No announcement yet.

fresnel lens

  • Filter
  • Time
  • Show
Clear All
new posts

  • fresnel lens

    I'm going to try my hand at turning a fresnel lens from acrylic. I'll probably start with 3mm material since that's what I have right now. I have a glass lens of about the right curvature so I'm going to use that as the starting point to lay out the angles on the rings. This will be a pretty coarse fresnel, but it's not intended to be of optical quality, just good enough to focus an led emitter to about a 15 degree beam.

    I've seen how a diamond cutter can produce a near-perfect finish right off the bat, but I'll have to accept the best finish I can muster using a well-prepared carbide cutter. I'll probably try flame polishing, and I'll try the wipe-on CA trick also. I could also try using epoxy finishing resin to coat the surfaces.

    Anybody have any experience doing something like this? I remember decades ago my dad making some plexiglass lenses on the wood lathe, but I don't remember how he polished them up to a useful condition- and they weren't fresnel lenses which would be problematic to polish up I'm sure.
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  • #2
    WD40 seems to work well for keeping clear surfaces when machining acrylic.


    • #3
      Instead of carbide tooling you might get better results with really good carbon steel tool sharpened and polished like no other. Take your second-best wood chisel or plane iron..


      • #4
        Evan had a post about chemically polishing plastics. I haven't tried it, but it looked like it'd work.

        I've used Novus plastic polish to restore clear plastic after CA debonder hazed it. I think a kit of small bottles (three different "grits" of polish) is about $10 on Amazon.


        • #5
          The chemical treatment is probably vapor polishing. Usually methylene chloride is used, but sometimes acetone. The part is exposed to concentrated vapor which condenses on the surface, dissolving and flowing irregularities like scratches.
          For just a little more, you can do it yourself!


          • #6
            I don't think you want acrylic for machining...Lexan is a better choice. IMHO



            • #7
              I use to build fixtures for the semiconductor industry and we could get a clear smooth finish on table saw cuts and other shop saws with a jointer to smooth the edge then flame polish with a propane torch. I realize this was edge work but if you get the tool real sharp and proper feed rates I would think the flame polish would do the trick.
              Also, lexan has coatings on the material if I remember correctly so you may want just plain acrylic.

              Mr fixit for the family


              • #8
                You might make a polished aluminum form and hot press the acrylic into it.


                • #9
                  I believe that for a simple collimating lens of this sort that you just need a basic circular arc as your reference for the angles of the steps. Setting those angles accurately is going to be the real challenge. Also if you have not considered it yet I'd suggest the "plano-convex" lens as your inspiration since it has one flat side and one curved. It will be far easier to cut just on one side than on both.

                  And again if you don't remember this as with any plastic or brass you want a slight negative top rake. Or at MOST a neutral top rake.


                  • #10
                    To get angles attach a 6ft bar in line with the topslide, mark a point in line with the end of the bed, calculate chord lengths for the angles to measure between the reference point and the end of the bar.


                    • #11
                      This sounds like a real interesting project.

                      The chemical polishing may be as simple as suspending the lens over a shallow dish with the appropriate solvent in it and letting the vapor do the work. You could experiment with some small test pieces. I would look up Evan's post on that, he usually had some good ideas.

                      BC, fresnel lenses are usually "plano-convex". I have never seen any that weren't. One side flat and the other side has the rings. I used quotation marks because on an overall level both sides are flat. You would not want to put those rings on both sides because the light travels at an angle after hitting the first ring and could easily strike the opposite face at the junction between two rings. That would result in scattering and a great loss of light. One flat side is the only way to go.
                      Paul A.

                      Make it fit.
                      You can't win and there is a penalty for trying!


                      • #12
                        To start with, I'm not exactly sure what plastic I'm using- it was something I had on hand and probably was (is) lexan. I know that plastic and glass have differing refractive indexes, so there will be some experimentation involved- I'm ok with that. The glass lens I'm modelling this after looks to be parabolic, so that will throw another curve into it.

                        I may be better off with HSS instead of carbide- I was just hoping to be able to get a really good sharpen on a piece of carbide, then not have to touch it for the duration of the cutting. One of the more interesting parts of this experiment will be setting up a fixture for the cutter so it can be made to describe a proper arc. Seems that a ball turning attachment with an angle readout would do the job- I'll have to correlate the angles with the cross slide feed amount.

                        Because the focal length will be very short, in the area of 10 to 15 mm, the outer rings will end up with a very steep angle. The cutter will have to be very pointed in order to not interfere with the vertical wall on the inside of each ring. In fact I'll have to cut with both sides of it- one to define each wall, and the other to define the curve for the next inward facet. With this in mind, I'll probably have to use HSS so I can get the right cutter shape without knocking the very tip off. Should be fun.
                        I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-


                        • #13
                          A short focal length may imply a small lens. In hat case, how big do you intend to make the "segments"? If they ae pretty small, the difference between a straight slope and a curve may not be much, and you may get good enough focus without requiring curvature. If you have only a few, curvature may be more of a necessity.

                          Keep eye on ball.
                          Hashim Khan


                          • #14
                            I was wondering about that. It would be interesting to place an led module inside the fixture that I use for turning the parts. I could power it through the spindle- seven AAAs in a tube centered in the spindle bore- enough to give light without needing heat sinking- and get an instant indication of the effect of the angle being turned. A quick wipe with CA to clarify the facet and you'd see the result. I'd have to block off the rest of the lens to see how any one ring bends and gathers the light, and see how a straight slope behaves. I can see it being just fine.

                            It seems to me that the fewer rings I need to make, the more light is going to be gathered and focused by the lens. The more rings, the more surface area is going to be taken up by anomalies created by the tip of the cutting tool, and by the coating material if that's how I clarify the lens surfaces. A flame is also going to create anomalies in the sharpness or crispness of the corners.

                            Obviously there are a number of factors that can be juggled. With many short facets, a straight slope is going to be fine, and with few facets I will probably need to cut a curve. The lens in this case is going to be about 1.35 inches diameter, and can be as close to the led module as 10mm.
                            Last edited by darryl; 11-07-2017, 03:37 AM.
                            I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-