3 or 4 Cree XR-E R2 MR11 kit??????

[frozenwarp]

Hi all,

I'm about to order my LED'a for my custom cannister light, first off some background info so u know what i'm working with.
14.8V 4000mah LiPo battery
Custom Aluminium Cannister and light head (so no fixed size im up for idea's)
pretty much the rest is Salvo parts (glands, latches etc)

NOW my question,
I'm planning on ordering a Cutter MR11 Cree XR-E Kit but i cant decide whether to get the 4 LED kit or the 3 LED Kit, Even though the optics are apparently 4deg for both i would guess due to there being more room the beam may be a little better from the 3 LED one compared to more light but worse beam from the 4 LED kit

Just looking for thoughts, comments and ideas

I would like to keep the whole thing fairly small but if u have ideas that are way better but bigger please still feel free to share as u may convince me to go larger


Thanks in advance,
Bradley

 

[MrSpock]

It all depends on the driver. R2s require 1000 mA at 3.5-3.7 V each. If you are going to use a buck driver, it will take approx 2 V for itself. Now do the math:
3.5 x 4 + 2 = 16 <- Marginally enough to power the head, since actual voltage from the battery will be around 16V fully charged.
3.5 x 3 + 2 = 12.5 <- More than enough power = less efficiency.

I actually just built the light with almost the same configuration - I used DX reflectors and lenses glued to the Modamag heat sink. I wound up with DIY Li-Ion 14.8 battery (8 x 18650, 2P4S) and 4 Cree R2 connected in series, driven by LED Supply 1000 ma buck driver. The setup is extremely bright and focused on the land, but I haven't got a chance to test it underwater yet.

 

[Packhorse]

I hear the 4 deg cutter optics are no where near 4 degrees.

Having said that all things being equal I'd go for more LED's. If you drive them at a lower current they become more efficient.
BUt for what its worth I think the best set up for MR11 is a single P7 (in my experiance) or maybe a MC-E.

 

[MrSpock]

Having said that all things being equal I'd go for more LED's. If you drive them at a lower current they become more efficient.

That's not entirely true. having the same voltage and lover current will not make them more efficient. It will make LEDs dimmer. Most efficient setup with 4 R2/Q5 connected in series and 1000mA buck driver should have 16.8 V of input voltage.

As for use of P7 or MC-E - they require 1400 mA @ 3.4 - 3.9 V
http://www.cree.com/products/pdf/XLampMC-E.pdf
I quite not figured out how to drive 3 of them with a reliable driver.

 

[Packhorse]

Well yes. You do need to take into account which driver you are using and of course which battery.

If you are running a array of 3 LEDs and another of 4 LED's and the power consumption is the same for both arrays the array with 4 LED's will be brighter. There for it must be more efficient.
This can plainly be seen by looking at any current vs lumen chart for LED's.
Take into account the lower Vf to run the LED at a lower current and it gets even more efficient.

For example

If you run 4 Q5's at 750ma you get a total of 820 lumen ( 3.6v X 750ma 10,8 watt)
If you run 3 Q5's at 1000ma you get a total of 735 lumen. (3.7v X 1000ma 11.1watt)

 

[MrSpock]

Packhorse, I don't understand your logic here:

the lower Vf to run the LED at a lower current and it gets even more efficient.

There is a direct correlation between current and LED light output. Voltage in this case is not that important. Take a look at: http://www.cree.com/products/pdf/XLamp7090XR-E.pdf page 8 - "Relative Flux vs. Current" chart. That's why you need to use a driver. And BTW: kaidoman data isn't really scientific. ;-)

As for the efficiency as a relationship between power spent (current multiply by voltage) to produced light - it's not what people looking for in a dive light, isn't it? Light output and reliability are much more important factors IMO.

To get the most efficient schema for XR-E using a buck driver you need to use the following formula:

N x 3.7 + K = Vb; where N - number of LEDs, Vb - nominal voltage of the battery pack, and K is a drive coefficient, usually 2 V for DC buck regulators.
The current depend on the driver type, not the applied forward voltage.

For example, if you use the same 1000 mA driver, same 16.8V battery pack, you'll have 25% less light output if you use 3 LEDs, compared to 4 LEDs.

Most people seems to use 1000 mA for XR-E attached to a heatsink. This one : 03023-D-N-1000 is among most popular (and reliable) ones.

 

[Hallmac]

There is a direct correlation between current and LED light output. Voltage in this case is not that important. Take a look at: http://www.cree.com/products/pdf/XLamp7090XR-E.pdf page 8 - "Relative Flux vs. Current" chart.)

Voltage is important... that is half the equation for selecting the proper driver.

As for the efficiency as a relationship between power spent (current multiply by voltage) to produced light - it's not what people looking for in a dive light, isn't it? Light output and reliability are much more important factors IMO.

In the LED world efficiency is discussed based upon how much energy is lost as heat. All conversions, so far anyway, lose some of the stored energy in the form of heat. The closer to the required Vf of the emitter array the less energy the driver has to burn off. For the Cree a Vf of 3.6 to 3.7 volts is necessary per LED. Burn time and light output are directly related to the current applied. As you increase the current, LEDs heat up. For a truly efficient LED product you would need a power source that provided the correct Vf, and only enough current to create light without increasing the temp of the emitter. While efficient in power consumption LEDs would not be very practical in terms of output. Since we divers have a fantastic avenue to deal with heat we can get away with pushing up the amps to get the illumination we are striving for. Thus, we waste energy in the form of heat.

I think you are both correct in your discussion, Packhorse is talking about driver and electrical efficiency and you are speaking of storage and efficient product use based upon what is currently being done and practiced.

By the way, Packhorse has a very effective multi die Cree dive light in use. I know for a fact he has spent considerable hours playing with LED lighting for underwater use.

 

[MrSpock]

By the way, Packhorse has a very effective multi die Cree dive light in use. I know for a fact he has spent considerable hours playing with LED lighting for underwater use.

I know! I used some of his knowledge in my project. Respect and kudos for sharing! :japanese:

 

[MrSpock]

Voltage is important... that is half the equation for selecting the proper driver.

Of course it is! But what I'm saying it's not important in case of using a buck driver, since the Vf on every LED will be maintained automatically because of the current regulated by the driver (assuming the driver input voltage is in specs). The Ohm's law is still the same: I = V/R.
Here are some interesting reading bits: http://www.leddynamics.com/LuxDrive/datasheets/3021-BuckPuck.pdf

I agree on the heat issues, but it's not just an LED problem. Heat dissipation is an issue in every analog electronic project. If you built custom amplifiers, you know what I mean ;-)

 

[Packhorse]

I know KD info is not really all that accurate but the diminishing returns in terms of lumen for more power are correct. That is all I was really trying to point out.

I think we can both agree that the right choice in driver/ battery voltage is probably the main consideration in efficiency. But running the LEDs at a lower current does increase efficiency and also the life span of the LED. But lets face it the life span of the LED is not really the biggest concern is it.