Planning an aspheric scratch build

Yeah, punctuation and grammar helps a lot, especially when there’s so much text.

QUOTE
As I said before, these multi-optic systems give more lumens but also increases the size of the projection.
A single optical stage would not cause that circle of spill light, and would create a smaller projection while collecting less lumens.
ENDQUOTE

i might have to throw a flag on the play.

size of the emitter projected? is purely a function of two things… actual emitter size, and the focal length of the lens.

it doesnt matter if you use one or 9 lenses? the overall focal length of the single or compound lens, will determine the emitter projected size… now, THAT STATED? i do know what you meant by that statement…

QUOTE
it is caused by the large amount of optical stages, and refraction between the stages.
ENDQUOTE

eh?

When you have multiple optic stages vs single stage, (assuming the final lens is at the same distance) that means the other optical stages are closer to the LED.
Since they are closer, the focal length is shorter, which means the delta angle is bigger and results in a larger die projection.

Of course if you have the closest lens at the same distance at the single lens then the focal distance is the same and the size would be the same, but then there isn’t really a point to the multi-lens system because it would take up more space without increasing the light captured.
(Assuming same diameter)

PS- there is a quote button at the bottom right

The main point of multi-stage optics is to have the system collect more of the light from the LED without increasing the lens diameter to gigantic sizes.
It’s easier to move a lens closer to the LED than to make a giant diameter lens in order to capture more degrees of light.

It’s easier to move a lens closer to the LED than to make a giant diameter lens in order to capture more degrees of light.

oh, no! we have violated Coghlan’s law of absolute throw!

we made something better, without increasing diameter!

snicker

now… mind you, i am impressed you know whats going on with dual lenses. thats a rarity in the flashlight world.

do you know how to calculate the focal length of a dual lens setup? optically,the 2 lenses being used are thought to be combined into one compound lens, with one overall focal length, and are treated as a single lens…

But the point of doing that is not throw, it is increasing optical efficiency by capturing more lumens…
It has more lumens, but projects them over a larger area, therefore making a bigger spot with more lumens, and not more throw.

With a multi lens setup you use the focal distance of the closest lens, every optical stage after that continues to increase the divergence.
You can use any ray tracer to see how adding stages affects the divergence.
Often with multi-stage optics you need a bunch of custom made precision pieces and there’s a ton of variables, I much prefer working with single stage optics to not have to deal with error/imperfections getting multiplied at each stage, and lower production cost.

ho… not so fast… slow down.

your right, but, you are making an assumption that doesnt necessarily need be true.

1) lets use as an example a single lens setup, one lens with focal length of 100mm
2) a lot of light as we know, wont hit the lens, its lost to the sides before it hits the lens 100mm away
3) what to do? well, we will use two lenses. one of them, the one with shorter focal length? will be closer to the lens, and will collect more light. it will focus it a little,keeping more light now to hit the bigger lens.
4) it is true, that if the same 100mm front lens is kept, the new compound lens? will indeed have less than 100mm combined focal length, and so the emitter projected will be bigger because of the lower focal length of this new compound lens. and yes, it will be brighter because that first lens collected light that would have been otherwise lost, so, efficiency is up and the bigger projected emitter is now brighter.
5) the assumption you are guilty of making? who said we will KEEP the original front 100mm lens?

to wit…

6) we will replace the single 100mm FL big lens? with 2 lenses, true… but two different lenses! we WANT the projected emitter size of the 100mm FL single lens? so, we obviously WANTED that size of emitter projection, right?
7)why couldnt we simply pick 2 lenses… such that theres a smaller lens with shorter FL near the emitter, giving us the efficiency we want…
8) we simply pick the correct new front lens? of a focal length that those 2 combine to produce a resultant compound lens… that is 100mm focal length in total!
9) we get a free lunch here…the rear lens collects more light earlier, we have the higher efficiency we like… if it combines with a front lens that the compound lens is 100mm FL overall? we get the same size projected emitter! and its still brighter!

i know, how hard would that be? easy… i found the proper equation(s) to predict it, ha ha… and i can prove it.

you dont poop your pants over a little algebra, do you? go on… dare me.

No, the divergence of the beam depends on the LED distance from the first optic.
Once the light hits that optic then the divergence can only get larger, not smaller.
You can ignore all the other stages, because the first stage close to the LED will already cause more divergence than a large 100mm lens farther away.

Here is an extremely exaggerated example:


In order to simulate a non-point-source of light, the rays are coming from a few mm above y=0, representing light coming from some place that is not the center of the LED.
The closer the first optic is to the LED, the more divergence there will be.



the green line would represent a point source of light right at the focal point, which results in perfect collimation.
the red lines represent the minimum and maximum angles of light coming from the LED.
Hopefully you can see why a long focal distance is important.

PS- you’re right, you can use math to figure this out, it’s just a bit of trig.

i’ll get to the equation in a second… but first? I mean… if i may?

we have a problem. we want it solved. that much is obvious… and we even know what we want.

just like everyone else? you are spending all your obvious considerable intelligence, abilities, and energy? explaining why it CANT be done. what is the point of that? what do you expect that to get you? problems dont get solved, by looking for more reasons why it cant be done.

your second mistake? your point of view is all wrong, from the get go… you wonder IF this can be done… i dont think like that. I simply assume it can be done, then i assume that i’m the man who’s gonna do it. you might find this way of thinking? “arrogant”. I call it “confidence”.

the problems you are creating by doing this? are that everyone ELSE keeps hearing a list of reasons why its not supposed to be able to be done. they keep repeating all the reasons why ot cant be done to each other and to more people. now, instead of a bunch of smart people working on the problem in a systematic manner looking for answers? all that is going on is everyone ends up sitting around talking about what cant be done.

so, everyone ends up doing more or less the same things over and over… wondering why things arent getting better. I mean, Einsteins definition of insanity? “doing the same things over and over, expecting different results”

now, with THAT proper frame of mind? why dont we simply list the things we NEED, then go looking for them. we already know everything we need, you just dont realize it. its just a matter of a little time, a little reading, a little leg work? to find ALL of those things we need… in one simple equation.

see? thats yet another problem right there. everyone goes looking for the answer, and if its not already done for you, and handed to you? in a nice equation in “chapter 7” somewhere in a class? they assume there is… no answer. That doesnt necessarily mean there ISNT an answer? that just means no one has found it yet… or? much more likely? the answer is already there, somewhere… its just in an equation, being used for something else, and you just dont SEE that it just needs rearranged into such a manner as to give YOU what YOU want.

=

on wikipedia, the basic optics page? everything is right out in the open, easy basic equations. i laughed when i finally SAW it. The answer had been staring me in the face all along. I must have read it a hundred times, and just didnt realize it.

we have 2 lenses. each has a focal length. there is some distance between those two lenses, or, they could be touching. there will be some distance from the first smaller lens back to the emitter. and the whole new compound double lens? will have an overall new focal length.

if we knew those things? we would be done, obviously, right? lets list them…

F = new overall focal length
F1 = lens #1 focal length
F2 = lens #2 focal length
D = airgap distance between lens 1 and lens 2
BFL = back focal length, I.E. the distance from the back lens to the emitter source

Like most incredibly confusing problems? the answer is hiding RIGHT out in plain sight, for all to see, on wikipedia. Its right on the optics page, we probably most of us read it before, learning about lenses. i wanna remember they were callef the “lens makers equations” or something like that…

here they are……

if the 2 lenses are touching? ( 1 / F ) = ( 1 / F1 ) + ( 1 / F2 )

if there is an airgap between them? ( 1 / F ) = ( 1 / F1 ) + ( 1 / F2 ) - ( D / ( F1 * F2 ) )

and the back focal length is always BFL = ( F2 * ( D - F1 ) ) / ( D - ( F1 + F2 ) )

there… theres the keys to the kingdom. hiding RIGHT in plain sight, on the basic math of optics wikipedia page. I realized? there was every one of my variables i was going to have, and whats more important? i knew the value for every one of them! the only thing i didnt know? was what i wanted… namely, what is the resulting new compound lens’s overall focal length, if two lenses are separated by an airgap with a back focal length.

simple algebra tells us, that given any equation, if we have all variables known, and one unknown? we simply solve the simple algebra problem, to isolate the unknown variable we want to know on one side by itself.

in this case, I was primarily interested in knowing the overall new focal length of the compound lens… so i simply solved the airgap equation for F… because i was unable to FIND it stated like that anywhere else.

But, let me back up a hair? originally, i was just writing down all the known variables, as stated in the given equations. all numbers. and the “F” was what i would solve for. i used paper and pencil and my trusty scientific calculator. I just ran the numbers over and over, with different focal lengths and different airgap each time, until it produced a resulting focal length closer and closer to the exact focal length i was looking for, and also had a close enough back focal length it created efficiency.

this WORKS? problem is, i was doing the probloems on the calculator and paper/pencil hundreds of times, adjusting the variables up and down a little at a time, one at a time… until i GOT the exact resultant focal length i was WANTING. This was highly inefficient…

so, i simply started SOLVING the airgap equation, in order to ISOLATE the “F” by itself ??? THIS was the ANSWER! i only had to plug in the numbers and it spit out the answer much quicker, with way less key presses on the calculator, and, way less scribbling with penciul and paper to get each answer.

this cut my work down considerably, but… since i use to be a rather high end software engineer back in the day? i was like, “dummy! simply write a short visual basic program, that allows me to enter values in for all the known variables and let the software instantly spit out the answer every time i adjust one of the values.”

which was of course exactly what i did.

after that? all i did was sit there amazed? as i plugged in different focal lengths and airgap values, watching the new overall focal length get spit out… so, when i saw the answer i wanted? i knew which two focal lengths i was looking for to buy, and also how much distance to put between the 2 lenses. i also instantly knew the back focal length, so… i also knew how far away from the emitter to place the first lens.

this? allowed me to do a-n-y-t-h-i-n-g i wanted to do!

1) i can take 2 identical 100mm FL lenses? touching (airgap zero…) and know that it makes ONE 50mm FL lens! back focal length is trivial, its 50mm…

2) as i separate them with a bigger and bigger airgap? the new focal length inches UP until it gets almost to 100mm FL, and as this happens? the back focal length is inching its way
closer to the emitter (system getting more efficient)

3) i can instantly change either or both focal lengths, and adjust the airgap… and get instant running answer as i adjust things…

4) if i know i can only have certain focal lengths of certain lenses available to me on my workbench? i can pick amopng those values, and adjust the airgap until i find the resulting new focal length I am LOOKING for.

5) maybe i want to glue a certain small focal length lens, lets say 5mm off of the emitter? i just pick values and move things around until i SEE a back focal langth thats 5mm… and a overall focal length i WANT.

==

as you can see? i now had the keys to the kingdom. i can make or do anything i want to. and? since i know how to MEASURE the focal length of ANY lens of any style i pick up and find? i dont have to depend on happy chinese specs, i KNOW the focal length of ANY lens i put my hands on if i want to know.

there is practically nothing i cant do with this, and quickly. when i start playing and building on the bench with visible emitters? i simply blow cigarette smoke into the light path in a dark room, and i can SEE on a piece of white paper the exact diameter lens i need to put there to capture ALL the light of the system without losing any. AND i dont have to use a bigger lens than i NEED. And??? if i know i “need” two 100mm lenses to get what i want? i can also use two lenses pout together like this to create ONE of them, use THAT for “lens 1” and start playing with making two other lenses be the OTHER “lens 2”… thereby easily and fairly quickly creating a 3 or 4 lens system.

which puts the lenses in a row, with more regular space, capturing and shaping the cone of light along the path, adjusting to not “lose” any as i go.

you see? if i try to do something “actually impossible”? the equation tells you! you get an undefined divide by zero answer. or the back lens is touching the emitter. or, you run out of space or get an error before theres enough airgap to yield the focal length you are looking for? i simply bump up one of the focal lengths and bring it back into shape.

since i already know how to predict the SIZE of a known emiters projection ata known distance (differant easy trig equation, childs play)? if you want an xml2 to be for instance 100 feet across at 300 yards? it will tell you wen you solve the chief soh cah toa, what focal length you NEED as the overall focal length to paint that size emitter at that distance.

if it calls for a 200mm diameter lens to capture all the light? poop to thatr! we just start adding lenses… click clik clik… until we have our system solved.

==

surplus shed? has so many different lenses of different diameters and focal lengths? its ridiculous, they are all cheap… my simple program makes it easy to play what iof for a few minutes, until you get the criteria you want.

a simple “jig” on a workbench and some cigarette smoke and a piece of white paper? allows me to do the prpototype build.

once i fab up the prototyoe out of pvc or cardboard tubes? i can cast a piece of stock and machine up to make it out of aluminum.

===

you keep telling me made up reasons why it “cant” be done? i already DO IT on a regular basic before! i have SEEN units that do stuff like this, its not impossible. and i have MADE the stuff before, it works.

3 or 4 equations, and i can solve hem for what i want to find by isolating any variable i want with paper and pencil… then enter the equation in the visual basic program so its yet another answer being spit out. ( i like it when my software keeps a running tally of how big an emitter is “painting” at 100 yards, i find it a handy reference)

if you wanted? i suppose i could list several emitters, and let them ALL spit answers out, or, let the person adjust the distances to see different results.

its just a couple basic equations, and a short vis basic programi made before, i could make it again in short order.

=

now then?

you know that phrase, “hey, he puts his pants on one leg at a time just like everyone else.”

well, hate to break it to you? i dont… i kind of hold them in front of myself and just JUMP into them all at once.

and yeah? i do this same thing with imaging lenses too, for night vision.

i buy a lens with longer FL than i want? and drop the FL by adding a lens behind it… which lowers the F number to wjerever i want it to be.

since the F number controls how “powerful” the night vision unit will be? it should be obvious my units are, uhm… “better”. and i can do things the other night vision builders cant do. or that they think is impossible or priced like unobtainium.

dont you see? i am an ENGINEER, i used to make “mathematical models” for a living when i was younger. i have 2 university degrees in mathematics and computer science.

look at my tag line? “with enough black coffee and cigarettes? all things are possible.”

thats not just a phrase to me… thats what i “do”.

you keep telling me, that it is impossible??

then riddle me this. Where exactly do you think compound camera lenses COME FROM? they dont come from planet pluto, do they? no, they come from earth. humans make them. they exist.

why do you suppose those equations are called the “lens maker equations”, eh? Because compounds lenses CANT be MADE ?

if i want a compound lens, that has a variable focal length? i simply choose the proper two lenses, then slide them from touching, to further apart. Then i readjust the emitter distance to re-focus up. Easy.

Didnt you ever notice there are a LOT of camera lenses that are 2,3 or 4 lenses only? fairly simple construction? Where do you think they come from? They get made by people doing what i do.

the reason its not being done in the flashlight world or almost never? simple… all of the optics engineers work for the camera lens and binocular and telescope companies. Also, no doubt? The chinese flashlight factories? Are going to be like the IR illuminators back on my night vision site? They are just interested in basic fabrication, fast money out the door. They have no vested interest in making the design or build more complicated. I, on the pother hand? am not a “businessm,an” in a suit that has to answer to investors. Its in my best interests to design something “better” and make it. Particularly enjoying it, if it IS more complicated or advanced.

again, not trying to be RUDE, but… you keep telling me theres some “delta problem” which you allow to keep you from thinking you can do anything. Since i was not aware it wasnt possible? I just found a few equations, solved for what i needed and set about making what i wanted to make.

its amazing what you CAN do, when you are unaware of how you are not supposed to be able to do it. Which is of course news to me? since i have already done this stuff before already, in real life. I just needed to learn fabricating and machining to make my units i already have designed and working already.

engineers like me? we arent particularly or necessarily any “better” at math? than anyone else that took math classes. In fact, most of us really arent the best at it, to be perfectly honest. Its the WAY we were taught to USE it thats different. Also? “most” engineers? are unfortunately like “regular math people”, in that they need an equation to use.

the TRICK is? to go out anmd FIND something that has all the variables you are going to use. Then usually, you have to rearrange (solve and resolve multiple times) the equations INTO the form you NEED. If it was sitting somewhere, ready to use? Iy obviously wouldnt be a “problem”, now would it?

and, for me, a big dose of my frame of MIND going in. I concentrate on what i DO know, and go on from there. Or, I have other “tricks”. I orften start at the “solution”, and concentrate on creating the LAST STEP (missing) that obviouly led to the solution. Then i figure out what the step before that must have looked like. Eventually i end up at the “beginning” and i am done.

Or i start at the easy beginning… AND at the solution… and i “come in” from the beginning, and i come in from the end… and i “finish up” in the middle. In the end, i have all the steps. the fact i solve them out of order? doesnt matter when i am done.

==

in real life? MOST people seem to find me “intelligent… but eccentric”

and i can live with that. and so can my CAT.

Sedstar,

I know that a lot of guys on NV forum used your theory and practical experiments to build their own product(and since they probably have better machinery they really made great NV equipment) and now all those guys are making some serious cash $ and all that because you are unselfishly sharing your knowledge to the world even into tiny details how do you actually do something.

The main idea is important so lets say cool futon builds of some members were grown from your original idea and yet that guys does not seem to pay respect to inventor.

Enderman,
You are also one knowledgeable man and since I am not on degree like lets say Sedstar or Djozz(as examples) I really can’t comment or argue your theory.

I only know what I build and what I see and when I strictly concentrate on pure die projection of SR multilens system (I don’t look other reflections around) only conclusion I have is wow… This is clearest die projection I saw in my life. Seriously you see tiniest phosphor imperfections. If you don’t trust me invest around 90$ build your own and then you’ll see.

Sunnranger is modified UF-T20 flashlight on which into their stock threads only by pure accident and without additionally machining you can fit Yukon 3x42(or even bigger) night vision lenses.

And in my theory those NV AR coated lenses when U stick white led into SR are preventing or obscuring throw capabilities that usually don’t happens with osram IR emitters.

42mm multilens IR SR is very close to performance of 67mm single aspheric. So I don’t even doubt that with right choice of lenses like Sedstar mentioned something fantastic and in small and acceptable size can be made.

Are you ok?
I never said something wasn’t possible, I just explained why I don’t like multi-lens systems.

Also, from your wikipedia” equation, it is already obvious that you can’t create a longer focal length than simply using one lens at the max distance.
Adding more lenses closer to the LED will always shorten the focal length. And I already explained what a shorter focal length does, it makes the delta angle larger because an LED is not a point source of light.
You keep talking a lot without providing any proof of what you’re saying. If you want to test out your theories then download a ray tracer and test it out, see for yourself if you can make it work.

PS- stop bragging about how many degrees or potatoes you have, nobody cares. One thing I learnt from university is that a degree says absolutely nothing about a person’s intelligence or ability to actually build something.

Nobody is stopping you from building a multi-lens flashlight, so go ahead and do it and come back with more results and less talk :wink: Maybe we should decide on a diameter and I will make a single-optic flashlight and you can try to beat it with a multi-optic design?

This is rude man… Why rudeness?

Edited
He was being condescending.

Now that’s the spirit… :+1:

Lets build it! I bet that multiple lenses system will beat a single lens one.

P.S.

I almost forgot on Ervin Anastasi…? Check this out: The other million bet beam

Sedstar,

Could this lenses be good contender?

I mean they don’t have AR coating like Yukon lenses? And it is about 4 lenses inside that? Is that worth a try? I mean I should machine the tube and pill but it should be doable?

Wow fellows, what a buttload of useless talk up there!

I think sedstar is trying to point out that the problem can be solved, and indeed it can. To the point that the emitter's size, and wether it is domed or not, would have very little impact on the final result for as long as the last lens (main collimator) were to be large and powerful enough with very low focal distance for maximum convergence and sufficient size for minimal loss of light:

Zoom lens @ Wikipedia

Of course, this implies a good quality host with fine threads for the lens' assembly.

Cheers

I disagree… This talk was very useful to certain point.

Multiple lenses flashlight/illuminators are not commercial available but they exists and I am one of plenty diyers that have build such system so I know that it works with IR light and that with little luck(proper lens selection) it should work with visible led light.

So there is no point that someone who builds only single aspherical system claims that multi can’t be done to guy or guys who have already build such systems.

There’s nothing that can’t be done. And if you can have for example 500kcd flashlight in a 50mm head diameter than it is worth a try.