Give me more light!

Discussion in 'General Electronics Chat' started by MrSoftware, Jan 7, 2016.

  1. MrSoftware

    Thread Starter Active Member

    Oct 29, 2013
    503
    124
    I'm looking for some guidance choosing a new IR LED to give me significantly more IR output on my IR remote transmitter. I'm in a time crunch so I need to order all my parts to experiment with in one shot. ;)

    I've built this little Velleman IR remote transmitter (schematic below) and it functions, but doesn't put out nearly as much IR light as need, as this will be used outside in the daytime and needs to overcome natural IR light. I'm thinking my best option is more powerful LEDs to replace LD1 and LD2, does anyone have a suggested model?

    I think LD1 and LD2 are the weak link because of the following; using my scope I can see the voltage drop across LD1 and LD2 is about 3v during a pulse (the anode of LD1 never dips below 3v during a pulse), and this more or less matches the datasheet. Measuring at the cathode of LD2 I can see that T1 is pulling down to about 0V so I believe T1 is fully turning on. So as best I can tell, LD1 and LD2 are at full capacity and just don't put out much light. I could also try replacing R3 with a smaller value, but I'm not sure what that's going to do to the life of these LEDs. All suggestions are welcome!

    Data sheets:
    LD1/LD2: https://www.rcscomponents.kiev.ua/datasheets/l-934f3bt.pdf
    T1: http://www.micropik.com/PDF/bc635.pdf

    schematic.jpg
     
  2. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    IR doesn't work hardly at all in Sunlight... What do think you feel on a sunny day?
     
  3. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,059
    3,821
    First, you could try to increase the power of the LEDs you have. Little 5mm indicator-sized LEDs are normally driven up to 100 mA at constant currnet and up to 1 A in pulses (very low duty cycle) of 10 uSecond pulses or less - see the datasheet, I think they normally have over-driving info.

    Also, Right now you are driving T1 with only 5mA to the base after you account for R2 and LD3 voltage drops. The base of T1 should be driven at about 10% of the collector current for good switching on a transistor with Hf of 100.

    I would change the circuit as follows...
    Move LD3 out of the current position. If you need an LED-ON indicator, let the microcontroller switch it from GP0 or GP1, those are not doing anything.
    change R2 to 220 ohms to give you 47 mA of base current (remember, that is without LD3).

    Then, you can change R3 to 22 ohms (assuming you pulse the GP2 output at reasonably low duty cycle). You can also add several more IR Emitters on your circuit to increase brightness coverage area (many IR LEDs are narrow angle beams so read the datasheet to know how focused you need to be).

    EDIT: your IR-Emitters have a 50-degree (very broad) beam. That is 50 degrees on each side of the center line. Look for something under 15 degrees will greatly improve power.

    If you pulse the output at a given frequency (38k hz is common), then you can filter the receiver to focus on 38k and have a better chance of it working in the sunlight. Not great but, if you can shield your receiver so it does not get direct sunlight, you have a chance.
     
    MrSoftware likes this.
  4. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    3,792
    945
  5. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    You should also consider shrouding the receiver to reduce the ambient light falling on it and adding lenses to focus the beam.
     
  6. SLK001

    Well-Known Member

    Nov 29, 2011
    819
    229
    Take a digital camera and look at the viewfinder while you transmit. The sensor in the camera is IR sensitive, so you can see if your LEDs are really as dim as you believe so.
     
  7. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,059
    3,821
    Not all cameras will do that. Most from the 1990s and early 2000s but more recent cameras (e.g. iPhone) seem to use a different technology or are able to filter out point sources of IR.
     
  8. crutschow

    Expert

    Mar 14, 2008
    13,019
    3,235
    You could use a higher power IR LED such as this or this.
    An IR laser would have an even higher power beam but would likely be too narrow for your application since you would have to point it exactly at your target.
     
  9. MrSoftware

    Thread Starter Active Member

    Oct 29, 2013
    503
    124
    Thanks everyone for the tips, I will apply it all. @crutschow and @Kermit2 those are some serious IR LEDs!

    I realize IR isn't the best for outdoors, but we need line-of-sight-only control. This is for an add-on accessory device mounted outside a car for handicap people and we want to make sure there are no accidental button pushes from inside the car while the car is driving, but not burden the handicap people with a master on/off switch on the device. This is a proof of concept at this point, but has to be ready for a demo this weekend. Time will tell how it shakes out. :)

    @GopherT - I will try all of your tips, thanks! The processor is pre-programmed and I can't change it, but the data seems to come in bursts. Several bursts per second, and during each burst the signal is at 35kHz, square wave, and the duty cycle seems to be about 50%. I will try your circuit suggestions with the current LEDs first. Then if I need even more light, what do you think about replacing the LEDs with these (they're available locally = today):

    https://www.radioshack.com/products/radioshack-infrared-led-emitter-and-detector?variant=5717550213

    I think this is the same, emitter only:

    https://www.radioshack.com/products/high-output-infrared-led?variant=5717550469

    The data sheet is lacking, but they're a more narrow focus than what I have now and the same wavelength, and support 3x the forward current.

    Here is the receiver circuit that we'll be using:

    receiver.JPG
     
    GopherT likes this.
  10. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,059
    3,821
    @MrSoftware
    You can add additional levels of safety so it does not activate while driving. There is a switch in the vehicle to test whether it is in PARK or not. Use that in your logic. Also, the reverse logic. Make sure the car cannot be removed from PARK unless the unit is in a stowage state. For example, the car cannot be taken out of PARK unless the brake is pressed. Link those two pieces of logic. May not be needed for your prototype demo but at least mention it as a future upgrade.
     
Loading...