Is there a limit in how much I can decenter the PD and height in plano prism lenses in order to reduce thickness? Since the prism value is the same in all parts of the lens, it shouldn't matter, right?
Printable View
Is there a limit in how much I can decenter the PD and height in plano prism lenses in order to reduce thickness? Since the prism value is the same in all parts of the lens, it shouldn't matter, right?
You can move the lens along the base apex line to reduce thickness. Moving up or down will have no effect on the thickness as it the same 90 degrees away from the base apex line.
Why would it need to be decentered for thickness and not just surfaced to the correct thickness to start with?
We are assuming it was surfaced to the correct thickness but when edging you can position the lens so the minimum edge (apex) is not at the edge of the lens. This will give you a finished product that is not as thin as possible. To put it in a number format. If the base of the uncut is 6 mm and the apex is 1 mm and you place the 1 mm edge at the edge of the frame you will have the thinnest result. If you position the lens where the apex is 2 mm you will have a lens that is both thicker on the apex by 1 mm and thicker on the base by the same amount 1 mm.
Quote:
Originally Posted by Lensman11
Yes you could, but why would you? Send a tracing, specify a lens thickness and block accordingly.
Go back to the original question. Can you move the lens to optimize thickness. Very few opticians can calculate the center thickness on a Plano prism. Remember in this case the center thickness is neither the thinnest point or thickest point so how you position the lens will surely effect the end result.
Move it from where? Moving it(decentering) assumes there is a correct place to move it from. Why layout the lens in a suboptimal position then move it to a more optimal position?Quote:
Originally Posted by Lensman11
I guess the premise of the question doesn't make sense to me. Obviously I understand the answer to what the question implies.
Obviously.Quote:
Remember in this case the center thickness is neither the thinnest point or thickest point so how you position the lens will surely effect the end result.
Why would one need to for this?Quote:
Very few opticians can calculate the center thickness on a Plano prism
This is my last reply to this thread. You stated simply order the lens with a stated center thickness send a tracing and block accordingly. I believe that most could not calculate the center thickness on a Plano lens with prism which is not part of the answer to the question. It seems the question was will moving the lens by an extreme amount have an effect on the resulting thickness of the final product. The answer is that it definitely will.
I certainly did not say anything about ordering center thickness.Quote:
Originally Posted by Lensman11
You absolutely can decenter a low power at 180 *spherical* lens for either cutout reasons or thickness considerations. To answer the OP’s question, as Kwill mentioned, if the surfacing lab fabrication was correct, there would be a point that the lens cuts out with minimal thickness, say 1.mm. You can certainly adjust your cutout to minimize the thickness.
But don’t over/under decenter aspheric/atoric for cutout or to induce prism. By deviating the lens OC on these designs, you are moving the optical off axis corrections that these lenses are designed for. Basic rule: only use spherical lenses for over/under decentration.
I hope this helps.
In a plano prism lens, p.d. and height is irrelevant. You could tell the lab anything and it will come out the same.Quote:
Originally Posted by Airegin
Why?
Because: p.d. and vertical height is the desired X-Y location of the optical center. And in a plano prism there is no optical center.
I hope you were never sent out by older lab workers to find a box of "optical centers"...
Well suppose you order a couple plano lenses (not blanks) with 8 prism base out, each with different PD values. I would expect a lower thickness with larger PD values, even though there's technically no optical center. That just seems like common sense to me?Quote:
Originally Posted by drk
I assume this means ordering a job cut and edged by a remote lab, and not uncut lenses, right?Quote:
Originally Posted by Airegin
As Drk stated the PD for this job is irrelevant. The lab should surface and edge the thin edge of the lens to match the frame type. There is no OC, so there is no way to verify any kind of PD. Grab a prism trial lens and put in the your lensometer. Orient the prism horizontally, then move the lens left and right. Nothing will happen.Quote:
Originally Posted by Airegin
Exactly. That's how all orders are done where I work. So even though the PD doesn't matter I'd still expect a different cut depending on the PD I input.Quote:
Originally Posted by Kwill212
Ok. Let's see if we can get you to answer you own question. How would you verify the PD of the finished job you proposed, before dispensing?Quote:
Originally Posted by Airegin
I can't. But I'm pretty sure the lab just uses the center of the lens blank and that physical point will be located at the OC I put in.Quote:
Originally Posted by Kwill212
That is not how a lab makes lenses. If you can't verify the PD the lab can't layout a PD.Quote:
Originally Posted by Airegin
Well it was just a guess. Our software does predetermine the blank size based on the PD and height I order. That would change how the lenses are cut.Quote:
Originally Posted by Kwill212
Quote:
Originally Posted by Airegin
Yes for jobs that are not plano. The minimum blanks size for a plano is equal to the ED+2mm. Your software might not include this exception. This is why we should work to understand the underlying optics and not just rely on our fancy instruments, devices, software, etc. I would recommend contacting your lab for a better understanding of how lenses are made. Ask for an in person tour if possible. Short of that, pick of some books on surfacing and dispensing.
Thank you, I will do just that!Quote:
Originally Posted by Kwill212
Maybe a little late to this discussion,
All Plano lenses have an optical center, this is where the power of the lens is 0.00 diopters, but the lens surfaces are not coplanar. They are concave or convex unless the refractive index of the medium is 1. This optical center can be measured, and its movement can be seen is a simple lensometer albeit exceedingly small measurements at large movements on the XY coordinates. Great LMS systems calculate these small deviations and calculate Plano prism movement in the calcs in 1/100s of a prism diopter increments to make what could be argued as the hardest lens to make. Spherical Plano lens can expose many lab deficiencies, waves, poor diamonds, variation in tool paths, haze, poor blocks, hot alloy and so on.
Adding prism to the lens design adds a multitude of new phenomena. Say CR-39™ would produce a concave lens design whereas a 1.6 index would produce a convex lens design.
Remember we were taught that plus lenses, the image follows the base while the minus lens the image follows the apex. Kind of leads us to a conundrum that a Plano lens follows the rules of a plus lens.
Maybe it is time that we do not use the word Plano anymore as its Latin translation equates to a parallel lens surface design that has no power (cannot happen).
So, calculations should be made to treat this as a plus lens were the thin edge should calculate as ED +2mm. And finish decentration should be done with respect of z80 edge thickness guidelines.
Special notes: higher powered prisms will measure power variations at differ lens thickness due to thickness variations. I often wondered if the digital lens designers would consider this with an atoric design. Prism lenses made with hi-index materials can read a false cylinder in a focomerter. This will manifest itself as a Sphero cylinder that is locked at 180 degrees regardless of physical rotation. This is easily seen by taking a stock hi-index spherical lens and moving to 4+ BO prism. While maintaining prism and direction, rotate the lens and note this phenomenon. This is caused by light breaking down into its spectral components (ROYGBIV) and focusing at differ lengths. The little switch on the back of a lensometer the turns the reticle green is meant to minimize this effect (called a red/green filter).
Ask any surface lab person about trying to make the following.
+6.00- 0.25 X 30 hi-index progressive les design. The power measuring circle is 8mm from the prism reference point so (power* decentration /10) Thanks Mr. Prentice, give us 3.2 Diopters of prism that floats the .25 cylinder all over the place. As an industry I bet we throw thousand of lenses into the garbage due to this effect. I sure would like to hear some Lab Rats sound in on this.
Newer lens mappers and auto lensometers eliminate or have setting to minimize. A want list item for every lab and office should be a measuring device that can interpret digital lens design and eliminate high index function errors.
Chris
This was a great reply a lot of time and thought went into your response. I would like to add a few comments and a few questions.
I would consider myself a lab rat being I have been running a lab for 40 years ( currently retired) and was involved in the development of one of the first lab computer systems and we were credited with making the first sucessfullink to generator eliminating operator input.
It is kind of impractical to worry about 1/100 diop of prism you can measure it nor can you be that accurate in manufacturing. If you add prism at the generator instead of blocking in it you will never end up with the desired outcome. The error induced-in fining is large and mostly not considered in the manufacturing process. The exception is free form as no fining is required.
Please explain further what a cr39 lens is minus and 1.6 is plus they both have front and back surfaces parallel
Why would you add 2 mm to the edge calculate thickness that would make plus lenses thicker than necessary. To optimize thickness you 0.5 mm at the bevel apex and 2 mm at the bevel base which is 4 mm in from the apex.
Higher prisms will read odd cyl powers and axis somewhat because of the thickness but mostly because the lens is not perpendicular to the lens stop. If hold the lens manually keeping flush against the stop and a lensometer with a prism compensator lensometer you will see much better results. The lens holder was to hold a lens with equal thickness on each side not the case with a purism or decentered high power lens. This might be difficult wth the temples hitting the side of the lensometer.
If you a lensometer with a prism compensator you will get more accurate results. A lens mapper does not have this problem because the lens is not tilted.
Hello Lensman11,
Sounds like you and I were cut from the same bolt of cloth. I’m also a retired optical vet having started back in the 70’s. Admittedly, my mind isn’t a sharp it once was. Let me try some better answers to your questions…
- “Please explain further what a cr39 lens is minus and 1.6 is plus they both have front and back surfaces parallel” I kind of misspoke here, I’m mixing tool index with lens index. This is another huge mistake we made in our industry. Say a lap tool marked 6.00/6.00 diopter is wrong, as the steel, iron, aluminum can’t bend light. Again, the definition of a 1.00 diopter is the deviation of light 1cm at 1m. Metal wont bend light. Most lens calculation are done with radius of curvature and then convert, wrongly, to Lap diopter power. Let’s try this... If we make a lens with 2 truly parallel surfaces this lens will result in power due to the index of refraction. The greater the index and thickness the more it needs to be offset with a steepened minus curve. This can also be observed when viewing the unrounded calculations as well as observed in a lensometer. As you move the lens around you will see OC movement.
- “Why would you add 2 mm to the edge calculate thickness that would make plus lenses thicker than necessary.” At some point we need to define what our wanted edge thickness is going to be. Initial calculations run to a knife edge then add wanted edge to recalculate the rear tool curve for said thickness (this is really an infinity loop calculation you need to stop at some diminishing point of error reduction). Ideally you calculate wanted edge thickness as a function power and edge type with regards to z80 standards. Take a +6.00 diopter drill mount lens, a 2mm strap would look like a brick. The drill holes would be insanely deep resulting in poor mounting stability. The LMS system I’m most familiar with used multiple edge types in combination with different material to respect the tensile strength as well as optimizing cosmetics. Again, I oversimplified my original response.
- “Higher prisms will read odd cyl powers and axis somewhat because of the thickness but mostly because the lens is not perpendicular to the lens stop. If hold the lens manually keeping flush against the stop and a lensometer with a prism compensator lensometer you will see much better results.” This is true! I just had a thought, what if we measured from the convex side and not use the lens stop? I need to give this some thought. What I was trying to say is that high index prisms and lenses of power break down light into its spectral components and a lensometer can’t resolve. The mires will be out of focus as movement increase from the OC.
Non-related but might be great for discussion is the digital claims of reduced power error due to 1/100 tooling. The front curve deviation on a progressive lens is .18 diopter this can instantly place a lens “out of tolerance”. Going back to 1.50 tooling in 1/8-degree increments, say on a flat top the rounding direction with regard to the front surface error could result in a 24% rejection rate. Does anybody else’s hair stand up when they hear 1/12 diopter?
I had some friends that used to describe these lens making nuances as “magic”. And the were/are amazingly talented. They know who they are.
Best to all and I’ll try to be a better communicator,