believe those old products do not have good features of PSRR. And I can't believe this could be a thesis of master level, maybe just a course project level.

    1. Ok - I do understand your low opinion of the thesis. It indeed misses any proper analysis and basically is just a measurement report of a device the student clearly has no insight into. I blame the advisor for picking such a weird topic to start with! As with respect to your critique on the sensor's PSRR. That is clearly unfounded given the data that suggests otherwise. So if you criticize a product, please claim reproducible data rather than just wildly bashing out - that seems to suggest that this critique is rather personal than technical.

    2. I think the academical quality of a thesis is not the point here? The presented fact that the sensor gives you a decent image with 1v 'ripple' on the supply is indeed remarkable. Many sensors will give you the weirdest stripes when there is only 10s of mV of ripple.


All comments are moderated to avoid spam and personal attacks.