XPG3 The upgrade?

[Enderman]

Generally speaking people seriously into throw aren't as concerned with high CRI. Not that aren't some out there but that would be a niche within a niche.

Can confirm, if there was an option for a pure green LED that produced higher intensity I would use that in my thrower xD

 

[parametrek]

All it is is a hemispherical reflector with a hole in the middle, so if you really want one you can get someone like phoenix or optiforms to make you one and it should work just as well.

Plastic christmas tree balls are silvered on the inside, right? A couple minutes with an xacto knife should do the trick for a crude collar.

 

[dc38]

Plastic christmas tree balls are silvered on the inside, right? A couple minutes with an xacto knife should do the trick for a crude collar.

I believe a few of us came up with this idea several years back, lol

 

[dc38]

Hey mr. SAAB...the issue is the viewing angle. With the dome on, according to the specs sheet, xpg2 is 115, whereas the xpg3 is 125. 10 degrees difference.

**i see you have already noted and accounted for that**

 

[dc38]

If you are willing to lose a few LEDs, instead of a phosphor transfer, try a complete removal. See if the unfiltered blue is any more intense or a higher wavelength than with the xpg2. Maybe they are using a higher wavelength of blue in tandem with a new phosphor with a higher activation point.

Surface brightness may have decreased, but overall brightness imcrease in conjunction with the higher viewing angle may compensate for that to offer more lunens per watt overall.

 

[degarb]

Got to go find a you tub video of this Christmas ball trick.

How long soak in h2o. . I haven't Google your dedome procedure. . Just perm of silicone

Sealant Type
Exposure Time Before
Condensation Detected
3M 606NF Acrylic 24 Hours
TET 1034-01 Epoxy Data Not Meaningful
SIKAFLEX 221 Polyurethane 72 Hours
DOW 738 Silicone Rubber 24 Hours
Condensation also appeared inside the epoxy sealed TEM
within 24 hours of exposure. This TEM, in fact, absorbed moi.......

 

[degarb]

Enderman an has quite the YouTube channel. If same guy, I am impressed with editing, electronics, and the pvc cannon. . Great for pizza delivery guys, electric pole repairman. Maybe looking for dead bodies at night on a mountain side. . I may be forgetting a few othr occupations.

I impressed myself the other night when my homemade xpl hi in a 42 mm allowed me to see a slight ghost in a paint job at 70 foot distant and 40 foot high. Only 12k candela neutral, 309 lumens, as I need 8+hour runtime and 2 cell max. . However, I could see ghosts on a 18 foot high ceiling better with a 12k candela Fenix cool led back in May, confirmation of depth of field argument that the late, great transportation sector guy espoused. But, my throw is nothing in comparison. . I guess over 100 meter viewing is the target audience. . Would be fun on a walk to have this range. May need to build into your walking stick.

Enderman, my hats off to you. Which is saying a lot, as it was under my headlamp. Though, the wearability of your lights is a tiny issue for me... I couldn't tell from skipping through much about the waiven collar orientation. . It looked oriented inward, which could not be right. . Obviously, I have no use for the big cannon. . But, wondering if this collar has been adapted for zooming 26mm class headlamps and flashlights.

 

[Enderman]

Enderman an has quite the YouTube channel. If same guy, I am impressed with editing, electronics, and the pvc cannon. . Great for pizza delivery guys, electric pole repairman. Maybe looking for dead bodies at night on a mountain side. . I may be forgetting a few othr occupations.

I impressed myself the other night when my homemade xpl hi in a 42 mm allowed me to see a slight ghost in a paint job at 70 foot distant and 40 foot high. Only 12k candela neutral, 309 lumens, as I need 8+hour runtime and 2 cell max. . However, I could see ghosts on a 18 foot high ceiling better with a 12k candela Fenix cool led back in May, confirmation of depth of field argument that the late, great transportation sector guy espoused.

Enderman, my hats off to you. Which is saying a lot, as it was under my headlamp. Though, the wearability of your lights is a tiny issue for me... I couldn't tell from skipping through much about the waiven collar orientation. . It looked oriented inward, which could not be right. . Obviously, I have no use for the big cannon. . But, wondering if this collar has been adapted for zooming 26mm class headlamps and flashlights.

Thanks for the compliments yes that is my channel, and yeah the flashlight at it's current size is not very practical for daily use xD
The collar is supposed to make a "dome" over the LED, so that any light going to the sides gets reflected back inside, and light can only leave through the small hole.
Basically, a backwards reflector, that is spherical instead of parabolic.
Even with the collar there is still a lot of light wasted because the lens I use has a pretty long focal length and there is light that does not hit the lens.

AFAIK zooming lights (like my P5r.2) use TIR lenses which basically go around the LED to pick up all the light, not just the light going forward.
TIR lenses will not focus as tightly as a regular aspheric lens like I'm using, but that gives a more usable spot, and also much higher lumen efficiency.
If I was to estimate, there is not even 500lm coming out of my light.

 

[saabluster]

Got to go find a you tub video of this Christmas ball trick.

How long soak in h2o. . I haven't Google your dedome procedure. . Just perm of silicone

Sealant Type
Exposure Time Before
Condensation Detected
3M 606NF Acrylic 24 Hours
TET 1034-01 Epoxy Data Not Meaningful
SIKAFLEX 221 Polyurethane 72 Hours
DOW 738 Silicone Rubber 24 Hours
Condensation also appeared inside the epoxy sealed TEM
within 24 hours of exposure. This TEM, in fact, absorbed moi.......

The old XRE would be good with a three day soak in h2o. Then run 1A to the LED briefly. It will flash the water right next to the die to steam and separate the two. Newer formulation encapsulants do not work with water. At least at the soak times I tried them at. Found several chemical cocktails that work on the newer stuff. To some extent everyone caught up to all my little tricks in this department. Well not all. I've never come out and said anything before but people using gasoline is just gross. Leaves a residue embedded in the phosphor layer too.

 

[saabluster]

Thanks for the compliments yes that is my channel, and yeah the flashlight at it's current size is not very practical for daily use xD
The collar is supposed to make a "dome" over the LED, so that any light going to the sides gets reflected back inside, and light can only leave through the small hole.
Basically, a backwards reflector, that is spherical instead of parabolic.
Even with the collar there is still a lot of light wasted because the lens I use has a pretty long focal length and there is light that does not hit the lens.

AFAIK zooming lights (like my P5r.2) use TIR lenses which basically go around the LED to pick up all the light, not just the light going forward.
TIR lenses will not focus as tightly as a regular aspheric lens like I'm using, but that gives a more usable spot, and also much higher lumen efficiency.
If I was to estimate, there is not even 500lm coming out of my light.

Just a suggestion. Ditch that 100mm lens. At least all the ones I have ever gotten were junk. Look on eBay for tv projection lenses. Most of them are Delta branded. If you don't care about beauty and formfactor that is the route to go for serious power. There is a reason I am way way way beyond 1.4Mcd in my personal projects. That is a major one. Really good optics.

 

[Enderman]

Just a suggestion. Ditch that 100mm lens. At least all the ones I have ever gotten were junk. Look on eBay for tv projection lenses. Most of them are Delta branded. If you don't care about beauty and formfactor that is the route to go for serious power. There is a reason I am way way way beyond 1.4Mcd in my personal projects. That is a major one. Really good optics.

What size though? I don't really want to go over 4"...

What distance are you taking your measurements at, and are you using the proper way of measuring candlepower/cd/lux or are you using the basic "multiply by distance^2" method?
http://nightsword.com/uniformbeamcalc/
I realized I was taking my lux measurements by the basic way and that is completely incorrect, it was giving me values way lower than they actually are, I should be getting like 3Mcd or more when calculated correctly....
Although I need to go outside and do a test at 50 or 100m to get the true values.

I looked at professional grade optics before, the only 100mm aspheric lens is from thorlabs and it costs $1150.
It NEEDs to be an aspheric lens, I've already wasted money on a lens from edmund optics because I thought it was aspheric but it wasn't.
I don't know which tv lenses you're talking about which are aspheric, but any leads/info would be appreciated.

Currently I have found this one, and although there is no information on whether it is aspheric or not, I think I'm gonna risk it:
http://www.bmisurplus.com/products/44316-plano-convex-lens-4-in.-diameter

 

[saabluster]

What size though? I don't really want to go over 4"...

What distance are you taking your measurements at, and are you using the proper way of measuring candlepower/cd/lux or are you using the basic "multiply by distance^2" method?
http://nightsword.com/uniformbeamcalc/
I realized I was taking my lux measurements by the basic way and that is completely incorrect, it was giving me values way lower than they actually are, I should be getting like 3Mcd or more when calculated correctly....
Although I need to go outside and do a test at 50 or 100m to get the true values.

I looked at professional grade optics before, the only 100mm aspheric lens is from thorlabs and it costs $1150.
It NEEDs to be an aspheric lens, I've already wasted money on a lens from edmund optics because I thought it was aspheric but it wasn't.
I don't know which tv lenses you're talking about which are aspheric, but any leads/info would be appreciated.

Currently I have found this one, and although there is no information on whether it is aspheric or not, I think I'm gonna risk it:
http://www.bmisurplus.com/products/44316-plano-convex-lens-4-in.-diameter

The Delta 260, 265, 79 and many others are almost all in the same range. 4-4 1/4" diameter. Would probably slot right into that pipe. They are close focusing so capture a ton of the light. They are aspheric but not in the way you are, at this point, used to thinking about it.

Throw lights are not just an "interest" for me. I have lived, eaten, and breathed all things throw for a decade. I have gone down all the obscure alleys in search of solutions. Optically, and at this size, I have found nothing that will beat it. I have bought the extremely expensive lenses trying to get my next throw "fix". Only to find that they are rubbish for our application. With one exception which was a small lens that I put into one of my production lights.

My test distance is roughly 30'. It absolutely does not require going to 100meters to get the "true" values. As long as your setup is good. This is not the thread to get deeper into it so I will leave it at that.

 

[dc38]

Fiber optics?

P.s., I am aware of the precollimator, but I was thinking more along the lines of using a larger lens closer to the led to focus the image onto the entire back surface of a smaller projecting lens, which would refocus the rest of the light into an almost pencil beam. Like a reverse telescope.

 

[Enderman]

The Delta 260, 265, 79 and many others are almost all in the same range. 4-4 1/4" diameter. Would probably slot right into that pipe. They are close focusing so capture a ton of the light. They are aspheric but not in the way you are, at this point, used to thinking about it.

Throw lights are not just an "interest" for me. I have lived, eaten, and breathed all things throw for a decade. I have gone down all the obscure alleys in search of solutions. Optically, and at this size, I have found nothing that will beat it. I have bought the extremely expensive lenses trying to get my next throw "fix". Only to find that they are rubbish for our application. With one exception which was a small lens that I put into one of my production lights.

My test distance is roughly 30'. It absolutely does not require going to 100meters to get the "true" values. As long as your setup is good. This is not the thread to get deeper into it so I will leave it at that.

Well I'm not sure what you mean by "not in the way I am used to thinking about it", because there are aspherical lenses, and spherical lenses, and what I need for a thrower is an aspheric lens...

Do the numbers on the delta lenses mean anything like focal length or something? Obviously not diameter if you say they are all ~4" but I see a lot of other numbers like delta 77, 78, etc..
Also, I'm not really looking for "close focusing" Because even though you capture more light and the lux is the same you end up with a bigger spot, and I would rather have a thin beam than a high divergence beam.
For example, the lenses used on the deft-X flashlights which is really thick, you can see in the pics just how much divergence there is due to the short focal length. The one I currently use is 14cm focal length.
I just looked at a youtube video of a delta 78 lens, and it actually seems like it has a good thickness, not too thik, so maybe your definition of close focusing is exactly what I need

I know throw lights are your specialty, you're the guy who built the deft X and victor lights, right?
But the problem is that the lens I am using is very flat has a convergence point at least 10m behind the light, so my lux measurements at 5m are off by millions.
Should I make a separate post to talk about this?

Anyway, I just received my XP-G3s and will do some tests today to see how disappointing it ends up being xD

 

[Enderman]

Fiber optics?

P.s., I am aware of the precollimator, but I was thinking more along the lines of using a larger lens closer to the led to focus the image onto the entire back surface of a smaller projecting lens, which would refocus the rest of the light into an almost pencil beam. Like a reverse telescope.

It's unfortunately not possible to collimate light into a smaller diameter beam then the origin :/ when you reduce the beam size you increase the divergence ending up with a non-collimated beam that just spreads out more than before.
I think it is possible with a theoretical point source of light, but that's not possible irl :C

 

[saabluster]

Fiber optics?

P.s., I am aware of the precollimator, but I was thinking more along the lines of using a larger lens closer to the led to focus the image onto the entire back surface of a smaller projecting lens, which would refocus the rest of the light into an almost pencil beam. Like a reverse telescope.

You can do that but it would increase the apparent source size as opposed to the collar. The end result being less throw. Not sure what you are asking in regards to the fiber optics.

 

[saabluster]

Well I'm not sure what you mean by "not in the way I am used to thinking about it", because there are aspherical lenses, and spherical lenses, and what I need for a thrower is an aspheric lens...

Do the numbers on the delta lenses mean anything like focal length or something? Obviously not diameter if you say they are all ~4" but I see a lot of other numbers like delta 77, 78, etc..
Also, I'm not really looking for "close focusing" Because even though you capture more light and the lux is the same you end up with a bigger spot, and I would rather have a thin beam than a high divergence beam.
For example, the lenses used on the deft-X flashlights which is really thick, you can see in the pics just how much divergence there is due to the short focal length. The one I currently use is 14cm focal length.
I just looked at a youtube video of a delta 78 lens, and it actually seems like it has a good thickness, not too thik, so maybe your definition of close focusing is exactly what I need

I know throw lights are your specialty, you're the guy who built the deft X and victor lights, right?
But the problem is that the lens I am using is very flat has a convergence point at least 10m behind the light, so my lux measurements at 5m are off by millions.
Should I make a separate post to talk about this?

Anyway, I just received my XP-G3s and will do some tests today to see how disappointing it ends up being xD

Yes there are aspheric and spheric lenses. And then there are aspheric and freeform aspheric. Technically you do not need aspheric lenses for flashlights. You would just need more spheric lenses to accomplish the correction that one aspheric can provide. The Delta lenses have multiple lens groups with insanely complex correction in freeform lenses. In an older test with the DEFT-X light engine I got to 2.8Mcd with the Delta 102. I have no idea if there is any significance to the numbers they assign the lenses. Not sure why you'd want as thin a beam as possible. If that is what you like then fine but in my experience it is far more impressive to have a wider beam even if slightly less powerful than a smaller beam. As in as long as the wider beam is even remotely close in intensity it will appear to the eyes as more powerful.

Yes I am the designer of the DEFT-X and Victor. You are in my forum.

You can still get accurate readings with that lens at 5 meters. As long as your setup is calibrated for it. You would simply need to focus the projected image at the plane of the sensor and assure the sensor was fully covered by the beam. If the beam just covers the full sensor then it would be a mean rather than peak reading. This can be made finer by covering up the sensor with black electrical tape leaving a small opening so that you are capturing the intensity readings of smaller parts of the chip. All you need then is a correction factor.

 

[dc38]

You can do that but it would increase the apparent source size as opposed to the collar. The end result being less throw. Not sure what you are asking in regards to the fiber optics.

Could you use fiber optics to route the light into the most desired pattern before focusing it? For example, "straight" path of light to make it more predictable instead of dealing with the wide viewing angle and hoping that you catch all the light. LED, FIBEROPTIC, ASPHERIC instead of LED, SPACE, ASPHERIC

 

[Enderman]

Yes there are aspheric and spheric lenses. And then there are aspheric and freeform aspheric. Technically you do not need aspheric lenses for flashlights. You would just need more spheric lenses to accomplish the correction that one aspheric can provide. The Delta lenses have multiple lens groups with insanely complex correction in freeform lenses. In an older test with the DEFT-X light engine I got to 2.8Mcd with the Delta 102. I have no idea if there is any significance to the numbers they assign the lenses. Not sure why you'd want as thin a beam as possible. If that is what you like then fine but in my experience it is far more impressive to have a wider beam even if slightly less powerful than a smaller beam. As in as long as the wider beam is even remotely close in intensity it will appear to the eyes as more powerful.

Yes I am the designer of the DEFT-X and Victor. You are in my forum.

You can still get accurate readings with that lens at 5 meters. As long as your setup is calibrated for it. You would simply need to focus the projected image at the plane of the sensor and assure the sensor was fully covered by the beam. If the beam just covers the full sensor then it would be a mean rather than peak reading. This can be made finer by covering up the sensor with black electrical tape leaving a small opening so that you are capturing the intensity readings of smaller parts of the chip. All you need then is a correction factor.

That's really cool, the deft x was a great inspiration when I was just getting into flashlights

My goal isn't as thin as possible, but as collimated as possible while still being usable. Obviously I could get a 2m focal length lens and have far better collimation but the amount of light hitting it would be very small.
From all the lenses between 70 and 100mm I have tried, the fasttech one works the best by a long shot, but I'm still looking for something of better optical grade because the spot is not very nice.
I like a near-cylindrical beam better than the large spread that comes from those ~78mm lenses used in the deft, victor, and "black bullet" flashlights, the first pic in the black bullet link shows just how much the beam diverges.

Also about the delta lenses, are you saying that you use the whole lens assembly in front of the LED?? Not just one of the lenses?
I looked at this video and they do seem ~4" but I think instead of using an aspherical lens there are two spherical ones and a spherical aberration compensation plate (which is that thick glass thing)
So if you used the whole assembly then yeah it would not have spherical aberration but you're still using like 3 or 4 pieces of glass and reducing the light transmittance at each one...a single aspheric lens would be much more efficient no?
The delta lenses on ebay are really cheap though so maybe I'll just buy one and see for myself if it will beat the 100mm lens I currently use...

 

[wimmer21]

This thread is fascinating.

Good luck with all your trials, Michael. I'm sending all my good vibs to Garland!