What's new
DPR Forum

Welcome to the Friendly Aisles!
DPRF is a spin-off of dpreview. We are a photography forum with people from all over the world freely sharing their knowledge and love of photography. Everybody is welcome, from beginners to the experienced professional. From smartphone to Medium Format.

DPRF is a community for everybody, every brand and every sensor format. Digital and film.
Enjoy this modern, easy to use software. Look also at our Reviews & Gallery!

Spectral Quantum Efficiency of the Olympus TG-6 Waterproof PS camera

B

Bernard Delley

Member
The TG-6 is a handy waterproof PS camera. I am glad it is still around in the dwindling market of PS cameras. It has not been updated since its appearance in 2019? With its 12 Mpixel BSI-CMOS sensor it remains at reasonably current technology. The sensor is 6.2 mm wide. A larger sensor with the same maximum equivalent focal length of 100mm for the 4x zoom might be really be hard to squeeze into a waterproof camera of its handy size. For the present measurement it is crucial that this camera supports true raw output: no Bayer demosaic, no raw preprocessing to speak of.

Quantum efficiency is the number of electrons produced per photon falling on the pixel area. It is a statistical average number smaller than 1.
Details of the measurement and calibrations will be discussed in the follow up.

Here is the result of my preliminary measurement:


Olympys_TG-6_QEprelim.png

~
 
The basic setup scheme is shown below. Positioning the transmission grating (500 lines/mm) flat on the lens (normal to the optical axis) results in an almost perfectly linear projection of the visible spectrum to the sensor with less than 2nm error across the spectrum. The final calibration can be made by replacing the broadband halogen light source by laser pointers. I used 405 532 and 656 nm pointers.
g-spectrum_scheme_rev.png
 
Another calibration is needed to relate the number of electrons in the pixel to the reported RAW number for it. This is done with the statistical evaluation of the measurements going int o the Photon Transfer Curve. TRhe parameters derived from this are inserted in the fig. cgain ; 2 electrons correspond to 1 digital raw number increment. There is also the black level, the Full Well capacity and the read noise, as well as response non-uniformity.

Olympys_TG-6_PTC.png
 
Is this camera using the Sony IMX477? I note that there is an Arduino Kit camera using this Sony sensor. This information would be relevant to it as well.
 
The calibration on how many photons fell on the pixel during the exposure time is the most challenging. One cannot simply place a photo-spectrometer at the pixel position and measure the photons in PS camera. I use a full frame lens in a proxy dummy camera where the X-Rite iStudio 1 photo-spectrometer is moved along the sensor with a micro meter screw. This leaves a number of factors to be considered, and tends to factor in calibration errors. One thing that could have been measured in principle, but did net get done by now, is the transmission of the proxy lens. It is a high transmission more or less near transmission one. For this reason one must call this measurement preliminary for sure.
Including this calibration one could call the result "External Spectral Quantum Efficiency" where all the photon losses due to the PS lens and the sensor stack and due to any reflections would still be counted as part of the input. The main part of this calibration is the spectral transmission of the grating into the diffraction order that gets used. This is shown below:
spectral_calibration.png
 
Brian said: Is this camera using the Sony IMX477? I note that there is an Arduino Kit camera using this Sony sensor. This information would be relevant to it as well.

I have no info on the exact chip. I just wanted to know better what I have got with this camera. The early and sharp IR cutoff for this camera is new to me. The level of QE looks quite reasonable and plausible to me, also the general spectral shape.
If one has the Arduino camera, then similar measurements could be done, or easier would you know where Sony has published its spectral response ?
The smallness of the sensors would still require a proxy dummy camera in bigger format, unless one has a specialized highly sensitive photo-spectrometer with fiber probe.
 
Found a datasheet:
IMX477 is a little larger sensor than in the TG-6. Sony is very cautious showing only Relative Sensitivity (Includes neither lens characteristics nor light source characteristics).
It is peak normalized to 1 for green and gives thus no information on Quantum Efficiency. I seem not to see all pages of the data sheet. The spectral response of the Sony goes deeper into the red than TG-6
 
Thankyou for doing this- "Way Back When" I spent a lot of time converting measurements taken on Film to "Watts/Steradian" (1970s), and the 1980s working with custom made Digital IR sensors, writing code and doing the experiments to calibrate them. That was 40 years ago.

These days- can drive you crazy when manufacturers will not release the data sheets and treat the sensor used as some trade secret. Kodak published all their data sheets, including those for the Leica M8 and M9. BUT- I know the Arduino uses the Sony, and the data sheets give a start. The Black Level of the Sony looks very close to your measurements.

On the Leica- I end up rewriting the Black Level stored in the DNG file- the sensor was much more sensitive than what the Camera gave it credit for. Picked up more shadow detail,
 
You must log in or register to reply here.
Back
Top