Need to Refocus Projected Image Through Microscope Objective

[bstafford]

Greetings-

As will become abundantly clear from this post, I know very little about optics, so any and all help will be appreciated.

I am a visual neuroscientist and I am attempting to project a visual stimulus through a camera port on a microscope, off a mirror, through the microscope objective, and down to the sample plane. I am facing two issues (that I am aware of so far).

1) The visual stimulus is being generated by a miniature DLP projector. The minimum focal distance of the projector is 23.6" at which distance the projected image is a 15" diagonal rectangle. The camera port into which I want to project the light stimulus is 1" in diameter. So, I need to add an additional lens close to the output of the projector to re-focus the image so that it can fit through the camera port. When I do this, however, the image from the projector will be out of focus since the lens will not be at the focal distance of projector. The closest I could get a new lens would be flush with the edge of the projector. This is ~1" from the projection lens and the diameter of the opening at that point is 1.25". The projection lens is listed as having an F/2.0; f=17.67 mm fixed lens.

2) Inside the microscope there is a lens element that has a 229.01 mm focal length. As far as I can tell I need the out-focus-image that I have captured from my projector to somehow end up focused (or relatively close to focused) when it reaches this lens.

If anyone can provide any input regarding types of lens to use and where to position them in this set up I would really appreciate it. Even input like: "You have not provided enough information to figure this out, please provide the following additional information:" would help.

Thanks in advance for any input/help.

 

[voko]

Does the camera port incorporate any optical elements? Is it basically a tube with the objective piece at the other end? (Any flat mirrors are irrelevant for the purpose of the previous question.) How are you going to attach the projector to the camera port?

Oh, and you could give us links to the specs of the microscope/objective/projector.

 

[Andy Resnick]

A few questions/comments:

<snip> I am attempting to project a visual stimulus through a camera port on a microscope, off a mirror, through the microscope objective, and down to the sample plane.

Let me make sure I understand- you are imaging (projecting an image of) a DLP onto a sample plane, de-magnifying the image with a microscope objective, yes?

1) The visual stimulus is being generated by a miniature DLP projector. The minimum focal distance of the projector is 23.6" at which distance the projected image is a 15" diagonal rectangle. The camera port into which I want to project the light stimulus is 1" in diameter. <snip>

Yikes- let's simplify this. Sometimes it helps to think in reverse- you have a object on the sample plane that you want to project onto the DLP. One immediate problem comes to mind: the DLP projector lens *magnifies* the DLP, which is the opposite of what you are trying to do. So, if you can't remove the projector lens, you need a 'demagnifier' lens (a reversed telescope) that can take a (say) 15" object and reduce it to about 1" (the size of the microscope optics), which is going to require a ginormous front element (at least 15" in diameter). Alternatively, you can only project a small central portion of the DLP onto the sample plane. A *far* better option is to remove the projector lens and replace it with a reasonably-sized collimator lens to directly couple the DLP to the optical path of the microscope.

2) Inside the microscope there is a lens element that has a 229.01 mm focal length. As far as I can tell I need the out-focus-image that I have captured from my projector to somehow end up focused (or relatively close to focused) when it reaches this lens.
<snip>

Ok, the tube lens- that's a fairly precise specification- I assume you did not measure that. In some ways, if you can use the tube lens then you may not need the collimator lens either- placing the DLP 230mm away from the tube lens will result in a well-corrected image of the DLP (less any vignetting) projected onto the sample plane. In other words (thinking in reverse), put the DLP where the camera sensor usually goes.

Regardless, you must eiher figure out how to remove the projector lens or resign yourself to only using a small area of the DLP.

 

[bstafford]

Does the camera port incorporate any optical elements? Is it basically a tube with the objective piece at the other end? (Any flat mirrors are irrelevant for the purpose of the previous question.) How are you going to attach the projector to the camera port?

Oh, and you could give us links to the specs of the microscope/objective/projector.

Thanks for the response!

To answer your second question first, here are the best available specs for the monitor (Dell M109s):

http://www.dell.com/support/my-support/us/en/04/product-support/product/dell-m109s/manuals
http://www.dell.com/us/business/p/dell-m109s/pd

The microscope in question is a Zeiss Axio Examiner D1 and the objective I will be using is a 10X EC Epiplan. Limited specs for the objective can be found here:

https://www.micro-shop.zeiss.com/?a=v&f=o&id=442030-9902-000&l=en&m=a&p=us&ss=1

To answer your first questions, there are two lenses between the camera port and the objective. One is at the end of a tube to which the camera port is attached (lens 1). The second is a tube lens within the body of the microscope (lens 2). I have now been waiting for over two weeks for technical specs regarding these optical elements in the light path. I may never get them because they consider this proprietary information. I have received technical drawings of the light path between the camera port and lens 1. This lens typically helps project the image in the sample plane up to the camera. Based on the drawings, the distance from this lens to the camera is 229.01 mm. Thus, I assume that the focal length for this lens is 229.01 mm. I have a campus machine shop that will build me a shelf of my liking that extends from the camera port that will allow me to align the projector with the center of camera port.

Again, my hope it to insert an additional small lens between the projector and the first lens in the light path (i.e. lens 1 from above; the one whose focal length I think is 229.01 mm). I do not have enough room in my set-up to position the projector at its minimal focal length (23.6"). Instead, I am hoping I can place this lens close enough to the projector such that the image it projects will not be vignetted at all. When I do this, of course, the projected image will not be in focus. I am hoping that the combination of a lens with specific characteristics and precise positioning of the projector and lens that the projected image can somehow be refocused onto lens 1.

I hope these answers help.

 

[bstafford]

Thanks so much for the reply!

A few questions/comments:



Let me make sure I understand- you are imaging (projecting an image of) a DLP onto a sample plane, de-magnifying the image with a microscope objective, yes?

Currently, this is how things are set up. However, there are two other lens elements in the light path, inside the microscope, for which I have limited-to-no information. One is at the end of a tube to which the camera port is attached (lens 1). The second is a tube lens within the body of the microscope (lens 2). I am operating under the assumption that if I can get my DLP image in focus at lens 1, that lens and the tube lens will preserve the image from there through the objective and down to the sample plane. Does this make sense?

Yikes- let's simplify this. Sometimes it helps to think in reverse- you have a object on the sample plane that you want to project onto the DLP. One immediate problem comes to mind: the DLP projector lens *magnifies* the DLP, which is the opposite of what you are trying to do. So, if you can't remove the projector lens, you need a 'demagnifier' lens (a reversed telescope) that can take a (say) 15" object and reduce it to about 1" (the size of the microscope optics), which is going to require a ginormous front element (at least 15" in diameter). Alternatively, you can only project a small central portion of the DLP onto the sample plane. A *far* better option is to remove the projector lens and replace it with a reasonably-sized collimator lens to directly couple the DLP to the optical path of the microscope.

Unfortunately, I cannot remove the projector lens. My hope it to insert an additional small lens between the projector and the first lens in the light path (i.e. lens 1 from above; the one whose focal length I think is 229.01 mm). I do not have enough room in my set-up to position the projector at its minimal focal length (23.6") away from the camera port. Further, I don't have room to place a huge 15"+ lens element there either. Instead, I am hoping I can place this small lens close enough to the projector such that the image it projects will not be vignetted at all (or at least not very much). When I do this, of course, the projected image will not be in focus. I am hoping that the combination of a lens with specific characteristics as well as precise positioning of the projector and the lens, that the projected image can somehow be refocused onto lens 1 inside the microscope.

Ok, the tube lens- that's a fairly precise specification- I assume you did not measure that. In some ways, if you can use the tube lens then you may not need the collimator lens either- placing the DLP 230mm away from the tube lens will result in a well-corrected image of the DLP (less any vignetting) projected onto the sample plane. In other words (thinking in reverse), put the DLP where the camera sensor usually goes.

Yes, the 229.01 mm measurement came from the one technical document I have managed to pry from my microscope manufacturer. In an ideal world, what you describe is essentially what I would do. Project the DLP image onto a huge ~15" lens 23.6" from the projector. The huge lens element would then de-magnify that image down to 1" in size to a point that is 230 mm from the lens at the end of the camera port tube. Instead, I am hoping that if I capture the out-of-focus, but un-vignetted DLP image, and project it, perhaps, to a specific point in front of lens 1, it will end up as a well-corrected image of the DLP at the sample plane. Of course, as I mentioned in my first post, I know very little about optics, so perhaps this sounds insane.

Regardless, you must eiher figure out how to remove the projector lens or resign yourself to only using a small area of the DLP.

This is sort of what I am afraid of. However, I have colleagues who are doing this successfully, although they are projecting their DLP image through the microscope condenser (not an option for me unfortunately). Their light path is quite different from mine and, presumably, they can get more information about the optical properties of the condenser. Thus, I am hoping I can somehow get this to work.