header image
sub header image
header divider
create holograms header



SB Transmission



SB Reflection

MB Transmission

MB Reflection




Testing for Vibrations

I recommend that you read this whole section on Testing for Vibrations before you proceed with setting up the interferometer. At the end of this section are two important notes: one dealing with fine adjustments and the other dealing with a technique called retro-reflection.

The first optical arrangement you will set up is a Michelson interferometer as shown in Figure 14a. The interferometer produces a bulls-eye pattern on the screen S called an interference fringe pattern. Its purpose is to visibly detect, with your eyes, any vibrations that occur on the table surface that are emanating from the floor and any movements that can occur in an optical setup's components (the black fringes, as shown in Figure 14g, move if there are vibrations and/or component movements present). Once you understand what causes the interference fringe patterns to move, you will be able to avoid these vibrations and movements during your exposures. In a multi-beam optical setup, all components from the beamsplitter to the object scene and plate holder (including the beamsplitter, object scene, and plate holder), cannot move relative to each other during the exposure. In a single beam optical setup, no movement can occur between the object scene and the plate holder. The bottom line: if the fringe pattern is not moving, you'll have a successful hologram. Because of limited table space, having an interferometer set up during your exposures is not practical. You will now set up the Michelson interferometer.

Michelson interferometer setup image
Figure 14a: Michelson interferometer: laser (L), diverging lens (DL),
beamsplitter (BS), mirrors (M1 & M2), and screen (S).

Viewing the optical arrangement in Figure 14a, position the laser L at one end of the table's longest length and centered and place the laser high enough on the table so its output aperture beam is at a height of 9 inches. The beam will travel to a beamsplitter BS where it is split into two beams (a 50/50 plate beamsplitter works great for this setup and the coated side of the beamsplitter should face the incoming beam). Leave the diverging lens DL out of the setup for now. One beam is transmitted through the beamsplitter to mirror M2 and the other beam is reflected at 90 degrees to the right to mirror M1. The distance, or path length as it is called, between each mirror and the beamsplitter should be almost the same, to within 1/8 inch. If you make them both exactly the same, you're bulls-eye pattern may be distorted. These path lengths can be determined using a tape measure and should be as long as possible for the table size. The interferometer's sensitivity increases the further the mirrors are from the beamsplitter. Both mirrors should then reflect their beams back to the beamsplitter and strike the beamsplitter at the original incident beam's position. Part of mirror M2's reflected beam will then be reflected by the beamsplitter to the screen S and part of mirror M1's reflected beam will be transmitted through the beamsplitter to the screen S. Two beam dots should be visible on the screen as shown in Figures 14b and 14c. The screen S is a piece of 4 inch x 5 inch white mounting board placed in the plate holder.

Note: Make sure the center of the beamsplitter, both mirrors, and the screen are 9 inches above the table. Use the retro-reflection technique to insure the beam is 9 inches above the table throughout the setup.

Your goal is to superimpose (overlap) these two dots so that the center of the interference fringe pattern (center of the bulls-eye) is visible as shown in Figure 14g. The dots can be closely superimposed by grossly moving mirror M1 slightly up and down, and/or sideways by adjusting the mirror mount in its connector and/or moving the lead base of the table mount, respectively. It is best at this point to have just one beam dot barely overlapping the other beam dot or having their edges touch. Try to get the two beams as horizontally level to each other as you can. In Figure 14b, the left dot is slightly higher than the right dot. They should have been more level. This is not critical, but it will make your final adjustments a bit easier. You'll be refining this momentarily using a fine adjustment technique.

Two beam dots on screen image    Two beam dots on screen image
Figure 14b: Two beam dots on screen            Figure 14c: Two beam dots on screen
with room lights on.                                            with room lights off.

Once you get the two dots somewhat overlapped as discussed above, you will now need to magnify the dots to see the fringe pattern. You do this by using a diverging lens (DL). A plano- or double-concave lens with a -15 mm focal length and 12 mm diameter or a 10x microscope objective is suitable for this purpose. There are two possible positions in the optical setup where you can place the lens: either between the beamsplitter and the screen, or between the laser and the beamsplitter. You will start by placing the lens between the beamsplitter and the screen. This position will enlarge the dots significantly on the screen S and allow you to see the fringe pattern easily as shown in Figures 14d and 14e. By enlarging the dots this way, you can easily adjust mirror M1 to help you completely overlap the two dots (later on, the final position of the diverging lens will be placed between the laser and beamsplitter to de-magnify the fringe pattern so you can see the whole, centered bulls-eye fringe pattern).

Magnified beam dots on screen image    Magnified beam dots on screen image
Figure 14d: Magnified beam dots with            Figure 14e: Magnified beam dots with
fringe patterns, room lights on.                        fringe patterns, room lights off.

Because of the huge magnification of the fringes with the lens at this position, it is not possible to see the whole bulls-eye pattern. The fringe patterns you’re seeing in the above figures are the outer edges of the bulls-eye pattern, not the center of the bulls-eye pattern. Again, adjust mirror M1 to get these two dots more overlapped (this is the moment to use the fine adjustment technique covered in the note on fine adjustments covered towards the end of this section on vibrations). As you get them more overlapped and approach the center of the bulls-eye pattern, the fringes become fatter and less numerous as shown in Figure 14f. Once you reach this point, let the table and components settle down for a few seconds so the fringe pattern is not moving or barely moving, then touch the table and watch the fringes move. You'll notice how very sensitive the interferometer is when the table is touched.

Large interference fringe patterns image
Figure 14f: Fringe patterns closer to the
center of the bulls-eye pattern.

To see the whole bulls-eye pattern, you need to place the diverging lens between the laser and the beamsplitter, making sure the diverging beam passes through the beamsplitter and is reflected back to the beamsplitter from both mirrors to the screen. You can move the lens around until you see the divergent beam surrounding the beamsplitter on the screen as shown in Figure 14g. You should now see the whole bulls-eye pattern within the beamsplitter's shadow on the screen. If you don’t, move mirror M1 until the bulls-eye pattern is centered using the fine adjustment technique. With the whole bulls-eye pattern visible, it is easier to analysis what is causing the fringe pattern to move, as discussed next.

Bulls-eye pattern centered image
Figure 14g: Bulls-eye interference fringe pattern centered.

These interference fringe patterns can be used to analyze any vibrations occurring on the table surface and/or any movement of the optical components. There are three types of fringe movement:

By studying the movements of your fringe patterns and relating the movements to causes in your environment and components, you'll get a good feel for what the most favorable conditions will be when making your exposures. The only thing I can't control is traffic driving past my house. This is why I usually make my exposures in the evening between 10 pm and midnight. The bottom line is: the quieter the table surface, the brighter the hologram.

Note: Fine Adjustment

To get the center of the bulls-eye pattern exactly in the center of the beamsplitter's shadow on the screen takes some practice and fine tuning. Seeing the center of the bulls-eye pattern in the center of the beamsplitter's shadow on the screen means you have both dots exactly overlapped. Make sure the diverging lens has been placed between the laser and beamsplitter. Figure 14g shows how the bulls-eye pattern is no longer magnified with the lens in this position and you can see more of the bulls-eye pattern.

To get the bulls-eye centered, I work with just one of the mirrors and its mount. Referring to the setup in Figure 14a, I would work with mirror M1. The reason I work with this mirror is because if you move the mirror right or left, the pattern on the screen moves the same direction. This is also true when moving the mirror up or down. The reason for this is because the reflected laser light from mirror M1 goes straight through the beamsplitter to the screen. Because the reflected beam from mirror M2 is reflected (not transmitted) to the screen from the beamsplitter, right and left positions switch. So if you move M2 right, the screen pattern moves left. This causes a bit of confusion and who needs confusion with everything else you're doing here.

Before you start to move mirror M1, first look at the curvature of fringe pattern on the screen. The center of the bulls-eye is in the direction of the fringe pattern's concave curvature. As an example, if the concave curvature is to the right, so is the center of the bulls-eye as shown in Figure 14h. Additionally, the width of the fringes gets thicker in that direction.

Concave curvature image
Figure 14h: Concave curvature of the fringe pattern is to the right.

To move the center of the bulls-eye to the left, you need to move mirror M1 to the left. To move the mirror gently and slightly, tap the front right side of the lead weight slightly with your index finger as shown in Figure 14i.

Tapping lead weight image
Figure 14i: Moving mirror M1 left with index finger.

Repeat this tapping gently until you see the center, or portion of the center, of the bulls-eye as shown in Figure 14j. Continue tapping until the bulls-eye is completely centered as shown in Figure 14g. If the concave curvature started out to the left, then gently tap the front left side of the lead weight to move the bulls-eye pattern to the right and center.

Center of bulls-eye pattern moving left image
Figure 14j: Center of bulls-eye moving left with tapping.

If the concave curvature is facing upward, you need to move the mirror itself downward to move the center of the bulls-eye downward. To achieve this, I grab the acrylic base of the mirror mount by its ends with my thumb and index finger and rotate the whole mirror mount clockwise downwards around the 8-32 short rod bolt as shown in Figure 14k. This is a very slight rotation. If the curvature is facing downward, you need to rotate the mirror mount upward or counterclockwise. Again, this is a very slight rotation. Since you're rotating the optic mount counterclockwise, you're actually loosing the mount from the 8-32 bolt in the short rod. You don't want to loosen the mount from the bolt so much that it is no longer rigidly attached to the bolt. Loosening the connector that holds the short rod to the table mount pole won't work since that is a gross adjustment.

Moving bulls-eye pattern up or down image
Figure 14k: Rotating mirror up or down.

Sometimes the concave curvature faces a diagonal direction towards a corner of the beamsplitter shadow. In this case, you would have to make both up or down and left or right movements of the mirror.

Note: Retro-reflection

All of the optical setups on this site will almost always have the laser output aperture at 9 inches above the table surface. There are two reasons for this:

  • in the multi-beam white light reflection hologram setup, the reference beam will need to impinge on the recording plate from below the 9 inch high plane at an angle of 56 degrees. If the center of the recording plate was just above the table surface, this would not be possible.
  • in all the optical setups, the laser beam should always be parallel with the table surface and with each other to make sure that the polarized direction of the laser's beam is the same for all the beams traveling around the table. The importance of the polarized direction of will be discussed later.

To insure that all the beams are parallel with the table surface as they traveling around the table from one component to the next, you will use a technique called retro-reflection. Looking back at Figure 14a, as you position mirror M2, adjust the reflected beam from M2 back through the beamsplitter to hit the laser aperture just 3-4 mm to its right or left (but not back down the laser tube) as shown in Figure 14L. Do the same with mirror M1. This will guarantee that the all the beams are parallel with the table surface and 9 inches about the table. I'll cover retro-reflection in more detail when you set up your first single-beam transmission hologram arrangement.

Moving bulls-eye pattern up or down image
Figure 14L: Retro-reflecting laser beam back to laser aperture.


Revised 5/2/2017