LED's for Stage Props

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
Ok Gurus,

I would like to start using LED's in stage props. So the question is can I use this :

http://www.fasttech.com/product/1341207

I liked the battery bank because it has multiple power outputs so I could possibly test various configurations.

to drive this :

http://www.fasttech.com/products/1574600

It would be a simple circuit with a button switch between the LED and the power source.

I was hoping that I would not have to go out and buy a multimeter and become a low voltage tech to get this done.

Regards,
Dead Goat
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
That could work with the addition of a resistor, easily calculated and sourced. It would work best on the 5V/2A line, which would limit you to about 66% of the LED's rated drive. That's still pretty bright, might be fine depending on what you are looking for.

Running on the 3.7V output, you should be able to get higher current and better efficiency, but it's tricky to get the right resistor and the light output dims as the batteries discharge.

With the addition of a buck-type switching regulator, you should be able to drive the LED to the full 3A off the 9V/1.5A output. If you want to go that route, just say so and someone will help you find an appropriate driver.

I would definitely NOT attempt to hook the LED directly to any of the outputs of that battery pack, unless someone with specific knowledge of that pack assures you it's okay.
 

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
I see, how much does the circuit change if we are talking about a momentary flash of light for 5 seconds.
The effect would need to be used a few times in a show so the LED would need to survive about 500 cycles.

I am taking a step back for education here to learn about calculating for a resistor.

Ohm's law

V/R = I

Assuming a 2000lm LED with a forward current of 500ma

33.7v / 100 ohm = 337ma

So if the forward current on an LED is 500ma would a 100ohm resistor be appropriate?
Does the ratio of actual current to forward current determine how hard an LED is driven?
 
Last edited:

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
So it seems that a resistor in the circuit would either burn up or waste the energy to heat loss. So a buck-boost driver it is.

I am definitely learning more than I want to about this. May I should consider hot wiring the switch on a tiny monster.
 

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
So based on what I know now I am thinking about 2 LEDs in parallel with a DC to DC buck driver.

Mini DC-DC Converter Voltage Stabilizer/Regulator Module Max 3A draw
http://www.fasttech.com/products/1253701

Cree XM-L2 T5-5B1 10W 995LM 4000-4200K Neutral White LED 1000ma
http://www.fasttech.com/products/1425003

Samsung ICR18650, 4.35V 2800mAh Protected Rechargeable Li-Ion Batteries (2-Pack)

If I wire 2 batteries in series the output would be approx. 7 volts to the buck, The voltage and current will be limited by the buck to 3.7v and 2000ma as measured with a multimeter (which I didn't want to buy but since the buck is adjusted with a POT is seems this is required.) Then the LED's follow the buck in parallel.

Just need a switch, and probably a heat sink.


Is this a better plan......feel free to jump in here guys.
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
You're not quite there yet. That regulator is a voltage regulator, not a current regulator. LEDs want to be driven with constant current, so you want a constant current regulator (which is exactly what any decent LED driver is). There are some available that are both constant voltage and constant current. These are often described as CVCC or CV-CC supplies.

These are better described as limited voltage, limited current supplies, because they have separate voltage and current limits and which ever one you hit first will engage. For example, say your load is a 1 ohm resistor. That means that for every volt on it, it will conduct 1 amp. Another way to look at it is that for every amp through it, it will have 1 volt across it. Now say you connect it to a CVCC supply set at 1V, 2A. When you connect it, the voltage and current will rise until one of the limits is hit. In this case you will hit 1V before you hit 2A. Thus the supply will regulate at 1V, and the output will be 1A. Now say your resistor changes a bit as it heats up (LEDs do this). Say the resistance changes to 0.9 ohms. The current will rise, but the voltage won't because it can't exceed the 1V setting of the regulator. Say the resistor changes a LOT, down to 0.4 ohms. At 1V, that would mean 2.5A. But the regulator will limit at 2A. Since the resistor will have only 0.8V across it at 2A, the output will have to drop to 0.8V. Now say the resistance continues to drop. The current will stay at 2A, and the voltage will drop.

This is how a CVCC supply works. It can transition smoothly from voltage regulator to current regulator, depending on the load and the settings.

A buck-type or linear regulator will have a voltage limit a little below the input voltage. This differential is sometimes referred to as the dropout voltage. ALL buck-type and linear regulators have this characteristic, though some are not well described. The value of the dropout voltage can range from a small fraction of a volt to several volts. In a regulator with an adjustable output voltage (really a voltage limit), the voltage limit can be adjusted in a range, typically 1.2V at the low end, and input voltage minus dropout voltage at the high end. In some cases the output voltage can be adjusted as low as 0V. Other low-end values are also possible.

Any type of regulator can be built with a current limit. This is not the same as a maximum current. A maximum current is how high you can go before the behavior becomes unpredictable. Exceeding max current can result in any of the following, and more: reg works fine at higher current; reg shuts down; reg reduces output voltage; smoke escapes and nothing ever works again. But when you hit a current limit, the behavior should be very predictable - the output voltage reduces and the current stays at the limit.

Here's an example of a regulator with both constant voltage and constant current adjustments: http://www.ebay.com/itm/5A-LED-Driv...209?pt=LH_DefaultDomain_0&hash=item5d4b488079

Okay, enough about regulators for the moment. At those power levels, you'll need a heatsink. Anything over 1-2W definitely needs a heatsink. If you're talking 10W per LED, you need substantial heatsink.

You also mention connecting the LEDs in parallel. When you do this, part of the current goes in one LED, and the rest goes in the other. If the LEDs are identical, then the current should split evenly, and if you have a 3A regulator, you'll get about 1.5A in each LED.

If you were to connect them in series, the full output current would go through each LED, but the voltage would be doubled.
 

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
DIWDiver,
Thank you for the time and effort. I am learning a lot.
I suppose emitters in series or parallel impacts this next question.
If I want to work with 18650's and the emitters are wired in series would the batteries need to be in series to hit the required voltage?
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
Yes, for a buck regulator the input needs to be higher than the output voltage, so if the output is 2x 3.3V, the input needs to be 6.6V or higher, which means the cells need to be in series.

There are boost or buck-boost regulators that could boost the voltage of a single cell (same as 2 cells in parallel) to drive the 2 LEDs in series, but they are much less common.
 

Dead Goat

Newly Enlightened
Joined
Oct 31, 2013
Messages
33
DIWdiver,

I like the regulator you posted above.
As I move forward I am quickly learning that optics may be my next area to tackle. No question yet on that.
I have noticed that some of the bike lights on fastech are just about what I am looking for, my concern with those is if they perform at stated outputs, and if they have enough heat sinking.
Just a thought flying through the brain. There will be many opportunities for various configurations so I will keep you posted. I think getting a build under my belt will mean more interesting applications vs. hacking finished goods.
 
Top