Author Topic: Transistor and LED without current limiting resistor  (Read 12204 times)

0 Members and 1 Guest are viewing this topic.

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Transistor and LED without current limiting resistor
« on: August 15, 2013, 10:19:28 am »
Hello everyone,

I have a transistor being switched by an ARM micro, this transistor is turning an infrared LED on and off.
My problem is, the LED is only bright enough to control a TV with it just 10cm from the IR receiver, any further and it doesn't work.
I'm powering the LED from 5v going through a 150ohm resistor. If I remove the resistor then the LED is bright enough to work from a distance of a few metres. I'm assuming the combination of PWM of the LED to send the IR code and the current limiting resistor is what's causing the LED to be so dim.

My question is, do I need the resistor? Everything works great without it.
Should I just reduce the value, maybe make it a 10ohm?

Thanks guys.  :)


- Anthony
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Transistor and LED without current limiting resistor
« Reply #1 on: August 15, 2013, 10:34:25 am »
assuming the led drops 1.5V, it should need 175 Ohm for 20mA current draw, which should be plenty

the fact your led hasn't spontaneously let out its smoke while omitting the resistor would lead me to think your switching transistor is in saturation, e.g. its being governed by the available base current, and that that is where your issue lies,
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #2 on: August 15, 2013, 10:51:21 am »
Thanks Rerouter.

The PWM of the led at 38kHz and the fact that I'm only turning it on for 0.5 seconds at a time is probably helping limit the magic smoke.

I'm using a PN2222 transistor (http://www.fairchildsemi.com/ds/PN/PN2222.pdf).

VCE(sat) (IC=500mA, IB=50mA) = 1V
VBE(sat) (IC=500mA, IB=50mA) = 2V
HFE (VCE=10V, IC=0.1mA ) = 35
HFE (VCE=10V, IC=150mA) = 100 (300 max)

Micro Output:
V = 3.3V
I = 6mA

The base resistor has a value of 520ohm.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Transistor and LED without current limiting resistor
« Reply #3 on: August 15, 2013, 11:06:32 am »
Maybe your IR led is the wrong wavelength for the IR receiver&filter in your TV, that would definitely cut the range back.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Transistor and LED without current limiting resistor
« Reply #4 on: August 15, 2013, 11:14:05 am »
the pwm would not help much....

i'm kinda agreeing with psi on his thoughts,

if you exceed the current spec of an led its output shifts to a lower frequency, could be what is improving the range,
 

Offline cravenhaven

  • Contributor
  • Posts: 40
  • Country: au
Re: Transistor and LED without current limiting resistor
« Reply #5 on: August 15, 2013, 11:22:24 am »
I suspect that you are not driving the transistor hard enough. The 520 ohm base resistor would only allow approx 5mA base current at 3.3V so is probably not in saturation which is where I think you need it to be. You really need to scope it to see what is happening.
If it was me I would be using a FET such as a 2N7000 and leave the resistors as they are to start with.
Also it might help to know which IR LED you are using to enable us to know what its operating characteristics are.
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #6 on: August 15, 2013, 11:24:54 am »
Hello Psi,

The only thing that varies between devices is the carrier frequency and encoding format, the wavelength of the IR light doesn't vary/matter so much. I know it's a brightness issue as I have looked at the light with a web camera and it's dim compared to standard remote controls.
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 3020
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: Transistor and LED without current limiting resistor
« Reply #7 on: August 15, 2013, 11:25:48 am »
Turn the transistor on solid with the current limiting resistor in place, and measure the voltage drop across your current limiting resistor, and thus deduce the current, no point chasing wild geese if there could be a fat one sitting in front of you.
~~~
EEVBlog Members - get yourself 10% discount off all my electronic components for sale just use the Buy Direct links and use Coupon Code "eevblog" during checkout.  Shipping from New Zealand, international orders welcome :-)
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #8 on: August 15, 2013, 11:32:07 am »
Hello sleemanj,

I will measure the voltage drop across the resistor and then get back to you.
Is there anything else I should measure to help diagnose this. I'm new to using transistors without a complete and tested schematic being provided.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Transistor and LED without current limiting resistor
« Reply #9 on: August 15, 2013, 11:35:26 am »
the wavelength of the IR light doesn't vary/matter so much.

It could matter quite a lot if you just picked a random IR led from somewhere and not one intended for remote control.

The IR sensor in the TV probably covers a pretty wide range of wavelengths but the IR filter on the front of the sensor can be quite narrow to reduce interference from other sources. CFTs etc.

If your IR led is 1300nm for example, it will only work point blank when used with a typical 940nm IR filter and sensor. It will also show up dimmer than other remotes when viewed on a web cam due to the webcams IR response falling off at higher wavelengths.

But i do agree that the transistor is a more likely reason for the issue.
« Last Edit: August 15, 2013, 11:42:20 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #10 on: August 15, 2013, 11:41:40 am »
Hi Psi,

Thanks, I hadn't thought of that.

The LED is a Vishay TSUS5400 (http://docs-asia.electrocomponents.com/webdocs/0e22/0900766b80e22d5c.pdf)
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Transistor and LED without current limiting resistor
« Reply #11 on: August 15, 2013, 11:45:08 am »
OK, its made to be driven far harder than i was thinking,

Try 33 ohms, which brings you up to 100mA,

 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Transistor and LED without current limiting resistor
« Reply #12 on: August 15, 2013, 11:46:07 am »
950nm is fine and should work no probs to control a TV

That led can handle 150mA  @ 1.3V

So driving it at absolute max would be.. 5V input - 1.3V = 3.7V/0.150A = 24.6ohms

I would try a 33R and see how that works.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #13 on: August 15, 2013, 12:01:50 pm »
Thanks guys,

Should I also investigate the base resistor value to see if this is causing problems or if this works will it suffice?

Using the calculator at http://kaizerpowerelectronics.dk/calculators/transistor-base-resistor-calculator/ and specifying the values from the PN2222 datasheet I get a base resistor value of 216ohm (220ohm). I am currently using a 520 ohm resistor based off of another online calculator I used a few weeks back. This is a huge difference and I'm not sure which to trust.

As I said in a previous reply, I am new to using transistors and I really appreciate you all taking the time to help me work this out.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Transistor and LED without current limiting resistor
« Reply #14 on: August 15, 2013, 12:03:28 pm »
I would try a 160R base resistor with that 33R led resistor.
(Assuming your 3.3V i/o can handle 20mA)
« Last Edit: August 15, 2013, 12:05:23 pm by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #15 on: August 15, 2013, 12:06:51 pm »
Hello Psi,

The specs for the I/O are:

3.3v logic
Source current of 6mA
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: Transistor and LED without current limiting resistor
« Reply #16 on: August 15, 2013, 12:11:31 pm »
ok, well that may not be enough to drive that IR led at full power given the gain of that transistor at saturation.

You could try two transistors in a darlington config.

Or, as others have said, try a 3.3V logic level N channel mosfet (with no gate resistor)
« Last Edit: August 15, 2013, 12:19:21 pm by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #17 on: August 16, 2013, 04:58:05 am »
In a circuit simulator I have set the circuit up with a 300ohm base resistor and 33ohm current limiting resistor and this gives me 1.2v @ 92mA for the LED with only 6mA being drawn from the MCU I/O pin. I will test this out on the actual circuit when I get home tonight.
If this fails then I might have to go back to the drawing board and use an N-Channel 3.3v Logic Fet as has been suggested (If I can find one; RS, DigiKey and Element14 have come up empty when specifying "N-Channel" and "2V Logic" or "3v3 Logic" as the filters).
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 3020
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: Transistor and LED without current limiting resistor
« Reply #18 on: August 16, 2013, 05:13:03 am »
If this fails then I might have to go back to the drawing board and use an N-Channel 3.3v Logic Fet as has been suggested (If I can find one; RS, DigiKey and Element14 have come up empty when specifying "N-Channel" and "2V Logic" or "3v3 Logic" as the filters).

I think the surface mount world tends to have more selection in logic level mosfets than the through hole one, try the SI2302DS
  http://de.mouser.com/ProductDetail/NXP-Semiconductors/SI2302DS215/?qs=sGAEpiMZZMvKM5ialpXrmhvhsJkB2qKw

In brief; N Channel SOT23, max current 2A, RDSon 85mOhm at VGS 4.5v, 115mOhm at 2.5v VGS, 830mW


~~~
EEVBlog Members - get yourself 10% discount off all my electronic components for sale just use the Buy Direct links and use Coupon Code "eevblog" during checkout.  Shipping from New Zealand, international orders welcome :-)
 

Offline ve7xen

  • Super Contributor
  • ***
  • Posts: 1192
  • Country: ca
    • VE7XEN Blog
Re: Transistor and LED without current limiting resistor
« Reply #19 on: August 16, 2013, 08:24:59 am »
So driving it at absolute max would be.. 5V input - 1.3V = 3.7V/0.150A = 24.6ohms
Since the duty cycle of the signal is going to be somewhere around 50% at a reasonably high frequency, it should be safe to increase the peak current substantially. The datasheet specifies a maximum peak current of 300mA (2x) with 50% duty cycle and a period of 100us (10kHz). It should be safe to run it at at least 250mA if the only mode possible is the PWM.

Once you've tested your code well, of course, so you can't easily burn the LED out by leaving it on.
« Last Edit: August 16, 2013, 08:28:22 am by ve7xen »
73 de VE7XEN
He/Him
 

Offline Stove

  • Contributor
  • Posts: 14
Re: Transistor and LED without current limiting resistor
« Reply #20 on: August 16, 2013, 09:05:22 am »
The LED can be driven much higher for 50% duty cycle: 300mA. Furthermore you are forgetting the voltage drop in the transistor, but granted it's rather low, 0.2V @ 300mA. I have put some graphs and calculations together in this image, please excuse the crudity of the drawing. Final result 12.3Ohm. If you really want the maximum out of it, find a way to measure the current and trim the resistor till it's 300mA. ve7xen was pretty close.
« Last Edit: August 16, 2013, 11:11:53 am by Stove »
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
Re: Transistor and LED without current limiting resistor
« Reply #21 on: August 16, 2013, 09:28:21 am »
The LED can be driven much higher for 50% duty cycle: 300mA. Furthermore you are forgetting the voltage drop in the transistor, but granted it's rather low, 0.2V @ 300mA. I have put some graphs and calculations together in this image, please excuse the crudity of the drawing. Final result 12.3Ohm. If you really want the maximum out of it, find a way to measure the current and trim the resistor till it's 300mA. ve7xen was pretty close.
That won't work, the MCU output is only 3.3 V so the LED current would only be 18mA with a 12R series resistor using an emitter follower.
 

Offline Stove

  • Contributor
  • Posts: 14
Re: Transistor and LED without current limiting resistor
« Reply #22 on: August 16, 2013, 10:26:08 am »
That won't work, the MCU output is only 3.3 V so the LED current would only be 18mA with a 12R series resistor using an emitter follower.
I'm sorry, you are right, the transistor should be like it is now (updated the image) and even then the transistor is too weak, 6mA * 35 = 210mA, so would have to darlington indeed.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
Re: Transistor and LED without current limiting resistor
« Reply #23 on: August 16, 2013, 11:05:40 am »
With no base resistor, it might overload the MCU output which could cause it to overheat, unless it's adequately protected.
 

Offline AMyattTopic starter

  • Contributor
  • Posts: 32
Re: Transistor and LED without current limiting resistor
« Reply #24 on: August 16, 2013, 01:43:43 pm »
Hi Guys,

Thanks for all your help. Would changing to the MPSA13RLRAG (http://www.onsemi.com/pub_link/Collateral/MPSA13-D.PDF) Darlington transistor work?

I am ALWAYS driving it with a PWM signal of around 38-40kHz at 50% duty cycle.

When first building this circuit I underestimated the base resistor value, the MCU shut down. I have no doubt it could have been much worse though.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf