15,000 Lumen Bridgelux C9000 Light Engine

wow, would buy custom made with this inside :o

Search Ebay for “600W DC-DC” There are some $20 DC-DC converter modules that are already constant current capable. They can output up to 80V (drive two C9000’s in series?) and they claim to by 95% efficient. I have a couple on order… http://www.ebay.com/itm/DC-DC-600W-10-60V-to-12-80V-Boost-Converter-Step-up-Module-Power-Supply-/170893208817?pt=LH_DefaultDomain_0&hash=item27ca076cf1

:heart_eyes:

I kludged up an external PWM dimmer onto the constant current driver… it failed miserably. Any dimming caused the LED to go into puke-flicker mode. Turns out that the stock program in the dimmer is PWMing at 100 Hz… way too slow to maintain a stable output. The constant current driver has a 1000 uF output cap, which 5A of current significantly drains between 10 millisecond PWM pulses.

I need to reprogram that dimmer chip for something like 10 kHz PWM…

I think that spec is only for an SST-90. Perhaps if you could get 15k lumens from an SST-90 sized emitter you’d be able to get that kind of throw, but the array is freaking huge. I expect closer to 100kcd, given that the bare emitter already gives off 15kcd.

I tweaked the dimmer firmware to do 32 kHz PWM. The driver behaves a bit better, but still messes up below around 35%. I tried adding a bigger cap (another 1000 uF) across the driver output… it just made things worse. I need to see what happens if I shorten and clean up all the external wiring… and maybe reduce the output cap on the driver.

I did manage to blow the switching FET in the driver by pulling around 220 watts out of it (maybe 19,250 lumens). I replaced it with an IRF3036 (2 milliohm, 190 amp rating, the original FET was 10 milliohm, 75 amp). I was expecting to see some decent improvement in the driver efficiency, but it was minimal.

Wowzers! That is sweet!

The BHR’s (Big Honkin’ Reflectors) should be in tomorrow…

I’m experiencing a similar problem with a AL8805 1A buck driver chip. PWM below 25-30 and it goes irregular blinky on me. I upped the PWM and still about the same deal. I might try a DAC output of my PIC, and see if that helps.

Maybe my problem is that it goes into a discontinuous mode? The inductor is only a 10uH one, so next up will be to try a higher value inductor. That, or my data sheet interpretation is messed up, it wouldn’t be the first time :slight_smile: .

EDIT:Heh, discontinuous mode not discontinuance

Just warn me when you turn this thing on and start aiming it at things. I want to pull the shades down and put on sunglasses. Ought to be able to see it here in Tyler, when you fire it up, LOL.

I figure you will have about 5 minutes before the chopper arrives with their little spotlight, to search for you.Wink

Chuck Norris’s edc! :bigsmile:

Meh, mine’s brighter than theirs… 8) The beam has enough photon force to push them away…

:bigsmile: Just don’t turn it on within 20 miles of an airport. The helicopter that comes for you will have Bruce Willis as well as Chuck Norris in it.

I got in the Big Honkin’ Reflectors (ok, 90mm is not that big) from Illumination Machines. Whacked 5mm of the bottom of one to get a 34mm hole. It cranks out 70,000 lux at 1 meter (around 10 degree beam width). The beam is as smooth as a baby’s behind. I’ll try and get some pics up soon.

Old-Lumens 1.3 mile SST-90 Aspheric Potato Masher Thrower does 39,000 lux at 1 meter. A Jacob A60 does around 35,000 lux at 1 meter. These lights have a considerably tighter beam.

Holy mother. This is very badass.

What kind of portable power source is used to power a light of this magnitude?

It can run on 12V to 30V DC. At 12V it pulls around 15 amps. I was going to (and still might) power it with a 4S A123 20Ah LiFePO4 pack, but A123 is in a bit of a financial meltdown at the moment and their cells may no be viable in the long term. Or any 4S-6S LiPo pack. Or a car.

Here’s a pic of the reflector sitting on the LED:

Here it is lighting up the sphere ’o many mysteries. What is not obvious is that the room is also lit by 10,000 lumens of overhead lighting. For some reason, those lumens take a back seat to the big guy:

Beam shot with reflector. Ceiling is 8 feet from the top of the reflector. Those PAR20 fixtures are 3 feet apart in the Y axis and 4 feet in the X axis.

Are you sure your sphere's not about to catch fire? It also appears that you have an interesting collection of electrical wiring and gadgetry there. Looks like fun.

Doing a little trig on the beamshot image looks like the beam width (full width) out of the reflector is around 16 degrees (with an SST-90 it is supposed to be 10 degrees). Works out to around a .25 lux spot around 600 feet wide at 2000 feet.

Funny you should ask… the lights sit on a couple of blocks of polyethylene foam to get them up to the sphere port. I started a run (with a different heatsink) and after around 5 minutes of observation went to work on the computer in the other room. Sometime later I heard a loud BANG!

What happened was one of the cheap Chinese clip leads to the fan decided to uncrimp the connection between the wire and the clip (hint, only buy leads that are soldered). LED lost cooling, temp rose to over 100C. The heatsink and LED then did their best imitation of the China Syndrome and melted their way through the foam blocks (perfect, form fitting rectangular hole). About 3/4 the way through the foam, one of the power connections to the LED unsoldered itself. The LED driver wanted to keep 5A to the LED, but saw no current flowing so it kept increasing the output voltage. A couple milliseconds later, the voltage exceeded the 35V output capacitor rating and the cap blew with a satisfying BANG (I’ve since added a diode that keeps the output voltage below 35V.