Help with optics, telecentric lens, and focusing at infinity

[jasonpatel]

Hi Guys,

I have a question about a telecentric lens I've been studying and was wondering if you guys could help clear some cobwebs in my head. For starters, I am simplifying the TC lens as a compound lens. Also, this lens has 2 degrees of freedom: (1) it can be extended and retracted (2) it can rotate (I believe rotating the lens changes its f# e.g. aperature).

So when I fully extend the telecentric lens (specifically http://computarganz.com/file.cfm?id=167) it is focused at infinity. I interpret this as it being being able to image an object at a distance of infinity and produce an image focused at the CCD behind the lens.



1. Is this a correct assumption? Will the TC lens focsued at infinity produce a focused image at the CCD? What does this mean for an object not at infinity? For objects close to the TC lens? Are their images still produced at the CCD?



Continuing, if I placed an object at some distance in front of the lens (not equal to infinity), I could create a focused image of the object at the CCD by retracting the lens until I see a focused image at the CCD.

2. How did the focus/image of the object change as the TC lens was retracted, going from focused at infinity to focused at the image location? Is it that the object is focused but the resultant image is not at the CCD until you adjust the lens (by retracting it) until the image is placed on the CCD?


Thanks so much for your help in advanced. This stuff is really bending my brain lol!

 

[Drakkith]

1. Is this a correct assumption? Will the TC lens focsued at infinity produce a focused image at the CCD? What does this mean for an object not at infinity? For objects close to the TC lens? Are their images still produced at the CCD?

"Infinity" means that objects that are 'very' far away, generally greater than a hundred feet or more, will be in focus. This isn't an exact measurement, and it is possible for objects such as stars to be out of focus even when the lens says it is focused at 'infinity'. (I do astrophotography and this is sometimes a problem)

To understand what is going on here, it is important to understand what focus is. Focus is where the incoming light rays are concentrated down to the minimum spot size on the sensor. But because of the way light and lenses work, we can't have everything in focus at all times. If you set your lens to focus on something that is about 20 feet away, anything closer or further away will be slightly to severely out of focus depending on how far or close it is. I can't give you an exact distance because that depends entirely upon the lens design and the F/Ratio of the lens at the time of imaging. Now, even though anything not at 20 feet will be out of focus, that doesn't actually mean it will be blurry. There is a finite tolerance to how blurry it can be before it is actually detrimental to the image. This range is known as the 'Depth of Focus' or 'Depth of Field'. Anything within this range has acceptable amounts of sharpness, even if they aren't perfectly sharp. Objects outside of this range will still be imaged on the CCD, but will be noticeably blurred.

Continuing, if I placed an object at some distance in front of the lens (not equal to infinity), I could create a focused image of the object at the CCD by retracting the lens until I see a focused image at the CCD.

2. How did the focus/image of the object change as the TC lens was retracted, going from focused at infinity to focused at the image location? Is it that the object is focused but the resultant image is not at the CCD until you adjust the lens (by retracting it) until the image is placed on the CCD?

All the light coming into the lens is being concentrated to a point which is a certain distance behind the lens. For objects that are in focus, this distance is equal to how far the CCD is placed behind the lens. IE they come to focus at the CCD sensor. Objects closer than those at focus have their light concentrated further away than the sensor, while objects further away have their light concentrated before they reach the sensor, thus making them blurry.

http://en.wikipedia.org/wiki/Depth_of_field
http://en.wikipedia.org/wiki/Focus_(optics)

 

[jasonpatel]

Awesome, thank you for the great response! I am trying to absorb it right now...

Two more quick questions:

1. So, then what does the "focused at infinity" setting mean on the telecentric lens? If an object infinitely far away will have light rays coming in parallel no matter what the setting on the lens is, why even have a "focused at infinity" setting on the lens?

2. Let's say hypothetically I created a system of N lens. After going through the imaging physics for each lens I find that the Nth (last) lens produces an image at infinity (not at the location of the sensor). Does this mean that the sensor will see no image?

Just to clarify, I am interested in infinity optics because I want to image the angular resolution of an object (electron beam). If you had any simple papers/reads on this I would be extremely grateful! (I only ask because it seems like you have some great experience in the imaging/optics field!)

 

[Drakkith]

1. So, then what does the "focused at infinity" setting mean on the telecentric lens? If an object infinitely far away will have light rays coming in parallel no matter what the setting on the lens is, why even have a "focused at infinity" setting on the lens?

Because if you didn't, the light would be focused incorrectly. Keep in mind that the depth of field for 'infinity' ranges from a few hundred feet to billions of light-years. Focusing the lens closer than this will result in a different depth of field.

2. Let's say hypothetically I created a system of N lens. After going through the imaging physics for each lens I find that the Nth (last) lens produces an image at infinity (not at the location of the sensor). Does this mean that the sensor will see no image?

Unless the light is tightened down to a very small beam that can form an acceptable image, then you will have a big blurry mess. With non-coherent light this is generally not possible, you must focus it onto the sensor.

Just to clarify, I am interested in infinity optics because I want to image the angular resolution of an object (electron beam). If you had any simple papers/reads on this I would be extremely grateful! (I only ask because it seems like you have some great experience in the imaging/optics field!)

What kind of resolution are you trying to get?

 

[jasonpatel]

Well, resolution is not the main issue right now, I guess wrapping my head around the optics is more so.

Basically, I need to find the focal length for an insertable/removable lens (constrained to the right 150mm from the object), inserted between the object and the telecentric lens (the telecentric lens can be moved, so x must be solved for as well) that would:

In one configuration, focus the TC lens at infinity therefore giving the angular distribution of the object. And in the other configuration give a focused image of the object to the CCD. Being that there are only two configurations: insertable lens being inserted or removed from the system at the same location (think of a mechanical lever that will slide the lens in place).

I'm thinking that I should start with the TC lens focused at the object. Then insert the lens and that would somehow focus the system to infinity giving the angular distribution. But, I'm kinda of confused as to where to start; i think I am over complicating it :(

 

[Drakkith]

I think your needs surpass my knowledge! I don't even know what angular distribution is lol.
Hopefully someone else here can help you. Just give it a day or two.

 

[Andy Resnick]

<snip>So when I fully extend the telecentric lens (specifically http://computarganz.com/file.cfm?id=167) it is focused at infinity. I interpret this as it being being able to image an object at a distance of infinity and produce an image focused at the CCD behind the lens.

1. Is this a correct assumption? Will the TC lens focsued at infinity produce a focused image at the CCD? What does this mean for an object not at infinity? For objects close to the TC lens? Are their images still produced at the CCD?

It's not clear if this is indeed a telecentric lens: lenses can be telecentric in object space, image space, or both. 'Telecentric' means the entrance and/or exit pupil is located at infinity, and as a result magnification does not depend on object distance: identical objects placed at varying distances from the lens will be imaged as having the same size, which is why telecentric lenses are often used for part inspection. However, the datasheet shows image sizes varying with object distance.

Typically, telecentric lenses have an *enormous* front element, because the field of view is equal to the size of the front element:

http://www.edmundoptics.com/imaging/imaging-lenses/telecentric-lenses/gold-series-focusable-telecentric-lenses/1630

Your lens appears to be the same as this:

http://www.edmundoptics.com/imaging/imaging-lenses/telecentric-lenses/55mm-fl-partially-telecentric-imaging-lens/1963

Note that it is only telecentric over a restricted object distance and the magnification varies with object distance. (Side note- check out the hypercentric lens!)

In any case, to answer your questions- yes. The lens is in many respects just like any other lens. The lens law (1/do + 1/di = 1/f) still applies.

 

[jasonpatel]

It's not clear if this is indeed a telecentric lens: lenses can be telecentric in object space, image space, or both. 'Telecentric' means the entrance and/or exit pupil is located at infinity, and as a result magnification does not depend on object distance: identical objects placed at varying distances from the lens will be imaged as having the same size, which is why telecentric lenses are often used for part inspection. However, the datasheet shows image sizes varying with object distance.

Typically, telecentric lenses have an *enormous* front element, because the field of view is equal to the size of the front element:

http://www.edmundoptics.com/imaging/imaging-lenses/telecentric-lenses/gold-series-focusable-telecentric-lenses/1630

Your lens appears to be the same as this:

http://www.edmundoptics.com/imaging/imaging-lenses/telecentric-lenses/55mm-fl-partially-telecentric-imaging-lens/1963

Note that it is only telecentric over a restricted object distance and the magnification varies with object distance. (Side note- check out the hypercentric lens!)

In any case, to answer your questions- yes. The lens is in many respects just like any other lens. The lens law (1/do + 1/di = 1/f) still applies.

Hmmm, this is great thanks so much. Its starting to all piece to together now. One quick question:

-If I had a two lens system of focal length f1 (front) and f2 (back), how should/would I position them to focus at infinity (assuming the observer is at a fixed position behind the back lens e.g. mimicking the ccd)??

 

[Andy Resnick]

Hmmm, this is great thanks so much. Its starting to all piece to together now. One quick question:

-If I had a two lens system of focal length f1 (front) and f2 (back), how should/would I position them to focus at infinity (assuming the observer is at a fixed position behind the back lens e.g. mimicking the ccd)??

If you mean a lens with front focal length f1 and back focal length f2, placing the lens a distance f2 from the image plane will locate the object plane at infinity.

If you mean you have 2 lenses, with the first being f1 and the second being f2, there is not a single solution to your problem- many lens spacings are possible (think of it as focusing a zoom lens at infinity while zooming).