Additional use for Fringe Locker
Correcting steel table thermal drift with a fringe locker and auxiliary laser
A fringe locker is usually used to sense the recording wavelength fringes at the film plane and feedback correction to a moving mirror or moving prism in one leg of the set up. This works best and should always be tried first. Some set ups require the use of scarce blue photons which are not readily absorbed by ordinary silicon detectors. The power available for fringe locking is sometimes frightfully low and a "hard" lock is not possible. One solution is to buy more blue photons, another is to buy more sensitive photo diodes or photo transistors and install them in the head, another is to concentrate the light at the detectors with a lens, and still another would be to add more gain at the input stage.
Even when there is enough light to make the locker work, there may be a diffuser or a diffuse object somewhere in the set up and therefore a diffuse wave that will not make a clean high contrast set of fringes to lock onto. For this case a few extra optics will often provide a second specular beam to use to form clean fringes. If none of these options are possible or attractive then there is still one more option that may work that requires several more optical mounts and optics but no more precious blue photons. For this option you need a simple 1 mw red HeNe or stabilized red or near IR diode laser as a secondary laser and enough optics to zig and zag down the same approximate paths taken by the primary blue laser light. The secondary light is required to share only one common surface with the primary light and it can usually remain on all the time and may even be considered part of the safe light fixturing.
The object of using the locker in the first place is to compensate for long term drift in the table.This is not always the case but it is the most common task required and the most difficult problem related to long exposures. The components themselves may also warp slowly during an exposure but nothing much can be done about that. When the components are tall or raised above the tabletop a significant distance then table bowing may be a more serious problem than simple linear changes. In this case it is best to mount the beam combiner high up on the film plane, perhaps 2/3to 3/4 of the distance from the table top to the top of the film holder or film or plate. This positioning produces a good compromise between bow and linear compensation and will work when using primary laser light or secondary laser light. The only surface that has to be common to both primary and secondary waves is the moving mirror on the bimorph but it is also a good idea to mount the secondary wave combiner on the film plane whenever possible. This will make two surfaces essentially common and if the splitter can also be common, then so much the better.
The above drawing shows a generalized locking scheme using a second red laser. The paths taken by the red light are roughly parallel to and of the same length as those taken by the blue light.Feedback from the red fringes formed by the combiner, BS3, will move the bimorph and change the path lengths for both colors simultaneously. Gross table drift will be adequately canceled but local or component drift and warpage will be unchanged. This system is very general and it would be fair to change the set up to record rainbow holograms and even use some reflection recording geometries in the blue zone shown without changing the red zone arrangement at all. Of course the acid test is to shoot a few shots and check for movement fringes and repeatability.
The general reflection hologram is something of a special case. Fringe sensitivity to path length changes is at a maximum, often 10 times more sensitive than a transmission set up so some special precautions may have to be taken to get adequate stability. One precaution would be to follow the primary path more closely than shown, perhaps even mounting the red mirrors on the same posts with the blue mirrors whenever possible. Another precaution would be to calculate the correct angle of incidence of the red light on the bimorph mirror. The red angle will be different than the blue angle but related by the difference in wavelengths. The fringes in red are larger than the fringes in blue and the correct motion of the bimorph to compensate for a wave of error in the blue is smaller than for a wave of error in the red. The drawing shows the red beam at a shallow angle with respect to the normal of the bimorph mirror and the blue beam at a larger angle. In general this will be correct because the largest wavefront displacement occurs at 0 degrees and the smallest at grazing incidence. For small angles of incidence, say 40 degrees or less for the blue, the approximate angle for the red can be found by multiplying the blue angle by the ratio of wavelengths. As an example, if you were using a HeCd and a HeNe, as many holographers do, and your 442 light was set at 40 degrees from the normal, then the 633 light would be at442/633*40=28 degrees. This is only 1.3 degrees from the correct value or about 5% error,which is not enough to cause trouble.
This little trick could also be compared to bathing a table in a constant temperature oil bath,which is sort of the brute force method of compensating for table drift. It might also be compared to buying a $30,000 research grade invar table, which also compensates for thermal drift by not having any to begin with. In any case, no matter how you look at it, if you have a thermal drift problem, and it is primarily the steel table that is growing 1 or 2 microns in a 20 minute exposure,then lock those fringes and relax. If you can't lock them directly then at least lock down the section of the table that you are using by building a secondary locking circuit. If you need a locker, I know of two manufacturers in this area, I haven't made a locker myself since 1983 but I use one frequently and consider it essential absolutely gotta have tooling.
Fringe locker manufacturers:
- Inovar Devices inc,
ph (435) 245-5061, fax (435) 245-6948
- Excalibur Engineering,
ph (435) 755-9221, fax (435) 755-9321
Last modified on 8/14/98