Simple setup: A Red LED in front of a small fresnel loupe/lens (credit card size). Using a screw I can adjust the distance between the LED and the lens. The LED is centered with regard to the lens.

My question: I pointed my assembly at a wall 1 meter away and adjusted the distance so that the resulting spot on the wall has the smallest diameter (increasing and decreasing the distance at this point results in the spot to grow and get blurry). Then I aimed at a different wall which is about 4 meters away and the spot got bigger - as expected. However, I was surprised to find that, with the different distance to the wall, adjusting the distance between LED and lens leads to a more focused spot.

What is the reason for this? Why and how does the distance between LED/lens and the wall I am pointing at matter?

A picture for the 3 cases: 1m (diameter 3cm, focus), then turned at 4m (12cm, out of focus), then distance adjusted (10cm, again focus)

Focus points at various distances

  • $\begingroup$ Hello, and welcome to Stack Exchange. What are the two spot sizes? And, pictures of the two could really help. $\endgroup$ Commented Sep 16, 2017 at 12:40
  • $\begingroup$ @DanielGriscom I uploaded the pictures and measured the diameters. $\endgroup$
    – zelyev
    Commented Sep 16, 2017 at 13:11
  • $\begingroup$ My original read was that you were surprised to get a smaller spot at 4m than at 1m. Now, I'm thinking you were surprised that you had to change the LED-lens distance with a changed lens-wall distance. Which is it? (The latter is due to simple optics.) $\endgroup$ Commented Sep 16, 2017 at 13:18
  • $\begingroup$ Yes, I am surprised that I have to change the lens-LED distance with a changed lens-wall distance - would really appreciate clarification on that. Doesn't focus mean that you send out the narrowest beam possible? $\endgroup$
    – zelyev
    Commented Sep 16, 2017 at 13:22
  • $\begingroup$ Is this just an application of the lens formula, $\frac 1 u + \frac 1 v=\frac 1f$ with $f$ the focal length of the lens, in that as you change the image distance $v$ you then have to change the object distance $u$ to get the image in focus? $\endgroup$
    – Farcher
    Commented Sep 16, 2017 at 15:06

1 Answer 1


A picture of a simulation that helped me to better understand what @DanielGriscom could have meant by "simple optics" in the comments.

I used https://ricktu288.github.io/ray-optics/simulator/ to simulate a point source and a moving lense. One can clearly see how the point where the rays get focused moves when the distance changes. The distance between LED and lense has to be adjusted according to the distance between wall and lense. The properties of the lense are the determining factor for how the system behaves.

You can save the following with your favourite editor as a .json file. Then you can open it with the simulator, select the different parts, and rearrange them.


enter image description here

  • $\begingroup$ Wow; that's a cool looking tool, but in five minutes of hacking I wasn't able to get anything like your extremely apropos display. Would it be better to give a higher-level link such as ricktu288.github.io/ray-optics? Or, is there any way to save your configuration such that other visitors can directly open it? $\endgroup$ Commented Sep 16, 2017 at 16:24
  • $\begingroup$ The first beam shown here is what we call a "collimated" beam. The radius of the beam does not change with distance from the lens. All of the other beams are converging, or focusing at some distance from the lens. As you've noticed the distance of the focus depends on the distance between the lens and the object. If you move the lens closer to the object you will get beams which are diverging, that is they are spreading out and never focus anywhere. If you change the focal length of the lens all of the the distances will change. In addition, the size of the collimated beam will change. $\endgroup$
    – Jagerber48
    Commented Sep 16, 2017 at 16:42
  • $\begingroup$ @DanielGriscom I added the .json data that can be imported with the simulator. $\endgroup$
    – zelyev
    Commented Sep 17, 2017 at 16:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.