What tools do I need to calibrate a colorimeter?

As I am not very familiar with tools used in optics, I need some help. I want to develop a colorimeter which will perform the following functions:

-to be able to calibrate a monitor (read RGB values using primary color on display)
-to be able to calibrate printer outputs
-to be able to read white point in a certain environment.

I have a sensor with RGB outputs and white led included. My question is what tool do I need to calibrate such a device in LAB space? Basically I think I need standard iluminants for primary colors, but I am not sure what device should I look for. A bit of directions would help a lot. Thank you.

/pisoiu

Hi,
You need in all cases calibrated "White" light _ and color standards or you must give your color measurement system to calibrate in a professional laboratory!
Problem is, that "white1" is not = "white2"!!
You can buy so an instrument i.e. from Minolta over 20 years, & I think from Leader too...
Check for colorimeter i.e. in automovi production...
It exits some standardrs for lighting the surfaces, its a NOT SO EASY subject!
K.
Because Mazet is one color sensor source in Europe, maybe they hase applications scripts & theory stuff over them subject. Of course Minolta, & some smaller German firms too...

Totally agree with you, it is not an easy subject. Yes, white illuminant like D50 or D65 will be required, but I think it is not enough. I think it is required a device which is able to generate also individual wavelengths, at least for the primary colors, with various intensities. I need to check the linearity of the sensor, and I cannot do this with only one intensity of the illuminant. Also, the gamut is important, because I need to calibrate the sensor for at least Adobe RGB color space. For display calibration and for white point indication, an ideal tool would be something (a surface) to stick the sensor, and on that surface a light is generated, with known coordinates in LAB space.

Oha!
You need a lot of investition! Do you have an optical spectrometer?
I snow ca. 25 years ago at a small firm in München different White standards....They are Ceramics and the differents between two examples___makes depressive!!
Im no more active in optical solutions, but I would search on Optical/Laser magazines & society sides for my infos, applications
For exact calibration you need an "Ulbricht Kugel", not some surface! Surfaces are for measurements/semi calibrations-or better; controls- declared...
K.

First, thank you for reply, it is a very good starting point, and after few days of digging in optics, it slowly begins to make some light in this subject. I see the need of an integrating sphere, however the biggest question for me so far is how to determine the exact color information of a source in LAB space. I mean, by using an integrating sphere, at the input port will be a light source, at one of the output ports will be my device and at another will be a photodetector. What type of photodetector should I use is still unclear. It should be able to detect the intensity of the received light as well as the spectral composition. Otherwise I do not see how can I compute color information.

First of all, don't confuse a colorimeter with a spectrophotometer. A colorimeter uses a set of wide-band filters (typically 3 filters) to generate CIE XYZ data. A spectrophotometer breaks the light into narrow-bands, typically 5mm, 10mm or 20nm wide. This may be done using narrow-band filters or a diffraction grating. XYZ data is also derived from the output of this sensor. With either instrument, the XYZ data is convoluted with that of a light source to produce data in a standard color space such as CIELAB, RGB, CMYK - or any of a host of others.

Both instruments can be made to operate in the Transmission mode or the Reflection mode - some instruments are designed to operate in either mode. In transmission mode the instrument measures the light emitted by a light source. It can be measured directly, or used to measure the transmittance of a material (like colored glass). In reflection mode the instrument’s light source illuminates a surface and the reflected light is measured.

There are several standard illumination geometries used. The most common are 0/45 degree and integrating sphere. In theory, the integrating sphere illuminates the sample with light at every angle. The 0/45 geometry has the light source and detector at a 45 degree angle from each other.

I'm not sure what your goal is, but producing an instrument that will measure a monitor and a printed image with reasonable accuracy and repeatability is a huge endeavor. You need to develop a set of "good' optics, an accurate detector and the mathematics to convert the data to useful numbers.

Thank you for reply. Perhaps I was not very clear with what I want to achieve, I will try to explain a bit better. I want to build a prototype of a device (maybe in the future I will produce in series) with the following functions:

-to be able to calibrate a computer monitor, like a classic spyder colorimeter for example and to generate calibration file for ICM engine in windows or equivalent for mac
-to be able to calibrate printer output, (like spyder studio from datacolor)
-to be able to read ambient light color temperature and indicate it in kelvin (like sekonic c-500 color meter for example)
-to be able to calculate flash output (likesekonic l-358 flashmeter)

The sensor selected for this use is this one: https://ro.farnell.com/jsp/search/productdetail.jsp?SKU=1618712 , but of course, if it proves not suited for my needs in the future, I may change it.

My first goal is to determine if this is achievable. The problem would not be the tools/expertise required in electronics and mechanics, here I have quite good expertise and resources available. The problem would be the tools required by optic part and all associated problems, color calibration, etc.

Hi pisu!
I become yesterday a electronics magazine, & I belive to found in it for you the best solution of "RGB-Calibrating"...
Check pls www.instrumentsystems.de, "Serie ACS-530" Temp stabilized LED-Modules, CIE conform to 0.1%)...
Good Luck!
Karesz