The idea is to build a not so expensive visible light spectrometer...
Already have an account? Log in.
To make the experience fit your profile, pick a username and tell us what interests you.
So, there are two versions for this spectrometer :
1 - The microcontroller version, which is cheaper but less efficient than the FPGA version.
2 - The FPGA version which allows to control timings more accuratly. It also has a more efficient ADC.
For the moment I used the optical bench from an old ocean optics spectrometer.
Well, I'm not an optical engineer so, if someone wants to help to build a cheaper optical bench, he is welcome !
Six years have passed since this project has been interrupted.
I made some progress this afternoon. Everything works fine.
Fill the doodle if you are interested in buying a PCB + components.
Survey will end october, 15. Then I send the PCB for production.
If you are interested in buying an IRIS board, register here :
http://doodle.com/poll/wc9ay3sep34uwekd
Price will be 100 euros without shipping costs including :
- PCB
- Components ( without TCD1304 )
( kits will be shipped in November, I'm slow... )
To be able to build the optical bench, I need :
- A slit ( http://www.babytronix.com/Slit-Mask.pdf ? )
- A software to be able to design the optical bench ( lensforge ? )
- Mechanical part to enclose the optical bench ( easy to do with a laser cutter and some plexiglas )
- a 40mmx10mm TEC to cool the CCD
- Mirror
- Gratings
Well, the bad news is that this project will sleep until december because I have too many others things to do... ( especially at work... ).
I plan to produce some board in september for people interested. I will open some doodle at the beginning of september to build a list for that.
The current setup allow me to acquire some spectra and the results are good enough to reach the initial goal of this project.
Anyway, I'm still very interested in building a much more accurate device but, to be able to to this, I really need help for the optical setup.
The entire design is on github, don't hesitate to use / modify it !
The minimum software has been developed.... I plan now to develop a v3 of this project to add :
- Peltier module for the CCD
- IOs
- My own optical bench ( if someone has a sugestion of software that allow simple optical simulation, I take it ! )
The version 2 of the IRIS spectrometer is works very well...
I use a number 13 grating from ocean optics.... since I'm only interested by the visible light wavelength, I plan to change the gratings...
I also plan to produce a second PCB with some IOs to be able to synchronize the acquisition with other subsystems.
But, well, I can say my main goal is complete ! ( github file will be updated asap )
Today I test the TCD1304. Everything works well ( I was able to capture a spectrum ). Anyway I need to change some timing in the VHDL code to have a better management of the CCD.
I add a figure of the noise coming from the grounded ADC input... Pretty good !
Next step : connect the TCD1304 !
- PC / FT2232 communication is OK !
- ADC sampling seems ok too...
- Multiple spectrum acquisition has been successfully simulated
Last step : spectrum acquisition !
:-)
I want to buy this assembled card. How should I approach for it ??? Please provide me the details of the same. Also let me know about the shipping charges involved.
Great project!
I'm also having problems finding components, however, I do have some slits. (I build 3/4 meter spectrometers and these were left over from units no longer made.) These were secondary slits but can prolly be used as a primary. Might need modified but better than 2 razor blades! http://www.babytronix.com/Slit-Mask.pdf
The Fastie-Ebert geometry won't work for me. The entrance slit orientation keeps me from having multiple optical cells while maintaining a 'straight in shot' (no fibers).
Thanks.
Well, I'm not an optician... This is a basic design to be able to acquire data from this CCD. Hope this project will be usefull if someone want to design his own spectrometer. This optical setup will be could for my needs. Anyway it could be very interesting trying to improve it :-)
count me in. I have some VSP3100 digitizers and NEC CCD arrays here but never mustered the courage to implement a proper spectrometer camera. What are your plans regarding cooling?
I worked with Andor spectrometer cameras, they can be cooled down to -80°C or lower which gets rid of all the nasty thermal noise (and then quantum efficiency starts to drop as well at some point ;-) )
It could be easy to add a Peltier cooler behind. TCD1304AP is given to works from -25°C. The main problem is to find a detector that is able to each such low temperature ( -80°C ). It's always possible to use hamamatsu S7031, not so cheap ( approx $130 ) but still possible and in this case we can reach -40°C.
Hi,
that's a great project you're working on.
The thing with the optical assembly is not so much one of getting the geometry right as it is to find the right components.
Planar gratings are the cheapest but toroid reflective gratings do away with additional mirrors (which also helps reduce stray light / reflections).
A good compromise is achieved in the Fastie-Ebert geometry (one spherical mirror, one small planar reflective grating) if you can work around the field curvature and low order aberrations.
j0hn
Roberto Innocenti
DIY GUY Chris
Michael Frank Taylor
Now drive Linear CCD, the best solution is rp2040