High Power LED (3W)- PWM driven by Electron

Hi All.

Would a circuit like this be an overkill to drive a high power LED?

I am calculating at 3.7v around .7 Amp… That is why I am picking a MOSFET. I didn’t have luck at digi finding a good transistor with this rating… Although I am not a pro, so I am guessing a lot here :slight_smile:


D1 goes to D1 on the electron (PWM in case I want to DIM it)
V_BATT goes directly to the battery positive.
GND… Well to ground. :slight_smile:

Any suggestion welcomed. This is for a beacon for a High Altitude Balloon. I am also worried that it will drain my battery :blush: The battery is a LiPo 6000 mah. But that’s another issue :slight_smile:

Hello,

Let me ping someone that might be able to help, @rickkas7 or @ParticleD are you able to assist?

1 Like

Greatly appreciated!

1 Like

No, that's not overkill. You certainly can't drive 700 mA directly from the device, and a MOSFET is easier to use than a bipolar transistor. It should be ok as long as the MOSFET you chose can handle that current, and its gate voltage is low enough to be turned on by 3.3 volts. I usually use a FQP30N06L just because they're readily available, and can be triggered by a Photon (though its 30 A rating is overkill). Usually, you need a resistor in series with your LED to limit the current through it, but without knowing what specific LED you're using, I can't say what value you should be using.

Using a MOSFET wouldn't cause you any battery problems, only how often, and for how long you flash your LED.

Hi @frlobo

You don’t really have enough information there for us to help you very much. What kind on LED for instance? At least the current and voltage specs would be helpful.

You have an N-channel MOSFET low-side switch and that part looks great. The polarity on the BCN (connector?) is backwards in your schematic, the + should be on the top connection pin 2. PWM from D1 is OK on Electron, but what you really want, I think, is a narrow pulse with a long repetition rate. Something like 100ms every 3 secs which could be custom PWM but might just be a timer or watching millis() in your code.

You do not have any kind of current limiting for the LED. I don’t think it is a good idea to skip that. 6000mAH is not going to last very long with a 3W LED but you knew that. Your battery will not be able to supply 3.7V over the entire current life you are trying for, so depending on the LED specs, you might be better served with a boost converter to drive the LED.

If you can fill in a few more details, people can help you more.

1 Like

I will work it out tomorrow and solve the issues you pointed out. Thanks for the help.

The resistor will be mounted externally near the resistor.

Can you elaborate more on the the boost converter? How will that save battery?

I was thinking more in terms of a buck converter to use an 11volt battery and power the system from the buck output for 3.3… of course loosing the charging festures and powering the electron thru the supply pins not the battery pins. But again… I am low on experience.

Which has the highest priority, your telemetry(Electron), or the Safety Light Beacon ?

You will be able to calculate the average power required for your LED Light per hour of flight.
You will be “guessing” at the average power for the Electron, Xbee/LoRa, etc.

I have no experience in High Altitude Balloons, but my gut feeling is to provide a dedicated battery for the Safety Light Beacon, if a flashing light is required for flight.

You have an interesting project.

Telemetry by far…

Aw @bko suggest I am borrowing his idea on flashing the beacon for say 100-200 ms every 3-5 seconds and disabling it completeley at above 35-42,000 ft to conserve battery (no need for the beacon that high).

I don’t know if PWM at say 50% duty cycle will consume less energy than at 100% even if it dims.

BY THE WAY: @bko This is the LED I am using: I plan to enclose it in a 3D printed small box with the resistor soldered to it… (I sent the PCB to the fab house without it!).

Thanks for the polarity notice. Fixed!

My battery is a 6,000 mah battery. I think the external battery idea for the beacon is a great idea… I will make my rough calculations and if needed add it.

@Ric Thanks for the recommendations. I will take a look at that MOSFET.

Hi @frlobo

Thanks for the extra info. If you are really choosing a green LED (white might be more visible) then the data at Sparkfun says

Additionally, each LED requires a forward voltage of 3.2-3.8V at 750mA.

So the way to read this the LED will light up at the forward voltage (Vf) at somewhere around 3.2-3.8V depending on the LED and temperature etc. and needs to be current limited to 750mA for continuous operation.

Since the forward voltage of the LED and the nominal voltage of your battery are similar, it could work out that you can just connect the two, but lets say you get an LED with Vf of 3.8V, then your LED will stop flashing fairly quickly as your battery goes from fully charged at 4.2V down to 3.8V. On the other hand if you get an LED with Vf of 3.2V, your LED will stay lit until your battery is essentially dead.

This is why I think you should consider a small boost converter to increase the voltage and keep the LED lit when the battery gets lower. The boost converter does not need to be a big power IC, you might just need a simple circuit like a Joule thief. If you have not seen the Joule thief, you should definitely search for it on the web.

Since you are pulsing the LED, you need less current limiting than you would for continuous operation.

4 Likes