I have been thinking about how I can make a larger format Hologram without getting sucked into the long exposure vortex and here is what I came up with.
For a 8" x 10" Hologram do the following;
Put the plate in the plateholder and mask off all but a 5" x 7" portion of the plate. For reflections you will have to do both sides.
Make your reference beam Gaussian.
Have a transfer mirror in between the spatial filter and the plateholder which is adjustable along the x and y axis.
Adjust the mirror such that the expanding or collimated beam just covers the 5" x 7" part of the plate that is not covered.
Do and exposure.
Adjust the mask for another (larger then quarter) 5" x 7" area.
Adjust the mirror for the reference beam to cover that second exposed area. (there will be some overlap and this is wanted).
Object illumination does not change.
Expose.
Repeat for last two (larger then quarter) quarters.
All four of the holographic "quarters" will have a reference beam that seems to originate from the same point in space so the four will act as one.
Well, what do you think? This could be very beneficial for large 24" by 24" holograms or larger.
I may have to try it. Might try a 12" by 12" with 4 - 7" by 7" exposed areas.
1) Are you intending this for masters, copies or both?
2) As you've show, the center exposure will either be 2x or 4x the rest of the plate. This will cause increasing loss of contrast and overexposure (at least with SH emulsions, not sure what it will do with DCG).
3) For "normal" exposures you don't want the light to come "from the same point" but rather from the same direction. You're going to end up with four different reference angles. If you're wanting to end up with a plate that contains a single image, this won't work properly. You can moderate this somewhat by using a large distance between the final mirror and plate but most of our tables probably aren't large enough.
---------------------------
If someone says it can't be done but they haven't tried it, don't believe them. http://www.dragonseye.com/Holography
1) Whatever it works for. Specifically H2s and in my case, single beam reflections.
2) Being that I spend minimal time on the drawings one needs to use it only as a guide to what I describe in the post. If the beams are gaussian to start with the overlapping area will be of less intensity then the other areas so the ratios will be approximately the same.
3)Not sure what you mean. If I replace the mirror with a spatial filter then the angles are different with respect to the four quarters and it is the point in space of the reference beam that I am talking about. I do not use a collimated reference beam for most of my holograms. Having a transfer mirror allows the origination of the reference beam to be further away.
As always thanks for the feedback. It may take some more thinking though but I feel it would be worth it if I can get it to work.
If the beams are gaussian to start with the overlapping area will be of less intensity then the other areas so the ratios will be approximately the same.
True, but my gut says they'll be different enough to introduce some odd artifacts. Still worth trying.
Not sure what you mean. If I replace the mirror with a spatial filter then the angles are different with respect to the four quarters and it is the point in space of the reference beam that I am talking about.
This method works fine (I think) for the H1 where you'll be illuminating it using the same setup but falls down somewhat if you intend to use it for an H2. Each quarter ends up with a different reference angle and unless you have sufficient distance between the plate and the mirror, it will be noticiable in the final hologram.
Along the lines of a montage, you could use a similar method only with seperate plates. By precisely aligning each plate (in the quarters you originally show) alone, you can put them back together like a puzzle and end up with with a single image. I believe Zebra imaging does something similar for their very large holograms.
---------------------------
If someone says it can't be done but they haven't tried it, don't believe them. http://www.dragonseye.com/Holography
I guess what I am trying to do is take a finished hologram and break it into 4 quarters. Then take each quarter and work backward to duplicate what that part of the plate experiences during exposure.
Reminds me of Adrian Lines' "Portrait of the artist as a dead man" Some of you old-timers might remember him and that work. He made several 8x10's, each one a part of a body, cut the outer ones so it looked like a human form and placed it in a coffin. The effect was of a hologrphic human form in a coffin.I believe he won some major prize for that one.
Sometime early 80's there was a holographic exposition at the Science museum in London. I was an intern at Richmond Studios then and I was helping set up the pieces. Adrian had a few pieces at the show and he had just finished setting them when the museum opened to the public. As Adrian was looking at his hologram, some guy walked in, looked at the hologram, turned to Adrian and said, "That's a 3D hologram. It's 3D" Adrian put on an expression of total amazement and said, "Really? How do they do that?" Then as this guy was explaining to Adrian how "..it was done", basically telling Adrian how he made his own hologram, Adrian looked more and more amazed. Finally, when the guy finished his explanation, Adrian just refused to believe the guy that it was "..3D". Rob Mundy, Paul Newman and I, watching this from a distance, were laughing our heads off!
Saxby 3rd edition p.189-90 describes a scanning arrangement using a cylinder lens to make an intense line of light that is mechanically swept over the plate, resulting in very short exposures (~10 ms) for any particular strip, vastly reducing worries about vibration and settling time. Spatial filtering a line would be easier, too - a slit at the cylinder lens focus ... I wonder if something like this might work for you?
I do not wish to actually have four separate pieces of film/glass that has a noticeable line. I want the finished hologram to be homogeneous as if it were one hologram. Especially if the subject matter is in the center of the plate.
Tom, I read that and it seems very doable but I was actually trying to perceive a new method that I have not read about and possibly generate some new subject matter to discuss. For some reason I envision the group as brain stormers and was hoping for some calculations or suggests on beam profile or some novel way of masking that may generate a novel method for creating large format holograms with low power and short exposures. Seems the thread did not take that path. Oh well!
But I will tell you that now that everyone seems to shy away from the idea, I will have to pursue and perfect it. Now I am all fired up about thinking it though and making it real before the PCG meeting. I would love to bring a finished product there but there is much I need to do to get the lab back in order.
Michael, If someone says it can't be done but they haven't tried it, don't believe them, right?
It seems, and I need to wait until I get home to use a better drawing tool, that the gaussian profile is reduced to a more even illumination. The drawing on the left is four gaussian profiles, throwing away the outer edge on a large black plate four times the size of the red plate. To me it seems there is a more even illumination. I know this is crude and a true gaussian is not as linear as I have drawn but this is what I see in my mind generally. Feel free to interject. All comments welcome.
1. If you're shooting the same object a number of times (4 in your diagram) and the object moves between shots, you'll get interference. Movement between shots is not the same as movement during a shot.
2. You're hoping that the Gaussian fall-off in the middle areas will re-inforce each other, but this needs careful balancing. It could over-reinforce and give you a brighter and/or noisier areas where the images overlap.
3. If this is a single beam denisyuk, the object will be lit from a different perspective for each shot. This discrepancy may not be bad for a small object but, by definition of large scale, the object will probably be large.
3a. Even if the object is small, the highlights of the illumination will vary as you change the reference.
4. Each illumination angle will throw light into the areas of other parts of the hologram. Let's say we number them exposures 1-4 clockwise from top right. Then the illumination of the object from reference 1 will also expose the parts of the plate in regions 2, 3 and 4. Similar for each of the other illuminations.
In the end, I don't think it's impossible, but think it needs careful thinkin out to balance the beams. I also think the choice of object is important, for example a perfect sphere may not work for reasons 3a and 4 above.
