need input for a 24v led light

My uncle is wanting me to make a premade 12v LED light work on 24v for his boat.  The light we are using is a Grote 60561 http://www.grote.com/product.php?product_number=60561, now the issue I am dealing with is I have no Idea what this light is constructed with, all the components are sealed.  I started safe and put a 1kohm resistor in and it barely lit the led, went down, and down, and I am currently using a 470ohm 1/2watt resistor.  With the 470 and 24 v The light output is really close to what it looks like with the standard 12v, but the resistor gets pretty hot.  I was thinking about trying the next ohm range down to see if I can get the led brighter,  but stepping up on a higher watt rating to hopefully run the resistor cooler.  Any ideas or suggestions would be greatly appreciated.  Thank you

hi, so the specs show 14 volts @ .2 amps...that equals a 70 ohm load, drawing 2.8 watts. it probably draws a little less current at 12V, but use the specs given. math time...... (24V - 14V)/.2A = 50 ohms. .2A^2 * 50ohms = 2 watts ..use 2 times plus for safety factor equals minimum 4 watts. so select a 50 ohm 5 watt resistor. now does the boat run at more than 24V nominal when the light is to be used? that will change the 24V to a higher value, which will change the resistor value (higher value). hope this helps, mark

You could use a current limiter. A 3-pin LM317 adjustable voltage regulator with a 1/2 Watt 5.6 Ohm resistor limits the current to 0.21A - ie, I-limit = 1.2V / R. The 317 will dissipate (say 14V x .21A =) ~3W so the LM317T (TO-220 package; 15W) is probably best. [ The 7.5W TO-202 LM317M also; but the small TO-39 is only 2W and therefore needs a heatsink, and the TO-3 (20W) is a PITA to mount. ] The TO-220 package as shown in the fig below is common and easy to use - you can solder the resistor between the center pin (the 117's output) and the outer "adjust" pin. Note that 24V input is to the outer #2 INPUT pin, and the output to the LED is taken from the other outer #1 ADJUST pin - NOT the 317's middle #3 OUTPUT pin/leg!    It's a simple and cheap circuit ($2?) and its has its own thermal limiting and over-current protection. It's as simple as this diagram from www.otherpower.com.... ... as found on Home Built LED Lighting. NOTE that I have just emailed otherpower.com seeking permission to use the diagram above. Note too that any the12volt.com automated Copyright watermark may be invalid. [ FYI: I spotted the above diagram and "just had to" use it. I was intending to link the normal & common Databook version else draw my own, but IMO the Otherpower.com's pic was just so good... And incidentally, I tried not to get distracted, but I kept finding great info on their site... ] You may notice the 1.25V versus 1.20V for my "I = 1.2V/R". That's a specification issue - the typical versus minimum "Reference Voltage". The original Databooks usually use the minimum, ie 1.2V. (Hence the 120Ù "reference" resistor value for a minimum load of 10mA for a stable output.) But that should trivial, especially when rounding to common preferred resistor values. And the max current isn't that critical anyhow - just decide if you want a bit above the LED's spec'd 200mA or a bit below. The LED may draw less anyhow - the programmed current limit is merely the upper limit - the circuit/LED can draw less current. Not that I wanted to present a Design-101 class... But, that's me!

Thanks guys for the info, I will play with it more today. I just came back in the shop and I have had the LED running on batteries since yesterday morning about 9am, it is still running with the 470 ohm resistor as i mentioned in my post. resistor is still hot, as i expected it to be. But i would feel much better going a safer route. As far as the voltage fluctuating, I am not positive but would assume the voltage will go slightly higher when under way, with alternators running. This is a 62' ocean sport fisher, when docked it is connected to shore power so voltage should be steady then.

If it's a 470R resistor, then the LED's 0.2A spec must be wrong - I think they mean 0.02A = 20mA. ( V=IR => .2A x 470R = 94V across the resistor, and nearly 20 Watts. 20mA means 9.6V at 0.2W. 12V across 470R means 25mA @ .3W, etc.) But if 470R is ok, I'd suggest a 1W or maybe even 2W or 5W version as their larger size dissipates heat much better. And you should be designing for a 14V drop - not that it makes much difference. (A 12V system is normally ~14 to 14.5V with alternator-charging, ie, 28 - 29V in a 24V system. Note that an automotive "12V" LED or load is designed to handle 14.5V etc.) PS - yet again, incorrect info published on manufacturer's sites - incorrect solutions from Rostra, ...., ...., simple LED specs... Yep, devolution is evolving as predicted.

Copyright update: With a big thanks to otherpower.com Dan Fink, the Executive Director of Otherpower, has kindly responded and given permission for the use of their image in my last reply. Alas I can't back edit my reply, hence this reply stating the good news. Otherpower's image/diagram not only saved me concocting something similar (though probably inferior), but it was also by far the best and most suited to my intent. (And I looked at lots of images!) And it was a double-bonus - that image led me to the otherpower.com site which impressed me greatly. For anyone else into alternative power, and batteries, etc, may I suggest a visit... Though still as yet only a cursory glance, I am keen to peruse their site in greater detail. (What a pity it's 3:30AM here now! And with no sleep last night, I really should get to bed. Now. Else maybe soon...) I suspect I have found another - but rare - reference site. Yay! Yep - it's been a good day.