# 10w LED Circuitry

Hi,

I have bought some 10w LED's (9-12v), and I am looking to make a flashlight. But my question is a more general one.

I have been watching videos about how to put it together since I am still learning a lot.

The LED can handle 10w, and if I connect it to a 12v battery, that would, according to my calculations, need a 14.4 Ohm resistor to get 0,833A out of 12v. And 12v*0.833A = 10w ( I will keep the number about right, and not care that a battery puts out 12.something V for this example)

The thing that I hear in the videos again and again is that I need a driver or a current limiter for the LED.

I do not understand why I need that, could it not be done just with resisters, by calculating them based on the max V of the battery?

I hope someone can explain this to me.

High power LEDs can become lower resistance as they heat up, allowing more current to flow, leading to more heat and self destruction. Thus a current limit supply is used to prevent overheating and runaway failure.

Take a look thru these
http://www.digikey.com/product-search/en/integrated-circuits-ics/pmic-led-drivers/2556628?FV=15c0002%2C7b80018%2C1f140000%2Cfff40027%2Cfff802d4&mnonly=0&newproducts=0&ColumnSort=1845&page=1&stock=1&pbfree=0&rohs=0&k=led+driver&quantity=&ptm=0&fid=0&pageSize=25&pkeyword=led+driver
Different current options, package options, internal or external switch.
All pretty much act like switching power supply to keep your LED from burning up.

I've been thinking about putting out a board with a high power LED controller. I wonder if the same is already available on Amazon or e-bay or AllieExpress to not make it worth while.

Hi,

The thing that I hear in the videos again and again is that I need a driver or a current limiter for the LED.

The current limit resistor is basically what your 14R4 resistor is, it has limited the current that can flow through your LED to a level that means you only dissipate 10W in your LED.

However as CrossRoads has pointed out the LED characteristic changes and you need a more dynamic current limit circuit to keep the dissipation to 10W.

Usually any LED over 1W really needs a proper current limit supply, these devices also help optimize the power used to run the LED and hence power consumption/battery life etc.

Tom...
CrossRoads, if I had choice of yours (if you built one) or alibobo's LED driver, I'd be lined up with the rest at your front door.
Make it Arduino compatible with PWM input to fade, brilliant!!.(pun intended)

Actually your initial calculations are wrong - you are assuming you want to drop 12v across the resistor for 10w - right off the top, you need to look at what the forward voltage is for the LED's you are using and factor that into the equation. You don't include a data sheet link, but you do mention "(9-12v)" - exactly what is that number? Are the LEDs already configured with an internal controller of some sort? Another consideration in driving LEDs from a battery like in a flashlight is it is much better to use some sort of LED driver that uses PWM for the LED - dumping most of your energy from the battery into the resistor only serves to keep the flashlight warm (which might not be a bad idea in very cold weather to keep your hands warm??? ) Generally with battery powered stuff, you are interested in the best efficiency to get the most light with the least drain on the batteries.

Hey. Thank to everyone who replied.

gpsmikey:
The package says input: 9-12v DC.

Would it be better to use a LM317, or would that be exactly the same as using a resister?

bfkmnemonic:
Would it be better to use a LM317, or would that be exactly the same as using a resister?

There are LM317 constant current circuits on the web, but that won't work with a 12volt supply and 10w LEDs.
10watt LEDs are made with three ~3.3volt LEDs in series, and an LM317 CC circuit needs at least 3.25volt overhead.
And it's linear regulation, like a resistor, so things will get hot.
Post a link to the LED.
Leo..

I've been thinking about putting out a board with a high power LED controller. I wonder if the same is already available on Amazon or e-bay or AllieExpress to not make it worth while.

Probably not yet, but it would be exactly 24hrs before 10 different Chinese companies would be offering the same thing after you put yours on the market.

Does this look like a good part to make a general high wattage LED controller from?
http://www.digikey.com/product-detail/en/diodes-incorporated/AL8860WT-7/AL8860WT-7DICT-ND/6226981
Works with a wide range of voltages, only needs a few external components, and can switch 1A loads - or there's a differen package that can switch 1.5A loads.

There is a whole family of AL88xx.
These boards use the AL8805.

Leo..

See? Not even 24 hours to show that is wasn't worthwhile to come up with a driver board.

I ordered 10w led cob chips from ebay. Mine are square, with an aluminum back.

I got confused with the current limit resistor too. In the past, I just calculated ohm law and picked a resistor from my drawer to drive the standard 3mm and 5mm ledâ€™s.

The forward current, and spent alot of time reading alot of opinions from a whole bunch of people that had a whole lot to say about power conversion and building constant current power supplies, from scratch and knowing how to calculate windings in a coil to convert voltage to amps. I didnâ€™t want to know the chemical compound makeup of a LED and how many electrons it exactly needs to weigh on a molecular level. I need to know how much metal are in the bearings of the engine in my truck, in order to drive it.

If you would like to drive a 10 watt generic china led chip, without learning how to build a power supply you can just buy THIS:

Constant Current 10w LED Driver

Itâ€™s \$1.00. Save you the time in scavenging for tiny parts you do not have one at a time, and / or burning heat in inefficient voltage regulators.

I bought about 5 or 6 different types and styles of 10w LED chips, and tried them all, gauging using a photography light exposure meter to gauge the brightness. It seems they mostly all output about the same amount of light, and the power supply itself changes the brightness (or burns them out). I run about 56 volts through them in the headlights of my truck. I use them to retrofit halogen work lights. Theyâ€™re great because theyâ€™re cheap and I can afford to see what happens when they get too much or donâ€™t get enough power. Buy them in packs of 5 or 10, play with it and see what happens.

Aren't basically all those Meanwell and whatever DC-DC drivers "driver board"? I imagine e.g. an LDD-300 is basically such a chip in a case. Or is it a non-integrated circuit that does the same?

I did my last COB-lighting project with LDB-300s. Those are apparantly available nowhere right now. I wonder if I should look for a buck-boost IC and build my own one for the upcoming project.

Good price and indeed available. Unfortuantely, I cannot find any shipping information to Germany, at least not without creating an account.

DocStein99: Thank you ... Just my thoughts..

Thanks to everyone who replied.

One more question though, there is something in this area that I do not understand, and that I cannot find anywhere. (OK granted, there might be a million things, at least)

Does the current always have to be limited, or is it something special with LED's?

I do not understand how a servo works then, because it will draw the current it needs for the job, but what happens when it is not using the maximum possible watt? How does it limit it self, and why do a LED not work like this?

With a servo motor (or any dc electric motor including stepper) the power to the motor will remain in brake and actually consume power, when motionless. It will actually get hot and over heat if too much current is delivered. Efficient motor servo drivers reduce current when motor is idle since it does not need to consume as much, as when in motion. If you have the time and some spare junk motors, try over powering them to see what happens (thats what I did). Too much heat will reduce the strength of the motor. The motor's power driver actually limits the amount of current delivered to the motor.

How a servo works - there is active circuitry that moves the servo in response to the pulse stream it receives:

LED does not have any circuitry. It allows current to flow once the voltage exceeds the diodeâ€™s Vf, and the current flow is unlimited if the surrounding circuitry does not limit it. If the voltage source can supply enough current, the LED will conduct it all until it burns up. As the LED heats up, its effective resistance drops and even more current can flow, helping lead to runaway thermal failure.

bfkmnemonic:
Does the current always have to be limited, or is it something special with LED's?

I do not understand how a servo works then, because it will draw the current it needs for the job, but what happens when it is not using the maximum possible watt? How does it limit it self, and why do a LED not work like this?

Just as Crossroads says, a servo motor has active circuitry that applied power to the motor using a specific control algorithm.

A motor is certainly among the class of items that needs care when powering though. A spinning motor produces voltage through generator action, and that voltage opposes the applied voltage and reduces the current substantially compared to what you would expect if you just measured its static DC resistance and used Ohm's law. However, there are still dangerous conditions. When starting a motor, it isn't spinning (obviously) so there's no generator action to oppose the applied voltage, so a large inrush of current is pulled that depends on the static DC resistance and winding inductance, and the current reduces as the motor winds up to speed. This can also happen if something jams and stalls the motor. No rotation -> no generator action -> huge current draw.

Jiggy-Ninja:
This can also happen if something jams and stalls the motor. No rotation -> no generator action -> huge current draw.

It's pretty much a short circuit at that point. Whatever the weakest link on it will burn up just like a fuse.

DocStein - The price is right but I don't see a pwm input on these - are they pwm controllable? Thx...