Can underground pipes affect holographic interferometry results?

[FissionMan1]

We are attempting here at Illinois Wesleyan to create a holographic speckle-pattern inteferometer. We have been working at it for some time and have come to a roadblock in terms of picture quality. We're using a high quality optics table so there are no table vibrations. All of the mounts are tightly and securely mounted. We are using an image multiply system and the standard image-subtraction with Labview. We have an image, but we're looking for other adjustments or suggestions to try to improve our picture. If anyone else is doing similar research we'd extremely appriaciate recommendations for better picture quality.

Follow the link below for the current quality of our efforts. In this image we are driving a metal plate at a resonant frequency of 1542 Hz. This is the best image quality we have right now and to show what we are supposed to see, I have outlined the pattern in red.

http://www.totalfission.net/pic.jpg

 

[Danger]

Welcome to PF, Fissionman.
I got to admit that I've never heard of one of these things before, but I have a question for you anyhow.
Is your resolution problem related to the holography equipment itself, or to the image sensor?

 

[berkeman]

Interesting background reading on speckle pattern interferometry:

http://en.wikipedia.org/wiki/Speckle_imaging

FissionMan1, how does the holographic aspect factor in? Can you provide a tutorial web link or two to the kind of setup that you're using?

 

[Danger]

Hi, Berkeman. Yeah, it's interesting... what little of it I can understand. This was the site that I checked up on just to know enough to ask my first question.
http://www-staff.lboro.ac.uk/~mcjmh/SIR/speckle.html"
Do I even want to know what 'temporal phase unwrapping' is?

 

[berkeman]

Hi, Berkeman. Yeah, it's interesting... what little of it I can understand. This was the site that I checked up on just to know enough to ask my first question.
http://www-staff.lboro.ac.uk/~mcjmh/SIR/speckle.html"
Do I even want to know what 'temporal phase unwrapping' is?

Cool link, thanks Danger. Now I see what the basic setup is, and where the holographic interference part is used. I'm going to ping an optics specialist friend of mine at HP Labs to see if he has any input.

 

[Danger]

That could be amusing.
Until I read that line about 'under 100 nanometres' resolution, I was about to suggest switching to a shorter wavelength laser. Apparently, though, that wouldn't make much difference. (And I still don't know whether it's the beam or the camera that's causing the trouble.)

 

[berkeman]

Follow the link below for the current quality of our efforts. In this image we are driving a metal plate at a resonant frequency of 1542 Hz. This is the best image quality we have right now and to show what we are supposed to see, I have outlined the pattern in red.

http://www.totalfission.net/pic.jpg

FissionMan1, what does the target look like in a real photograph? From the interferometric aspect of your question and setup, I'd guess that the pattern you've outlined in red is the topographic step of the target? What other target patterns have you tried?

 

[Danger]

What it reminds me of most is the sort of thing you see when powder is put on a vibrating plate to show where the nodes are.

Oh, well... 6:00. Time to head home for a beer. Back shortly.

 

[NoTime]

IIRC the entire image can be reconstructed out of any segment of the mixed reference beam/Image beam.
As a practical matter, both the image sensor area and sensor pixel count would contribute to the resolution.

Also, I vaguely remember from somewhere that the motion of the plate is instantaneously chaotic. The motion averages out to the expected pattern over a period of time. To get the expected image you would need to sum the images from several cycles.

 

[FissionMan1]

Berkeman, you're the greatest. We're so confused as of right now as of where to go.

As for NoTime, the problem with that is that our setup is too unstable to average the images over several frames. We're on an air table and have made sure that all the instruments are locked down tight and not on stands that are too tall, but without vibrating the object we still see some vibration lines. This impacts the image when we are vibrating because there are dark lines that move over the image on the order of 2 or 3 hz. I am unable to figure out why this is. We also have not yet accounted for speckle size vs. pixel size (read below if you don't know what I mean) which could compound our problems, but the main thing seems to be stabilizing the image.

For all of you wanting to know what this stuff is, check out Thomas Moore's paper in AJP on building a simple speckle pattern interferometer. That is the paper we are working off of for my research on vibrating instruments. Although at the moment, we're using a circular plate in order to calibrate the stuff.

 

[FissionMan1]

Also, berkeman, the target is simply a flat circular plate which I outlined in red. The red curves are the boundaries of the 6-pointed pattern which we see at this frequency resonance. Basically, at every resonance of the object, it produces holographic patterns that we record (this in time will be used for some other applications we're working on). At different resonances we achieve different types of patterns. If you are interested, read the AJP article I referenced previously.

 

[berkeman]

Can you maybe use a Pockels cell instead of the vibrating plate? Does the Pockels cell produce varying phase delay without any moving parts?

 

[FissionMan1]

http://scitation.aip.org/journals/doc/AJPIAS-ft/vol_72/iss_11/1380_1.html

Sorry, here is the article in full.

 

[FissionMan1]

From what I am seeing, Pockel cells are the vibrating object that we would put in front of the laser. Our goal is to be able to study many different kinds of things, not just pockel cells. But I really don't know what one of those are anyways, lol.

 

[Danger]

This impacts the image when we are vibrating because there are dark lines that move over the image on the order of 2 or 3 hz. I am unable to figure out why this is.

Could it be a strobe effect caused by the interaction of the vibrational frequency with the frame rate of your image capture?

 

[FissionMan1]

If it were a strobe effect it would affect the entire picture each time it happens, but unfortunately this is not the case. They are vertical lines that scroll across the screen.

 

[berkeman]

If it were a beating effect, the lines would be horizontal, I would think. Unless the raster scan of your video system is rotated 90 degrees for some reason.

 

[FissionMan1]

We do get horizontal lines as well, but that's when we are not purposefully vibrating the object.

 

[Claude Bile]

Have you tried imaging the beams using a different camera, or even just onto a sheet of paper and viewing them by eye? That should give a fairly strong indication as to whether the problem lies with the camera, or the intermediate optics.

Also, how stable is the optimised setup? Is the image quality liable to plummet if someone very slightly misaligns one of the optics?

Finally, Pockel's cells are crystals that have a refractive index that varies with an applied E-field. One can therefore modulate the refractive index of such a crystal (in time), by modulating the applied voltage.

Claude.

 

[Danger]

They are vertical lines that scroll across the screen.

Drat! I'll keep paying attention to this, but I'm afraid that my limited (ie: nonexistent) exposure to this subject is a tad disadvantageous.

 

[NoTime]

This impacts the image when we are vibrating because there are dark lines that move over the image on the order of 2 or 3 hz. I am unable to figure out why this is.

Could be acoustic energy coupled directly to the platform.
Perhaps building ventilation systems.
You could try measuring room vibrations with some audio equipment.
Covering the unit with some sound absorbing material might help.

 

[Danger]

Could be acoustic energy coupled directly to the platform.
Perhaps building ventilation systems.

This might be a good time for Russ to get involved, since it's totally his ballywick. Makes sense to me, though. 3Hz is below audible, and should be consistent with a large (say 20"ø) vent conduit resonance frequency. By the same token, should we even ask if this apparatus is anywhere near a railway track or airport? Even if it's isolated from shock through the supports, those things can also set up inaudible sonic waves in the air.

 

[NoTime]

Highways also.

 

[Danger]

Too right. Those 18-wheelers can kick up quite a ruckus. Come to think of it, even a sideroad gets pretty aggressive seismically when being graded.

 

[Claude Bile]

Some vibration dampeners have a resonance at around 2-3 Hz, maybe the OP should consider checking the man. specs of the optical bench.

Also, being in the 2nd or 3rd floor of a building doesn't help vibration-wise - from experience.

Claude.

 

[Danger]

Some vibration dampeners have a resonance at around 2-3 Hz, maybe the OP should consider checking the man. specs of the optical bench.

Would secondary isolation between the bench and the floor help any?

 

[Claude Bile]

The standard procedure is to put the legs in beds of sand.

Claude.

 

[Danger]

Have they checked the leg beds for sand crabs?

Sorry, but I'm just about at my wits' end here.
FissionMan, would you be able to supply some photos of your specific set-up, as opposed to the 'generic' diagram that my link showed? Close-ups of the various bits would be helpful. I'm just considering the possibility that there might be some visual clues that have escaped your notice.

edit: I just thought of something else. Is this apparatus in the vicinity of a large window? If so, outside noises or even just the wind could set up accoustic waves at the vibrational frequency of the pane. Hanging a heavy curtain of some sort would help in that case.

 

[FissionMan1]

Sorry for the wait boys and girls, I went home and had a bit of trouble setting up the dial-up. Anyways, as soon as I get back to school, I will get you some pictures. Unfortunately, my lab partner is probably not around since she goes home on the weekends and I don't believe my advisor is in the lab this weekend either, but rest assured that I will give you guys some pictures.

As for current stability. We're in the basement of the building on the concrete slab that consists the base of the building. Outside there is a lightly frequented road and there are no airports, train-tracks, or alien space vehicles outside. There are no air ducts, there are no radios playing, there are no windows, and we even use a curtain when we do our experiments to block out all the light from the rest of the lab!

 

[Danger]

Well, the explanation is obvious, then. You're haunted!

 

[Claude Bile]

Stethoscopes are nice when trying to hunt down rogue sources of vibration too. From the sounds of it though, the building should be fairly well insulated from environmental noise.

Claude.

 

[Danger]

Something else just crossed my mind. What about underground pipes? Those can carry vibrations for miles and transfer them directly into the foundation.