RR Rated M For Mature
HOME   rrTV-PHOTO   GALLERIES   MY GALLERY   HELP-FAQ
myHOME PM pmRR MEMBERS 378 ONLINE 18 EVENTS SEARCH REGISTER  START HERE
 
1 page524 viewsPOST REPLY
Midland Helicopters . HeliProz . Ron’s HeliProz South

.
.
Main Discussion > Electrical Enginering Help with Burned out LED's
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
Need someones help..


I have a night rig...450 Sport t-rex..
Picture of it here
http://rc.runryder.com/helicopter/t550953p1/

I installed a LED Array under the canopy...i pulled the LED from a small flashlight that ran on 3 1.5 Volt Batteries or 4.5 Volts

I wired this LED ARRAY str8 to the Lipo's Charging jack so basically its getting juice form a 1s lipo or 3.7 (4.2 Fully Charged)

When i first wire it it lights up GREAT...After a few flights the LEDS starting popping off like there burning out due to the motor drawing power (this is a guess) Once this happens they become nowhere near the brightness they were before this......
I replaced it once and the same thing happens...

So my question is what can i install to prevent this.. maybe a resistor? Diode?

I think what is happening is when you spool up this creates HVor LOW Voltage causing burnouts...So how can i prevent this, stabilize the voltage somehow...

Any Ideas will be apreciated!!!




Fly, Fly then Fly some more!!
11-03-2009 02:34 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
mikeflyz
Senior Heliman
Location: Westlake Village, CA

My Posts This: Topic  Forum
LEDs MUST be wired with a resistor. You'll have to find the current rating of the LED that you bought before you can find the right resistor value.

Use Ohm's Law (R=E(volts)/I(current rating of LED) to come up with your desired resistor.

Mike
MA Fury Extreme, JetCopter SX
11-03-2009 02:57 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
tadawson
Elite Veteran
Location: Lewisville, TX

My Posts This: Topic  Forum
He took the array from a flashlight - it should have already had whatever current limiting in it that it needed. I suspect, though, that since it was originally on "normal" batteries, that they may have been relying on voltage sag in those in their setup, and since the LiPo can output so much more current without voltage sag, that your array just plain can't take it. That, and the fact that if this was a cheap LED flashlight, most of those are crap to begin with, and were probably not designed for the duty cycle you are running either. You would need to measure the current drawn by the array, and then try to determine what LEDs are in it, and add dropping resistors until you get to the correct current draw . . . .

- Tim

The more I touch electrics, the more I grow to hate them . . . .
11-03-2009 03:17 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
u8mydog
New Heliman
Location: Ottawa, Ont

My Posts This: Topic  Forum
Are you sure that the batteries were in series in the flashlight or were they in parallel (1.5V)?

You could wire them back up in the flashlight and measure the voltage across the the leds. Then measure the current with a volt/amp meter.

Now take your lipo voltage minus the led voltage, then, divide the result by the current in amps. That will give you the required resistor in ohms. The wattage of the resistor will be the result voltage multiplied by the measured current. I would expect a 1/4W resistor would be plenty and easy to find.

msx/msr/b400/trex500/rap30/knight50
11-03-2009 03:20 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
GMPheli
Key Veteran
Location: W. Bridgewater, MA USA

My Posts This: Topic  Forum
I take it you have more than one led. Are you wiring them in series or parallel. You want parallel. In other words all the plusses hitched together, all the minusses hitched together and then each to the batt.
11-03-2009 03:46 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
dkshema
rrMaster
Location: Cedar Rapids, IA

My Posts This: Topic  Forum
OK kids, here goes. I looked up the specs on a high intensity, white LED, rated at 3.8 watts. I've got one of those flashlights, runs on 3 AAA cells, advertised as having a 3 watt LED.

The 3.8 watt, high-intensity white LED was rated to have a forward voltage of 3.2 typical with 350 ma running through it.

The forward voltage is the voltage you'll see dropped across the diode junction when it is "on".

A germanium diode has a forward voltage of about 0.3 volts. A silicon diode has a forward voltage of about 0.7 volts. LEDs get their color from their doping and construction, and the forward voltage varies with the color of the LED and a few other design goodies.

So, lets assume that you're running a white LED with a forward voltage of 3.2 volts, typical (could be as high as 3.8 volts, could be as low as 2.6 volts, depending upon the particular batch its die was processed in).

You're running it from a battery that has a fully charged voltage of 4.2 volts, a nominal discharge voltage of 3.7 volts, and a fully discharged voltage of 3.0 volts.

With a fully charged battery of 4.2 volts, the difference between the typical forward voltage of 3.2 volts and the fully charged pack is 1.0 volt (4.2 - 3.2).

As the pack runs down, it averages about 3.7 volts until it's flat. 3.7 - 3.2 gives you a half a volt of headroom between the voltage in and the forward ("ON" ) voltage of the LED.

As the pack continues to run down its voltage actually drops BELOW the required voltage to turn the LED ON (3.0 volts - 3.2 = -0.2 volts.

Luminosity is a function of the forward voltage and the current through the LED. As current goes down, so does light output.

If you don't limit the current THROUGH the LED, then there is a good chance you will burn it out.

Back to the LED. It has a forward voltage of 3.2 volts. Hooking it directly up to a power source that turns it on without a current limiting device is bad news. The diode turns on, conducts current, and without something to limit the amount of juice going through it, the diode rapidly heats up and the internal junction is destroyed.

Hook the LED up to your 4.2 volt fully charged pack. The difference in voltage between the pack and the forward voltage of the LED is 1.0 volt. The data sheet includes a chart showing the maximum permissible current through the LED vs ambient temperature. The LED has a maximum current rating of one amp, a minimum rating of 100 ma.

Assume you want to limit the current through the LED to 500 ma (1/2 amp) max.

Ohm's law, E = I x R (volts = current x resistance) is handy here. It can be rearranged so you can figure voltage, current, or resistance as the unknown value.

E = I X R

E/I = R

E/R = I

We know the battery is fully charged at 4.2 volts. We also know that out of that 4.2 volts, the LED will have typically 3.2 volts dropped across its leads simply due to the physics involved.

E, in the equation above is figured as battery voltage minus the LED forward voltage. 4.2 - 3.2 = 1.0 volt. To limit the current through the LED to 500 ma (0.5 amp), with a fully charged pack, solve:

E / 0.5 = R, where E is the battery voltage minus the LED forward voltage (1V).

1 volt / 0.5 amp = 2 ohms

You would need a 2 ohm resistor wired in series with the battery and the LED. What kind of power would the resistor need to handle?

P = V x I

2 ohms, 0.5 amps, 1 volt across the resistor.

V x I --> 1 volt x 0.5 amp, you'll want at least a 1/2 watt resistor. You would actually want at least a one watt resistor, to add some margin to add some life to the circuit.

-----

Great, you now have a healthy bright white LED with a 1/2 amp coursing through its veins. But the battery begins to discharge.

The resistance stays fixed at 2 ohms. The forward voltage of the LED remains relatively constant at 3.5 volts.

That 1 volt dropped across the resistor from a fully charged pack drops to zero as the battery reaches the diode's forward voltage of 3.2 volts. As the battery voltage drops toward the forward voltage of the LED, the current through the LED drops. At some point, it drops to zero. Before that happens, the LED has all but gone out.


In between, the LED gets dimmer and dimmer and...


And if you have no current limit resistor in the mix, the LEDs simply burn out.

-----

What you REALLY need is a constant current source to run the LED. You need a source whose voltage is at least 3.2 volts (to turn on the LED), but it needs to be a source the can continuously provide a CONSTANT current through the LED regardless of the overall source voltage.

And you need to insure that the LED devices are properly heat-sinked to keep them from simply burning up due to high internal die temperatures.

-----

* Making the World a Better Place -- One Helicopter at a time! *

Dave
11-03-2009 04:41 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
Bonez
Senior Heliman
Location: Eastern NC

My Posts This: Topic  Forum
dkshema man that was right on..I learned all that when i was building an LED strip for my saltwater fishtank..
11-03-2009 10:24 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
Hamo
Senior Heliman
Location: Ireland

My Posts This: Topic  Forum
Good stuff Dkshema but you haven't addressed the original question of this thread. The flashlamp has 3 X 1.5 batteries, 4.5 volts or 4.8 volts with new batteries. So why do the LEDs burn out with 4.2 V Lipo and work OK with 4.8 V Alkaline or whatever ?
Hamo
11-03-2009 11:14 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
dkshema
rrMaster
Location: Cedar Rapids, IA

My Posts This: Topic  Forum
Guess I'd have to find the flashlight you cannibalized and tear the thing apart to see what's inside. If you're actually burning out the LEDs, they are somehow getting too much current through them or they aren't properly heat sinked when removed from the rest of the flashlight.

-----

* Making the World a Better Place -- One Helicopter at a time! *

Dave
11-03-2009 02:12 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
Nice INFO!!


Thanks to everyone who chimed in here especially 'dkshema' with the detail answer!!

The thing that stumped me the most was that it burned out on 3 or 4th pack...never the first one or 2....( i used 2 flashlights so far as they are 2 bucks and change at Advanced Auto Parts)

Iam ataching some pictures of the Flashlight, a picture of the working LEDS the burnt LEDS and the OVERSTUFFED 450 this is going on... BTW I said this before, NO I CANNOT PUT MORE LIGHTS ON IT!

Ideally a constant power source would be the ideal solution, but because this is a small already OVERSTUFFED 450 more weight is not really desirable... Plus i really dont care if i burn them out while testing... The LED'S purpose are to light up the inside of the canopy as this produces a cool effect making the chopper i tiny bit more visible and i can fly without it just fine...

But its nice so I will try the resistor dkshema suggested...
a 2 OHM 1/2Watt Resistor wired in series and see how it does...

Now i just need to find a place to get one
Radio Shack has a 2 OHM 10Watt Resistor
http://www.radioshack.com/product/i...oductId=2062290
But thats not good right?

This place
http://shop.vetcosurplus.com/catalo...roducts_id=8322
But they want $16 for one $.29 resistor LOL

Will call local Cables N Connectors and see if they have one!

Thanks again for the help!!! Really apreciated!!
I will get the resistor and install it then report back on how it went...

Pete




P.S. CANNIBILIZED....HAHAH

Fly, Fly then Fly some more!!
11-03-2009 02:13 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
halley
Heliman
Location: Manchester, CT, USA

My Posts This: Topic  Forum
I'd be very surprised if the LEDs are wired for the math up there. Two ohms is not a very common load for LED circuits (or any power circuits, to be honest, hence the price). The LEDs you need should be a dime a dozen, almost literally. You should do some research on the types of LEDs in use rather than take the calculations as gospel just yet.

Usually, you wire a gang of LEDs or LED-strings in parallel, and a resistor per string. You're probably capped at one LED per LiPo cell per string, as the forward voltage for most LEDs is in the 2V~2.5V range, not 1V.

You can't really draw here, and I can't draw/post new images from work, but:

(+)-------|>|-------|>|-------/\/\/\-------(-)

A 2S Lipo's positive pole, which goes to an LED's anode, that LED's cathode connects to the next LED's anode in series, then the resistor in series, to the Lipo's - pole.

[ e d @ h a l l e y . c c ]
11-03-2009 02:27 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
halley
Heliman
Location: Manchester, CT, USA

My Posts This: Topic  Forum
You might also take a gander at sites like http://adafruit.com/ and http://www.SparkFun.com/ and http://MaceTech.com/store -- lots of others, but LED projects are a great way to get your feet wet with doing your own electronics.



[ e d @ h a l l e y . c c ]
11-03-2009 02:31 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
Hmm

Iam not positive about this flashlight array that i canibilized but it might have some sort of Resistor already soldered to it...

It might just be getting hot like posted above...

Its a cheapie Flashlight and iam willing to experiment

So what resistor (or other electrical component)
would u suggest to prevent burnouts?

I will try them all!



Fly, Fly then Fly some more!!
11-03-2009 02:35 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
halley
Heliman
Location: Manchester, CT, USA

My Posts This: Topic  Forum
Looking at your image, they've wired it up with two parallel strings, each with four LEDs in series. Maybe they did find some 1V drops, I can't see the resistor they've wired to it. Maybe they are doing a voltage doubling pump instead, but I doubt it.

It's not just a matter of "getting hot" (which is a sign of wasted power), it's tearing the LED apart with too much current. LEDs don't get hot if used within spec. Those flashlights will last for years if you keep giving them batteries. What's probably actually happening is that they're relying on the higher internal resistance of the NiCD batteries; the LiPos are much lower and thus giving a lot more current on the same load.

[ e d @ h a l l e y . c c ]
11-03-2009 02:39 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
Ok

Quote 
giving a lot more current on the same load.
Yup i figured this was part of the problem....but really boils down to whats can be expireminted with to resolve it (other than wire up your own LEDS, install another array..BTW in the front of the canopy there is 4 yellow LEDS, they are also wired up but there nowehere near as bright by design)

Ill check the backside and see if there is anythign else soldered in there.. this light is not mission critical, more of a ornamental illumination iam experimetnign with... If it works great if it doesnt then it doesnt...

So far i only have one possible solution to try -2 OHM 1/2Watt resistor...

Anythign else?

Worst case it burns out again...

Thanks for the replys!

Fly, Fly then Fly some more!!
11-03-2009 03:08 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
dkshema
rrMaster
Location: Cedar Rapids, IA

My Posts This: Topic  Forum
Note that the resistor I came up with in the math is based on some assumptions as to current, forward voltage, battery supply, and the data sheet for an LED rated at 3.8 watts, with a usable current range of 100 to 1000 ma. It's what is known as an "example".

Your mileage may vary.

You really need to know what it is you actually have to work with in order to make the necessary calculations.

-----

That flashlight is designed to run on Alkaline cells, I haven't looked at the specs on Alkaline cells for a long time, perhaps the thought that the design IS relying upon the internal impedance of three cells in series to limit current has some merit.

I'd look it up, but tonight I've got a ton of work I brought home to wrestle with.

-----

* Making the World a Better Place -- One Helicopter at a time! *

Dave
11-04-2009 12:52 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
So far so good

I soldered the 1/2Watt 2Ohm Resistor in series and did about 6 packs without burning out so all is well so far.. except my flying of course, so once i rebuild from a minor case of dumb thumbs ill test some more...

Thanks for the help!



Fly, Fly then Fly some more!!
11-09-2009 02:11 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
MutantGarage
Heliman
Location: Texas, USA

My Posts This: Topic  Forum
High power LED's usually need a constant-current regulator, as opposed to the constant-voltage power supplies you see most often.
Look on Deal-Extreme, they sell a lot of DIY flashlight kits.
The better regulators are based on switching power supply topology, they are far more efficient than a simple current limiting resistor.

LED's have efficiency curves as well, running an LED at 80% brightness may only take 1/2 the power (Vf*current).
11-10-2009 05:05 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE  
 
 
dkshema
rrMaster
Location: Cedar Rapids, IA

My Posts This: Topic  Forum
Constant current source was already mentioned in my post above. Now highlighted in BLUE

-----

* Making the World a Better Place -- One Helicopter at a time! *

Dave
11-10-2009 06:19 AM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   GALLERY
 
 
pkfrizzlefry
Senior Heliman
Location: Newington, CT (Central Connecticut)

My Posts This: Topic  Forum
Yup

In a perfect world a constant source would be ideal... but to keep weight down to a minimum the current from the lipo's along with the resistor is working just fine..

Thanks again!



Fly, Fly then Fly some more!!
11-10-2009 01:58 PM
PROFILE   PM   EMAIL   POSTS   BUDDY   IGNORE   HOMEPAGE  
 
 
1 page524 viewsPOST REPLY
MTA Hobbies . Model Rectifier Corp . ReadyHeli

.
.
Main Discussion > Electrical Enginering Help with Burned out LED's
 Print TOPIC Advertisers 

Subscribe to This Topic

Monday, November 23 - 8:12 am - Copyright © 2000 - 2009 runryder.com | email | link to rr | START HERE | NF