| Author | Message | ||
     Michel Coene (Michel) |
Ok guys, forget about grain for a moment; here is a funny one. A factory produces a fine, endless plate of stainless steel. Before heat treatment, it is degreased because the grease burns on the steel in the oven. The degreasing however, is far from perfect. The idea now is to install a freespace system behind the degreaser to check for _small_ traces of oil. The freespace system (Brimrose) sends a perfect vertical lightbeam straight down onto the steel. Under an angle, a big collector lens focusses scattered light on a detector. You can feel the problems comming: as the plate is very new and very clean, it is an almost perfect mirror which will be reflecting 99.99% of the light straight up again back into the AOTF. To increase our chances, I was thinking of sticking a mirror to the spectrometer, with the shiny side facing down to the steel plate, and two holes in it for resp. the outgoing and incomming light beam. This would (I hope)make the light bounce back and forth between plate and mirror, picking up some extra signal every time it hits the plate. Am I dreaming here? Any recommendations? + the environment is reasonably clean + the plate is very stable (on a 1 m diameter roll) + time is of no importance, so I can scan very long | ||
| Bruce H. Campbell (Campclan) |
Michel, Your use of a mirror sounds interesting. It may work. I would think you would have to have a curved mirror, one that would focus the reflections from it back onto the steel and then into the collection "hole." I can think of two alternatives you may want to consider. One would be to have the NIR beam at a small grazing angle. This would increase the effective pathlength and not have to deal with the reflections. The collection system would be also then at the same angle on the other side of the beam. The second alternative would be to use Raman with a tightly focused beam. The reflected specular light would be then discarded. With tight focusing, the effective sample volume would be very small and detection levels may be adequate for your purpose. Bruce | ||
| hlmark |
Mike - I like your idea for multiple-bouncing of the beam off the sample, since that will give you multiple passes through the sample, which will enhance the sample absorbance. The downside is that it will give relatively low energy throughput and since it may prove difficult to align, you might get different numbers of bounces at different times, due to alignment differences. Another possibility is to use a beamsplitter. This will give you much more raw energy (and with it, higher S/N ratio), but will work only if a single plass through the sample gives enough absorbance. Howard | ||
| Lois Weyer |
Mike - Brimrose has a Brewster angle system which I once used for a thin film on a metal roller. It illuminates the product and collects the signal at Brewster's angle instead of 90 degrees to eliminate the interference fringe. It might also work for your application, though. Lois | ||
| Gabi Levin |
Hi Michel, I saw your question. Two things come to mind: 1. Small residue of grease may have little absorption of NIR due to very small path length and low inherent absorptions. 2. Our Free Space avoids specular reflection, but that goes against you becase your surface is specular reflector. Thus, Lois is correct in susggesting the film analyzer. This is designed to collect light from the correct grazing qangle, eliminating interference fringes. Additional factor, our original design was for the MID-IR i.e., longer wavelengths, (2.5-3.7 micron) giving rise to higher absorptions, providing better signal. Thus, if your grease has signature in this region you will experience a better detection level. The next critical issue is the question if the expected residue is continuous or patchy. If it is continuous then you have all the time in the world to detect it and provide alarm. If it is patchy you may need speed to detect the patch. I suggest using the MID-IR film analyzer, scanning over minimum required range, speeding up the detection. I hope I can be of additional help at your discretion. Gabi |