Issue driving high power LED with transistor, seems correct though?

[XOIIO]

Hey, finally starting to use transistors, currently I want to make a stroboscope using a 10w LED I have, for the sake of portability it will just be powered by a 9v, and draws ~150ma (rough guess based on analog meter) rather than the 900 it does at 12v. (measured constantly on not pulsed)

Anyways, that's too much for the arduino to handle so I grabbed some NPN's to drive the LED.

Now, I burnt the first arduino board (ebay pro mini and I have a few extras, they are cheap enough its not a huge deal) using a chunky transistor I found, it was either the board going bad or (what I think) was the transistor being too slow, the LED would just be on constantly, it didn't work properly. I tried directly driving it too and that is what burnt out the first board no doubt.

Anyways, second day, I figured out transistor circuitry more, grabbed a pack of NPN's (just the radio shack sort of pack) that I had laying around and wired it up like so:



10v from the supply going to arduino raw (I am fairly confident of this, though it is possible it might have been on VCC), I could use an attiny and that was the original plan but then I'd also have to use a voltage regulator, and I am using a fairly small heatsink.

Anyways, after a while it stopped working, I tested with a 5v 20ma led instead of the much larger one, no luck, tried to reprogram it and it was dead.

Anyways...  I just want to double check that this is the correct way to wire it up, two things could have killed the arduino last time, either it was a current spike or voltage spike from the supply, I noticed it acting up a bit, or, and what I am thinking, is that I may just have put the 12v into vcc 

That could also have been from when I was trying the 5mm LED instead, and had lowered the supply to 5v, can't remember, probably doesn't help that I wasn't awake for very long before this lol.

Anyways, I just want to verify that this is correct, and not some issue with how I wired up the transistor. I doubt it but it's one of those things where I doubt myself anyways.

[Simon]

Well you completely lost me, you need a resistor from the arduino pin to the base of the transistor as the base to emmiter path is just like a diode with a 0.7V voltage drop so if you feed anything over 0.7V to it you cause a short circuit, blow your arduino pin and damage the transistor.

You seem to have wired it up like a mosfet rather than a BJT

[Simon]

what is arduino raw ? why is it connected to 9V ?

[sdg]

No chance it will work with the LED connected this way...

What limits I_B ?

[sdg]

what is arduino raw ? why is it connected to 9V ?

Power supply to the Arduino. It goes to a 3.3 V onboard reg.

[Simon]

The LED is shown backwards

[XOIIO]

Ah whoops, I always do that with the LED on schematics lol.

Well you completely lost me, you need a resistor from the arduino pin to the base of the transistor as the base to emmiter path is just like a diode with a 0.7V voltage drop so if you feed anything over 0.7V to it you cause a short circuit, blow your arduino pin and damage the transistor.

You seem to have wired it up like a mosfet rather than a BJT

Ah, didn't know I needed a resistor from the arduino to the transistor, I bet that's why it stopped working and then drew excessive current.

So it would just be like this then I assume? (ignore the wirepad info, I accidentally hit drop instead of adding the component  Have to re-download the library later.)



Going to catch a couple more hours of sleep before classes, so it will be a while before I can test this out.

[sdg]

using a 10w LED I have, for the sake of portability it will just be powered by a 9v, and draws ~150ma (rough guess based on analog meter) rather than the 900 it does at 12v. (measured constantly on not pulsed)

How do you know your LED will not draw more than 150 mA @9v ?
Datasheet, measure, extrapolated from 12v or wild-ass guess ?
Unless the datasheet says otherwise, I'd play it safe and add a current-limiting resistor in series with the LED.

[Simon]

yes the LED is a singe one will need limiting as it should see no more than 4V unless it's a module

[XOIIO]

using a 10w LED I have, for the sake of portability it will just be powered by a 9v, and draws ~150ma (rough guess based on analog meter) rather than the 900 it does at 12v. (measured constantly on not pulsed)

How do you know your LED will not draw more than 150 mA @9v ?
Datasheet, measure, extrapolated from 12v or wild-ass guess ?
Unless the datasheet says otherwise, I'd play it safe and add a current-limiting resistor in series with the LED.

I tested it on my power supply, didn't go past 200ma as I said in the post you quoted. When I get. Home i can get a more exact figure but it's well within the transistors limit

[XOIIO]

It's a 10 watt module

[Simon]

Module, do you mean it comes with a current limit resistor ? Any LED needs the current limiting in some way, what are you powering it with ? if it's a 9V battery then the battery's internal resistance will probably be acting as a current limit but this is not good as it overheats the battery.

[XOIIO]

Eh by the wrong thing so it didn't quote just replied, whatever.

Not sure if it has a limiting resistor or not, it's a module of 9 single led chips. I've ran it for a bit on a 9 volt, didn't notice any heating but I'll try it for longer when I get back.

It's not for any super long period of use use anyways, maybe a couple minutes at most.

[Simon]

what I'm trying to work out is if your LED's have a limiting resistor or if the internal resistance of the battery is limiting the current. You need to control the current with a resistor built into the module or your circuit. 9V batteries are not capable of supplying may amps so if you try to draw a lot of current you may hit a limit of the 150mA tou have found.

[sleemanj]

Not sure if it has a limiting resistor or not, it's a module of 9 single led chips.

It does not have current limiting.  However, at 9v it's a little below the actual forward voltage (it must be a 3s3p array, each die ie 3.something V), combine with a 9v battery's internal resistance, and the current (and light output) will be minimal.

At a minimum, you should select your base resistor such that given the transistor's approximate gain the transistor will perform in the function of a current limiter.

[Simon]

yes or put an emiter resistor in too and you will be able to control the current accurately and not need a current limit resistor.

[idpromnut]

It might be a good idea to have a closer look at how that type of transistor works. Ben Krasnow has a video on how to use NPN transistors: . I'd suggest you breadboard the circuit without the Arduino to see what is going on (use the V+ rail to simulate "HIGH" and GND as "LOW").

[rs20]

yes or put an emiter resistor in too and you will be able to control the current accurately and not need a current limit resistor.

Using an emitter resistor will be much less accurate than a conventional current limit resistor on the LED unless a) You omit the base resistor and have the base at 3.3v, OR b) you accurately known the current gain of the BJT.