Newbie question: how to drive this Lux III?

Hi Guys!

I've just got my first Luxeon LED in the mail. I'm going to mod it into a WolfLight host. I'm almost ready (got the proper heatsink, etc.).

I'm just not sure, which is the best way to drive this LED. It's a white SX1K binned Luxeon III.

So my question is: which would be the best setup?
a.) 3xAAA NiMH DD (most likely 800mAh Sanyo, I don't have them yet) or
b.) 3xAAA Alkaline DD or
c.) 3xAAA Alkaline resistored?

How much current would the LED get with each setup?

Thanks in advance!

Funny, I did this mod just last night. My first mod!
I cannibalized my FT-3C LuxIII so I'm doing direct
drive because I know what THIS led requires for
Vf . Well actually it already was directly driven and
I don't have any resistors yet so I'll probably put one
in and limit my mAs too 500 for respectably
balanced run/brightness levels.

A great led calculator can be downloaded from here
somewhere, called led_pro. Gives you runtime, current
values etc...

Thanks! Hey! This is my first mod, too! (Actually, first _Luxeon_ mod: I've already modded a Mag3D to 12NiMH, 35W Osram IRC MR16... quite bright /ubbthreads/images/graemlins/smile.gif )

What was your LED's bin code? Mine is xxxK, so I guess it's ok to DD with 3xAAA NiMH...

Well the bin code isn't on the back of my star for
some reason but the Vf at 1.2mA is 3.52V so I'm
guessing xxxJ bin.

My freshly charged NiMH*3 can have a total of
4.27V so to answer your question about the xxxK
bin it really depends on your individual LuxIII, you
need check the amount of current draw. Your
probably safe if you let the freshly charged cells
sit a day or two before using them so they can
discharge somewhat.

[ QUOTE ]
hacky said:
Thanks! Hey! This is my first mod, too! (Actually, first _Luxeon_ mod: I've already modded a Mag3D to 12NiMH, 35W Osram IRC MR16... quite bright /ubbthreads/images/graemlins/smile.gif )

What was your LED's bin code? Mine is xxxK, so I guess it's ok to DD with 3xAAA NiMH...

[/ QUOTE ]

Ok,
I've been testing my mod with
new Energizer AAA 850 mAH NiMH
cells I bought earlier today.

This LuxIII will only pull just
over 600mA on freshly charged
AAA cells.

I guess 3 AAAs can't deliver
enough current to damage the LED.
Not surprising I guess but I'm no
expert and I am just learning
as I go along.

Really simple mod, very cool.
I used a MiniMag reflector, nice
flood with a good hot spot. Next
I'll try one of the 17mm
reflectors for, hopefully better
throw.

Thanks again for the infos!

I've measuered about 600mA with 2 old Energizer 750mAH NiMH cells and an old NiCD cell (5 hours after charging).

I've get nearly the same reading (560mA) with slightly discharged alkalines.

I think I will direct drive it with 3 NiMH cells. 600mA sounds just about right.

I'm still looking for an appropriate reflector. I'll try to find some cheap light, and take the reflector out.

Hey, nice to see someone follows my path and uses the wolflight for modding.
As I said in the linked thread I use three sanyo nimh (800mah) in direct drive mode. This light is one of my edc-lights since I built it and it has never failed since then. Right now, it sees less use just because it's summer and it's getting dark late now. I'd like to see some pics of your assembly when you're finished. Please feel free to post them in the thread I started. It was meant to collect all mods that have been done to this light.

Chrisse

EDIT: I'm using a tv1j in it now and it seems o.k. I didn't measure the current going through, but the assembly doesn't get too hot when lit for a while so I think it's ok.

Hi Chrisse!

Your thread helped me a lot! Actually, I have a machined aluminium heatsink, which screws into the head. The heatsink has the Luxeon III Star glued on it (glued at the edges, heat conductive paste between the Star and the sink). I've choosen a Star, because this way I know the bin code for sure, and I didn't have heat conductive epoxy. I'll try to put some pictures up, but I don't have a camera, so it will take some time.

May I ask how much current does your LED get from the DD 3xNiMH mod? Thanks!

Hacky, glad I could help!

I'm using a star myself, mainly for the same reasons and because it sits perfectly inside the copper cap I used as a heatsink. I guess your approach is much better but I have almost no machining equipment or skills and always like to come up with simple and easy Ideas.
I will try to measure the current but I have to recharge the cells first and then try to remember how to measure current. /ubbthreads/images/graemlins/thinking.gif
Will try this as soon as I get home today.

Did it, the cells have about ten to twenty minutes runtime on them and my dmm reads about 0.66A.
Well, this probably means up to 750ma from fresh cells but I will test this in some days.

Chrisse

Thanks for measuring the current! /ubbthreads/images/graemlins/thumbsup.gif

Something came into my mind: IIRC, you have a xxxJ binned Luxeon III in your mod and get a maximum 750mA reading (decreasing as the cells discharge). That sound very good, you are driving the Luxeon at about specs for a long time.

Now my Luxeon is xxxK, thus has a higher Vf. This probaby means that I'll get an even lower maximum current reading (as I got a few posts above, but those were old cells, so I don't know if it counts), and much lower with the cells discharging. So the LED will be much underdriven. Now that's something I don't want to happen! /ubbthreads/images/graemlins/naughty.gif

Because of this, I haven't bought the NiMH cells yet. I wanted to ask your opinion first. So what do you think? Will the 3xNiMH setup underdrive my LED? You know... these cells are not cheap /ubbthreads/images/graemlins/broke.gif, so I'd like to go for sure before I buy them. Thanks!

[ QUOTE ]
hacky said:
Thanks for measuring the current! /ubbthreads/images/graemlins/thumbsup.gif

Something came into my mind: IIRC, you have a xxxJ binned Luxeon III in your mod and get a maximum 750mA reading (decreasing as the cells discharge). That sound very good, you are driving the Luxeon at about specs for a long time.

[/ QUOTE ]

Actually, the 750mah wasn't measured but only estimated, in fact the maximum reading might be lower.
[ QUOTE ]

Now my Luxeon is xxxK, thus has a higher Vf. This probaby means that I'll get an even lower maximum current reading (as I got a few posts above, but those were old cells, so I don't know if it counts), and much lower with the cells discharging. So the LED will be much underdriven. Now that's something I don't want to happen! /ubbthreads/images/graemlins/naughty.gif

Because of this, I haven't bought the NiMH cells yet. I wanted to ask your opinion first. So what do you think? Will the 3xNiMH setup underdrive my LED? You know... these cells are not cheap /ubbthreads/images/graemlins/broke.gif, so I'd like to go for sure before I buy them. Thanks!

[/ QUOTE ]

Actually the first lux I had in there was s??k, but I messed up the contacts and had to redo it with a new one. I don't know about the current that was going through it, but to my eyes it wasn't much dimmer than the one I now have. For me I'd go for it. You can still use those nimh's in a remote control or something similar if you're not pleased with it. Both versions I made where brighter on alkalines but since the heatsinking in my light is not perfect I prefer the slightly dimmer but safer nimh's.

Chrisse

(OFF: I've managed to fit in a trimmed down MiniMag reflector to the host. Now it has a nice hotspot (no too large, tought), and a huge sidespill! I like it! I'll try to take some pictures sooner or later!)

I have one more question: I've just found out, that a 16500 Li-Ion (or a 18500, it I can get one) would fit nicely (length-wise) into the WolfLight host.

Most likely I can get a 16500 3.7V 1000mAh Li-Ion cell. I think it would be perfect for my mod, because:
- it provides more voltage then 3xAAA NiMHs, so I can drive my xxxK binned Luxeon III harder if I want to (and I can add a resistor anytime, if I think DD is too hard)
- it has 1000mAh capacity, which is more then the 800mAh of the NiMHs (and if I somehow manage to get the fatter Li-Ion cell, then it's even more: 1300mAh).

What do you think? Is this a good idea? Any idea of how hard would the fully charged (4.2V) Li-Ion cell direct drive my LuxIII?

Thanks in advance for your help! You've helped me A LOT already! /ubbthreads/images/graemlins/thumbsup.gif CPF is great!

(I'm aware of the safety precautions regarding the Li-Ion chemistry...)

Hi hacky, if my understanding is not completley wrong, the current of the three aaa's is added together so it will give you 2400mah?
Then I'd be worried about heatsinking, driving the lux3 with 4,2 Volt from the start, for me It would mean too much effort, high risk, little to be gained. I'd stick with the nimh's.

Chrisse

[ QUOTE ]
chrisse242 said:
Hi hacky, if my understanding is not completley wrong, the current of the three aaa's is added together so it will give you 2400mah?
Then I'd be worried about heatsinking, driving the lux3 with 4,2 Volt from the start, for me It would mean too much effort, high risk, little to be gained. I'd stick with the nimh's.

Chrisse

[/ QUOTE ]

Ok, let my clarify:
Battery config one: 3pcs of 800mAh Sanyo AAA NiMH in series, using the original WolfLight battery container. If my understaning is corrent, this is the same battery config as you use. This would give 800mAh capacity and 3.6V nominal voltage, because the cells are wired in series. Also I think - because my emitter has a higher Vf than yours - this would drive my LED at around 600mA in the beginning _and gradually decreasing_. I would like to drive it a tad bit harder (as I think the heatsinking is enough).

Battery config two: One 16500 (16x50mm) Li-Ion cell, nominal voltage: 3.7V, capacity: 1000mAh (higher than the 800mAh above). So the original battery container would not be used. This would give 4.2V when taken off the charger. I don't know if it's too high to Direct Drive, but if it IS, then I could use an appropriate resistor. But at least, this config allows me to drive the LED harder and offers a little more run-time.

Battery config three: same as bty config two, but with a 18500 Li-Ion cell (18x50mm), which gives 1300mA. This would be even better, but it's unsure, whether I can buy this kind of cell here, or not. The other one is listed in a local webshop, so I can get that one for sure.

I just feel a little unexperienced /ubbthreads/images/graemlins/help.gif to choose the latter battery configs, as this is my first luxeon mod. That's why I'm asking you (chrisse, and other experienced modders of course), to gain a little confidence /ubbthreads/images/graemlins/wink.gif.

[ QUOTE ]
hacky said:

Ok, let my clarify:
Battery config one: 3pcs of 800mAh Sanyo AAA NiMH in series, using the original WolfLight battery container. If my understaning is corrent, this is the same battery config as you use. This would give 800mAh capacity and 3.6V nominal voltage, because the cells are wired in series. Also I think - because my emitter has a higher Vf than yours - this would drive my LED at around 600mA in the beginning _and gradually decreasing_. I would like to drive it a tad bit harder (as I think the heatsinking is enough).



[/ QUOTE ]

/ubbthreads/images/graemlins/ohgeez.gif /ubbthreads/images/graemlins/ohgeez.gif /ubbthreads/images/graemlins/ohgeez.gif

Well now and then I get to see that I'm still a novice myself. I always thought the capacity of cells in Series adds up. Shouldn't have slept in physics class when I was in school. /ubbthreads/images/graemlins/twakfl.gif
Now knowig this, yes capacity is increased with those li-ions but still I'd be worried about heatsinking and "overdriving" a k-binned Lux with 4.2V. It has been done with pilas and bare cells before so you might be succesfull.
A resistor sure would help a lot and let you be on the safe side, but then will eat up some of the gained capacity.
So all I can say is try it, if you absolutely want to.
Myself, I'm happy with brightness and runtime with three nimh's.

Chrisse

Well, I think I will try a Li-Ion solution. I'll post the very final results (hopefully with pictures) to your original thread! Thanks a lot!

I have to thank you, actually. Without you I'd still be telling everybody I've got 2400mah in this little light. /ubbthreads/images/graemlins/hahaha.gif /ubbthreads/images/graemlins/hahaha.gif /ubbthreads/images/graemlins/hahaha.gif
Can't wait to see the pics of your heatsink.

Chrisse

AAA NiMH batteries will put out 1A or more
of power, I made a mistake saying otherwise.
I made a tube of paper, slid that in the light and
tested the current without the series connector.
The Luxeon was drawing over an amp of power.
What this means is that the battery holder is a
major source of resistance.

Li-Ions would be cool but I don't know anything
about them, you would almost certainly need
added resistance.

Also, the SO17XA reflector is brighter than the
Minimag reflector. The hot spot is bigger but the
throw is greater. The sidespill is only slightly
smaller it's less bright but still useful.


[ QUOTE ]
roguesoul said:
Ok,
I've been testing my mod with
new Energizer AAA 850 mAH NiMH
cells I bought earlier today.

This LuxIII will only pull just
over 600mA on freshly charged
AAA cells.

I guess 3 AAAs can't deliver
enough current to damage the LED.
Not surprising I guess but I'm no
expert and I am just learning
as I go along.

Really simple mod, very cool.
I used a MiniMag reflector, nice
flood with a good hot spot. Next
I'll try one of the 17mm
reflectors for, hopefully better
throw
. [/qu
ote]

it bugs me that batteries are rated in mAH which is very confusing measurement of capacity since it's dependent on voltage to get Energy = WH or mWH... life would be a lot easier if batteries were rated in a rating of storage rather than of power. For example... an 850mAH AAA has 850mA • 1.5V = 1.27mWH... the total amt of energy you can expect to get from the battery... if you draw the power too fast, more of that energy is 'spent' on internal resistance but you can easily calculate for those loses once you do a test or three.

(3) 1.2V AAAs is only going to be 3.6.. so depending on your Vf, you are just about in the 'sweet spot' for max suggested 700mA... in testing you got 600 which falls in line with that thinking, so either you are drawing enough current to raise the internal battery voltage drop, or the Vf is high enough that 600 is just the current through that particular emitter at 3.6(ish) volts.

Since once converted to actual energy (WH) you just add them up and get 3.85 WH total capacity. now it's really easy to estimate runtimes... 600mA • 3.6V = 2.16W.. if no driver no driver loss... and with NiMH batts.. you can usually draw about an amp before you start really loosing power internal... so you can expect to get really close to the full capacity out of the battery.

So.. 3.85WH ÷ 2.16 W = 1.77Hrs... not shabby at all... you should be getting about 50 lumens out of a Lux III at that power level (estimated... i didn't look it up on the chart)...

I was trying to drive a LuxIII from (3) AAAs and learned quickly it was very difficult to get it to 3W, so i went to (4) so i'd be able to run it max power. Even so, i'll likely settle in at 2.5 or so to balance power, brightness, battery life and heat.

Batt option two will give you 3.7V • 1000mA = 3.7WH ... less than three 850mA AAAs (this simple math is why i ended up going with AAAs vs LiON batt in my design).

Batt option three would give you slightly more: 1300mA • 3.6V = 4.6WH... However in either of 2 or 3 until you run it you won't know the voltage drop and which would actually run longer or brighter, it's too close to call.

In any of these scenarios, you won't get max brightness from lux3, but will get a very decent runtime... bumping up the power from about 2 to about 3 will reduce the runtime by 1/2 not 1/3 as could be expected, that much more power will be lost 'heating the battery'

In any event, fun project keep us posted.