Project Excalibur - Next Generation LED Thrower (many pics) - UPDATE 2018-01-24

Just finished reading through the whole post :slight_smile:
It’s a pretty big coincidence that we both build a light with the black flat at almost the same time :stuck_out_tongue:
I am almost finished my build, just waiting on some liquid metal thermal paste.
.
.
.
“Minimum distance for accurate measurement of luminous intensity:

Many of you know that one cannot measure the true luminous intensity (throw) of a thrower in 1m distance it the value is supposed to be accurate. The reason for this is that in this distance in some lights the reflector is not completely lit up by the LED as seen from the sensor of the lux meter. One has to go further away and adjust the value for the longer distance. TLF member sma once studied this in more detail here (link is external). We both have the theory that it depends on the difference in size between the reflector and the LED. I have tried to calculate this for my light based on his results with an Olight SR-95 UT.”

I have done some testing, it’s pretty interesting actually.
If you take a piece of paper, poke 3 small holes in it, and put it in front of the reflector/lens of a flashlight you will get 3 separate squares (the die image) projected onto a wall.
In “infinite focus”, which is what you want for ideal maximum throw, these three squares will be centered exactly at the 3 points.

To get a proper lux reading when focused like this, the lux meter needs to be placed at the center, and the flashlight needs to be far enough away that all squares overlap in the middle, including the ones from the outer edges of the lens/reflector.

With this exaggerated huge LED example, all squares would overlap at 1.1m (so you would measure slightly farther than that)
.
However, with a focusable lens or reflector, you can actually change the focus and make all squares overlap at the luxmeter.
When they are all overlapping is when you get the “sharp die projection” image.
This allows you to take the measurement at a closer distance and get the same cd value when you calculate it back to 1m.
You need to keep in mind though that focused at the luxmeter will not give maximum throw. It should be collimated using that template I posted above for best throw.
.
For your flashlight, I calculated a minimum distance of 28.5m when it is collimated correctly.
If you want to know how I calculated this stuff just let me know, I can make a separate topic explaining everything.
To measure the maxabeam, I’m not sure how far it would need to be, because unlike an LED the small burning point is not all the same intensity like the surface of an LED is.

That’s a neat method with the 3 holes in the paper, enderman. The condition that there is some overlap of all three squares is equivalent to all of the reflector area being used. Like if you looked straight into the reflector at that distance it would be filled with LED/light.

Exactly :slight_smile:

I figured this out because you said that about looking into the reflector and seeing yellow.

:open_mouth: All I can say is…

https://www.youtube.com/embed/\_P5pmkJzAjo

Great job! :sunglasses: :beer: :beer: :beer: :beer: :beer:

Hi and welcome Michael. I greatly admire your skills and abilities. You must be a very patient person. :slight_smile:

Wow !

Very well done! From concept to application, brains to brawn as it were, and success in the end to make it all worthwhile!

Michael, your machine work is excellent and I hope one day to be able to get at least in the ball park, still on the learning curve with the lathe and now in a long delay from surgery but very ready to get back out there!

Mill work is beyond me for now, but obviously it’s the necessary other half to lathe work, combined there’s simply nothing you can’t accomplish and I love seeing it presented! Thank you both for all the effort , quality stuff right there! :smiley:

This must be the farthest throwing LED+reflector flashlight in existence right?
:stuck_out_tongue:

BTW to prevent your lenses from cracking in the future I would suggest buying some adhesive neoprene tape that’s 1/32 or 1/16” thick, then cutting a thin strip and putting that where the front bezel holds the lens.
It basically acts like an o-ring, putting soft pressure that isn’t too concentrated, and it creates a waterproof seal.

It has adhesive on only one side, so it won’t stick to the lens and make it hard to open.
If you want you can also put some on the other side of the lens, so that the edges of the lens are sandwiched between two layers of foam.

Very nice project and a great write-up. Great job, Michael and The_Driver!

Amazing build , nice nighttime bomber searchlight! :smiley:

So it is 1,5 milion candela reflector light and once when we’ll have commercial available LE UW Q8WP Osram emitters it will even surpass that results.

I am blown away by this. simply wow.

Very nice outcome, & I enjoyed the detailed build post.

I can appreciate how difficult it is to focus such a large reflector on such a small point!

I think I’ll have to update one or two of my throwers with an Osram black…

:+1: :beer: :beer: :beer:

I might have missed it but a few questions:
1 what is the depth of the reflector /or/ depth led mounted inside the reflector from the top?
2 what is this black thing in the center of the lens?

The LED is a 3V one right? Why choose this complicated driving engine?

Can you / do you want to give an estimate on the costs of this monster?

Thanks for the tip, we will take a look ;).

Concerning “the record”:
As far as I know this is the farthest throwing portable led flashlight which has a single LED with a reflector. Thats also what I meant when I wrote that it’s a lot of work and rather difficult to reach even higher values with this kind of light.

The farthest throwing led flashlight with multiple LEDs and reflectors is Lumensteins Lux Monster from Xandre. It is much, much bigger though (plastic spotlight host)! With around 6000 lumens it is also more practical, the spot is much bigger. It used to have de-domed XP-G2s and produced 2Mcd with 2000 lumens. Now it does 1.6Mcd with 6000 lumens.

The farthest throwing portable LED “flashlight” is Photons “Scheinwerfer”, it has a very large spheric lens and does 3Mcd.

I would have to check, I don’t know exactly.
The distance from the focal point to the outer rim of the reflector is 9.4cm (this is what I mean with “maximum focal length”).

Michael drilled a hole into the center of the lens because I had further ideas on how to increase the throw. There wasn’t enough time to implement this yet and we don’t know if it will work at all. I will keep it a secret for now ;).

Yes, the LED is 3V (technically it’s almost 4V, the Black Flat has a very high Vf).

This driver is of very high quality in terms of the components used. I don’t think there is a higher quality buck driver for LED flashlights. That is one reason why it is so expensive (also because it’s sold in Germany and the guy is trying to make money with it). It has all the features I wanted (over discharge protection, “nerdy” status LEDs, possibilty of connecting thermal probe) and can go up to a last 11A. I wanted to able to upgrade the light with a better LED and be able to utilize it fully. Newer LEDs are able to take more and more current.

No, not completely. One reason is that I was lucky on multiple fronts. If someone else were to build the same exact light they would pay much more money.
Most importantly you need a large, electroformed precision parabolic reflector and a matching head with large diameter. Where are going to get that and how much will it cost?

If you ask Peakbeam Systems how much a spare head with reflector costs, you will probably fall off your chair as we like to say in Germany. :smiley: I have heard that they want around 170€/$ for just the glass lens!

- you need multiple Osram Black Flat LEDs for testing, not all are good (but we had more good ones than expected)

- the driver you can replace with a cheaper one, but it needs to be a buck driver (the Convoy L6 driver, FX-30, might be the cheapest, it does 5A from 2S cells, but I don’t know how well it works with 3V LEDs). Otherwise the mtn max buck driver would be a very reasonable option.

- you need a DTP PCB for each LED (for testing)

- you need a variable constant current source for testing (lab power supply)

- you need a way to electrically isolate the batteries from the center solder pad of the LED without negatively impacting the heatsinking (we used a battery carrier which I already had)

- you need a machinist who makes all the parts

- you need to anodize the outer part of the body and gold coat the contact plates

- you need a large UCLp lens (or other ar-coated lens)

  • you need somebody who has a lot of time and is skilled at building lights

You calculate youself what you would pay ;).

Thanks for the detailed answers.

Isn’t the large SST90 amp hungry at 3-4V? Big die?

Dang I hope you’ll be able to compare your monster to the GT when it is done, they are pretty close on a lot of sizes.

Yes the Luminus SST-90, CBT-90 and SBT-90 are 3V LEDs which can take high current (up to 18A). They are actually the LEDs which this driver was originally developed for, I think.

The CBT-140 and the new CFT-90 can take even higher amps, officially up to 28A, but probably even more.

Please do make a thread on your ideas!
I don’t understand how to “read” your graph yet…

My light was focused very carefully by running the LED at 30mA and shining it on a tree 30m away. The measuring was done in 12m distance (I originally thought 8m would be the minimum distance, so 12m sounded good enough).

Did you take a look at sma’s test which I linked to? He showed that 2.5m is the real minimum for the Olight SR-95 UT. What does your method tell you for that light?

I know that my conversion factor is not really based on anything, but 100m distance is something I have read in other places regarding the Maxabeam with it’s Xenon bulb.

The Miller, The plastic center is probably a plug covering the hole for the return wire in the original short arc Maxabeam