Need fine engineering help please! desperate (thank you)

My LED supplier tells me that the XRE R2 has a viewing angle of 115deg and the Q3 has a viewing angle of 90deg. (despite whats written in the datasheet)

Is this correct? Does anyone know the viewing angle of the Q5? The supplier only has the specifics on the R2 and Q3.
 
Q5/R3, etc have nothing to do with the viewing angle.

The XR-E has a 90* viewing angle.

The XP-E/XP-G has a 115* viewing angle.

Your supplier is talking about two different types of LED's.


Q5/R2/R5 just refer to the BIN of the LED. A bin is the group of LED's with similar efficiencies. Theoretically, you could have a super efficient R5 XR-E, with the EZ-900 die, and a very poor Q3 efficiency XP-G. But either is not likely to happen...
 
I think what is called for here is a TIROS optic - a TIR with an integral aspheric at the business end.

Inova used them in flashlights for a good long while, and they produced some fantastically well collimated beams, lacking badly in the spill, with just some ringy wasted light from odd interactions. Can you afford to buy an old TIROS inova to play with?

I have *no* idea if this image is accurate to the angles involved, but you can see how the TIR lens and the aspheric are cast as one piece here:
Inova_TIROS_500px.jpeg

2008_4_10_51645_7151645.jpg


There are rings and ridges (not sure how to describe it) on this generation TIROS optic; they seem to be there to moderate the laser-like focus of a typical aspheric into a nice round spot, instead of the typical aspheric image of the die. Later (current) TIROS revisions removed the aspheric portion entirely, however.

Also - what about including a heat pipe or two in the light to pump heat to the outside of the case?
 
Last edited:
Double your light output with no extra heat

So how have our suggestions helped ?

In another thread I've done some calculations, based on analysis done by bbb74.
http://www.candlepowerforums.com/vb/showthread.php?p=3490781#post3490781

If you increase lens size so it captures +/-50 degrees instead of +/-40 degrees AND switch from an XP-G to an XR-E, the percentage of light captured from the LED increases from 44% to 94%.

Seems an easy way to achieve your goal in the original post, of 15% increase.

btw you never answered my question about the +15% goal. A person would have great difficulty detecting a 15% increase - is there some minimum lighting level you need to achieve to comply with an industry standard, or is it to match a competing product.
 
The apparent die size of an XR-E EZ900 is quite less than that of an XP-G, the XR-E EZ1000 is somewhere in between.

You could use an XR-E R2 EZ900 LED, as others already said. The spot will be smaller, but brighter, since that LED still has the highest surface brightness around, about 40% more than an [email protected].
The smaller viewing angle doesn't really help for spot brightness though.

If you increase lens size so it captures +/-50 degrees instead of +/-40 degrees
A strong aspheric lens usually can capture +/-35°, so 40° or even 50° is quite a lot...

If heat sinking is good enough for 0.85-1.0A, it might be good enough for 1.1A, too?
 
Last edited:
Re: Double your light output with no extra heat

So how have our suggestions helped ?

In another thread I've done some calculations, based on analysis done by bbb74.
http://www.candlepowerforums.com/vb/showthread.php?p=3490781#post3490781

If you increase lens size so it captures +/-50 degrees instead of +/-40 degrees AND switch from an XP-G to an XR-E, the percentage of light captured from the LED increases from 44% to 94%.

Seems an easy way to achieve your goal in the original post, of 15% increase.

btw you never answered my question about the +15% goal. A person would have great difficulty detecting a 15% increase - is there some minimum lighting level you need to achieve to comply with an industry standard, or is it to match a competing product.


The 15% increase that is needed isn't necessarily exactly 15%, its just my estimation. Basically brighter is better, thats all. The current model needs to be a bit brighter in order to be useful. Its currently bright enough in the center of the spot, but dies off at the edges enough to loose its functionality.

I've taken everyones help into consideration and am now testing prototypes with the XRE R2 with biconvex and plano-convex lenses. So far tests have indicated that this is exactly what needed to be done. I cannot thank everyone on this forum enough. I hope to find a way to return the favor.

With the XRE, I've got 2 lenses that almost work perfectly:
Biconvex 14mm diameter focal length =12mm
Plano-convex 15mm diameter, focal length = 18mm (coated)

Both can produce the correct sized spot and at the correct brightness. Ideally the biconvex focal length should be between 10.5-11mm and the plano-convex somewhere between 12-15mm focal length. (both in 15mm diameter-largest apatite as Dr.Jones also indicated. Indiana Jones?:thumbsup🙂 I cannot correctly calculate the perfect focal length though. I've tried all forms of calculations and have come up with little more than estimations.

I cannot for the life of me understand why first of all the plano-convex produces a much cleaner spot, and second, why a longer focal length plano-convex can produce the correct focus when compared to a shorter focal length biconvex. Clarification on this would greatly help.

The following is a picture of current tests and the lenses to be made. I greatly appreciate any input.

(Note: Diffuse focus because this is how it will be used. Spot on right has correct sized spot but too dark at edges.)
Plano-convex spot does not appear to be darker than the biconvex in real life.
biconvexlens.jpg

planoconvexlens.jpg

newbiconvexlensdeasign.jpg

newplanoconvexlensdeasi.jpg

There is a TON more information I've learned, but its too much for one post. For example, aspheric lenses are good, but have so many weird angles going on that they end up producing a good center spot, but the edges are significantly fuzzed at best. Also, why do smaller aperture biconvex lenses have a radius of curvature less than their focal length sometimes(this seems to be for very short focal lengths)? Why do plano-convex lenses act so differently when you place them outside their designed focal length when compared to biconvex? - all things that make it more difficult to even estimate the proper focal length. Basically I don't know if these new lenses will have too short of a focal length or now and have no idea how to judge.
 
Focal lengths are only easy to use for thin lenses, i.e. lens thickness is negligible compared to the focal length, which is not the case here. With thick lenses, things get more complicated: the effective focal length is measured from somewhere within the lens, and that spot actually is different for the two different lens orientations (with plano convex lenses).
That's one reason why there are two focal lengths given for the plano-convex lens: The back focal length is measured from the plane side to make positioning easier.

About plano-convex vs. biconvex: You have a strongly divergent beam on one side and a collimated beam on the other, and it may sound plausible that an asymmetric lens is better suited for such an asymmetric ray setup. (A better explanation might be that you try to split the required total refraction as evenly as possible between the two refractive surfaces to minimize aberrations.)

I think an aspheric would do an even better job at that relatively low f/#.

About your last post: Maybe it uses a different kind of glass with a higher refractive index?
 
Re: Double your light output with no extra heat

I cannot for the life of me understand why first of all the plano-convex produces a much cleaner spot,.

I think it's mainly due to low angle-of-incidence where the light in air strikes the glass at sharp angle. This is what happens on the outer parts of the lens when we have the lens so close to the LED.

Any imperfections in the surface of the lens will cause great deviation from the ideal raypath when this angle is very low. In a bi-convex lens this angle will always be much lower than a plano-convex lens.

The ideal to avoid low angles would be a Concave-Convex lens, but I've never had one that I could use to test this out. I do have a Plano-Concave lens that came out of a Video Projector that I may try.

Although I do have two similar Plano-Convex lenses - one produces a bright sharp image, the other a dimmer softer image ????
 
Last edited:
Yes, that's what I meant.

However, a concave-convex lens is not the best thing to have in general, but so-called 'best form' lenses are - as long as you stick to spherical singlet lenses: Thorlabs best-form lenses


Although I do have two similar Plano-Convex lenses - one produces a bright sharp image, the other a dimmer softer image ????
I don't know what those lenses really are, but maybe one is an aspheric? There's a thread on good and bad aspherics with pictures: Another aspheric lens question
 
Yes, that's what I meant.

However, a concave-convex lens is not the best thing to have in general, but so-called 'best form' lenses are - as long as you stick to spherical singlet lenses: Thorlabs best-form lenses

I can't see a lot of these appearing in torches -
- They're $80 even with no AR coating
- They have an Aperture of F1.6 (40/25) - the really bright torch lenses are generally F0.8.
- These are only collecting a narrow angle of light.
- With shortest focal length of 40mm for a 25mm lens, they make for big torches
- Because the lenses aren't close to the LED, they won't have the low incident angle problem of F0.8 lenses.
 
That's why you use aspherics for low f/# lenses.
I just wanted to point out that those best-form lenses are the best spherical singlet lenses.
Spherical lenses are much more easy to produce in high optical quality than aspheric ones, which is why they still have their use (compared to aspherics).
 
Hi Eatkabab,

It looks like we've been working on very similar projects. I've struggled with the same optics problem and may have a solution for you, or at least some fresh perspective. Please drop me a line at r dot c dot hutchinson at gmail dot com.
 
Did you ever finish this project? I might be interested in buying one, depending on the final price.
 
I sent you a PM but I'm not sure you got it. There doesn't seem to be a working "sent items" folder on this forum...
 
Got it, thanks. I sent you a PM too but I also don't know if you got it and it didn't show in my sent box(I think there is a box we are supposed to check to have it appear there). My DIY light is nearly complete, I will post a few pics and specs when it is finished. It will probably not look as nice as yours because I have just been ordering different pieces and parts trying to make things work trial and error style. My last obstacle is finding a universal mount, I have one for "through the lens" but not for "flip-ups".
 
Back
Top