RSS Feeds

Email a Friend Printer Friendly

Splitting Of White Light

April 23, 2004

Moscow scientists have managed to do simply and inexpensively something which normally proves complicated and expensive. The concept thought out and then implemented is a device which allows you to check the quality of ground and polished surfaces with unprecedented precision and rapidity and to detect every single defect of such surfaces. The effort has been funded by both the Russian Foundation for Basic Research and the Foundation for Promotion of Small-Size Enterprises in Research and Technical Areas.

The source of light for contemporary devices for the check of the quality of grinded surfaces are lasers, which are not cheap, but the correct interpretation of measurement results could be produced only by a specialist with extensive experience and intuition. The fact is that the interference acquired in this case does not allow to identify the type of defect - be it convexity or concavity. Now the situation can drastically change. The laser turns out to be successfully replaced by a common white light bulb, and intuition will not be needed at all. Therefore, the device will provide distinct results at the output, and there will be no need to guess what defects exist on the mirror surface.

The new device is called "digital interferometer with incoherent source". The device was invented and manufactured by the specialists of the Moscow Physical & Engineering Institute (State University), Moscow State Engineering University STANKIN and Research-and-Production Association "Energomekhanika" with financial support of the Russian Foundation for Basic Research and the Foundation for promotion of small-size enterprises in research and technical areas. The project was headed by Nikolai Vlasov, Professor, Doctor of Technical Sciences.

"The device action is based on interference of common white light (and this is reflected in its name). The essence of the work is as follows, says Professor Vlasov. The beam of light from one source is split into two absolutely identical beams. They are called the reference one and the object one. One of them is directed at the mirror which is initially of high-quality. The other is directed at the object under consideration, the distances to the object being equal. That is also the mirror, the surface quality of which should also be checked for concavities and convexities, and measured them, if any. If the " heights" or "hollows" are too extensive and exceed the allowable value - the mirror is culled, if the defects are nonexistent or insignificant - the mirror is permitted for work."

However, the question arises how the defects can be measured? So, both beams are reflected, each from its mirror. Then they are joined with the help of a semitransparent mirror to see what would turn out. Each of the beams has passed its way, the difference between the ways being the deflection of the object surface height from flat surface: if the beam hits a concavitiy, the length of its way increases, if it hits a convexity - its way decreases by the height of the convexity.

If the difference is divisible by the length of the light length, it intensifies - this phenomenon is called interference. If not - one beam extinguishes the other. The so-called fringe pattern is obtained - i.e. light and dark bands. Their number equals the number of times the light wave-length "goes into" the difference of ways. As the light source is normally the laser beam of a certain wave-length, this distance can be easily calculated. Nevertheless - and this is very important - the position of the bands would not help to determine whether the way passed by the beam reflected from the objects was longer or shorter. That is, if the mirror is convex or, on the contrary, concave.

For this purpose, the operates has to rely upon his/her experience and intuition. For example, when polishing the mirror, the pressure is higher at the middle of it - that means that the "relief drop" can be expected in that area. The sides of the mirror experience lower pressure - that means that convexities can be found there. That cannot be expressed in numbers.

That particular task can be fulfilled by a new device. If the lengths of the ways gone by light are absolutely identical, then regardless of the its wave-length, interference occurs upon overlapping of identical beams - since the difference of legths turns out to equal zero. The interferogram shows bright contrast band surrounded by the so-called black fringe - it is called the zero order band. That means that the only thing to be done is to make the two ways of two beams equal. To this end, it is sufficient to slightly shift the "correct" flat sample mirror at a certain distance in the known direction - either closer or farther away. That is performed by a special mechanism.

So, the method is as follows: the light goes in two beams, the fringe pattern is recorded - i.e. it is photographed, digitized and the zero order interference band is input in the computer memory. Then the mirror is relocated by one step and the procedure is repeated. The mirror is shifted to and fro. Several step-by-step images are acquired - i.e. a series of interference bands is obtained, processed and summed up. As a result, the answer if obtained in numbers whether the way length of the beam reflected from the sample mirror is longer or shorter that that of its twin reflected from the mirror that is the object of investigation. Therefore, the mirror quality has been checked. This method is very simple and precise.

"Certainly, our method is not deprived of drawbacks, says Nikolai Vlasov. For example, it cannot compete with the laser one if a nonstationary object needs to be investigated - we require some time. However, our method has no equal for an enormous number of common measurements and polishing quality control."

Informnauka (Informscience) Agency