scanning michelson interferometer

[mixedgas]

they are not state of the art, but linear arrays from TAOS (texas advanced optronic solutions ) are easy to clock. They have a nice demo board for 99$, but you just need 3 pieces of cmos to do the job.

The CW burleigh wavemeters I've seen move a right angle prism used like a corner cube 1-2 cm back and forth on a flexure pivot made of two pieces of spring steel. They have a little hene in them for the reference.

I scrapped a bunch of FTIR machines at my former employer, the mechanism was always a large diameter light weight voice coil on flexures. The processing power in a IR-FTIR was always dismal, maybe a 68030 or 68040 at best with a little dram. We scrapped them only when the sparcstation side of the software was no longer supportable. For the most cases the ad conversion was only 8 or 12 bit. So should be doable in a pic. I only mentioned the fring counting because its easy.

Feel like playing the lottery? Looks like a corner cube to me: http://www.goldmine-elec-products.co...?number=G16072

try the usual suspects, all electronics, surplushed, alltronics, electronic goldmine...

Steve

[RedlumX]

The first version of my CCD spectrometer indeed used a linear array equivalent to the TSL1402 from Taos, and yes, the necessary extra circuitry is very simple. But its resolution was not enough for me, ie, didnt match the capabilities of the grating. I remember it has something like 50nm pixel size only. For simple applications this may be fine, though. But I would recommend something with more than 256 pixels, in fact as many as possible. It must fit, of course, to the grating at hand; the grating is the bottleneck for this kind of projects ;-(

[mixedgas]

The first version of my CCD spectrometer indeed used a linear array equivalent to the TSL1402 from Taos, and yes, the necessary extra circuitry is very simple. But its resolution was not enough for me, ie, didnt match the capabilities of the grating. I remember it has something like 50nm pixel size only. For simple applications this may be fine, though. But I would recommend something with more than 256 pixels, in fact as many as possible. It must fit, of course, to the grating at hand; the grating is the bottleneck for this kind of projects ;-(

They make duals and tripples now, so no more 256 limit.

Steve

[abmikhatri]

Hi Shrad,

I am building a setup for FTIR in the range of 15um to 1mm, and found this thread online. I will be using Michelson interferometer with 90-degree off-axis parabolic mirrors for collimation.

I have to figure out the total mirror translation required for my spectral range (0.3THz to 20THz) and resolution of say 0.1 THz?

It will be really great if you can help.

Thank you!

Abdul