Arduino Forum

Using Arduino => General Electronics => Topic started by: markyj on Nov 26, 2016, 08:15 pm

Title: IR LED without resistor nano
Post by: markyj on Nov 26, 2016, 08:15 pm
I am trying to send IR signals using an arduino nano. I have got it all working, it sends signals fine, the device picks up the signals fine, my only issue is range.
I had calculated the resistor I needed through a calculator site for it and everything works great, just the range is the issue. When I removed the resistor from the circuit the range is much better but I'm worried about having that as a permament solution as everywhere tells me I need to use one but then I looked at some of the outputs of the nano and checked against the LED and found that the nano doesn't even produce the ma's of the max ma on the LED. So I was hoping someone could confirm if I can use the LED without a resistor. Im using a nano, it also has a radio reciever module on it too and an IR LED with the following stats:
- Forward Current: 50 mA
- Forward Voltage: 1.2V
- Max Forward Voltage: 1.6V
I was hoping that the fact the duration of the LED being on would be milliseconds occasionally might be ok enough to run without a resistor too. Could someone please let me know on their thoughts of this or if there are any better LEDs to use for this with better range.
Thanks
Title: Re: IR LED without resistor nano
Post by: DVDdoug on Nov 26, 2016, 09:22 pm
Quote
When I removed the resistor from the circuit the range is much better but I'm worried about having that as a permament solution
You are taking a risk.   I wouldn't go "into production" with a design like that, but it's up to you if you don't mind the risk of frying your Arduino (or the risk of an otherwise possibly unreliable design).

You are exceeding the current rating of the Arudino, and probably the current rating of the LED too.

You can drive the LED with a transistor or MOSFET, but you should calculate your resistor value to stay under the 50mA rating for the LED.
Title: Re: IR LED without resistor nano
Post by: tinman13kup on Nov 26, 2016, 09:27 pm
the nano doesn't even produce the ma's of the max ma on the LED.
If the led consumes more power than the nano can produce, you will burn up the nano while getting unacceptable performance from the led.
 The resistor is there to limit the amount of current the led can use, normally to prevent damage to the led. You need to power the led from a different supply, and use a small transistor controlled by the nano to operate it. You could also use a mosfet. Either way, you need a different power supply for that led.
Title: Re: IR LED without resistor nano
Post by: markyj on Nov 26, 2016, 10:43 pm
Im not sure I quite understand how it is i risk frying either if the led is rated 50ma but the ardiuno only provides 40ma. Does that not mean that only a max of 40ma will be going through the led? And a resistor would lower that 40ma even more? Please explain what I have wrong here, honestly my electronic knowledge is very bad as im better as a coder than electrician.
I am very much thinking about just going with it as its only a personal project and if anything does fry the parts to replace are cheap enough but i will then need to rethink my plan. Im kinda one those kinda people that plugs stuff in and hopes it doesnt blow up and if it does, dont do that again. Also the arduino will just be powered via standard usb cable too, nothing fancy. Plan b is to add the resistor and wire up multiple irs close to the devices like christmas lights that way it will definately work and be safe too.
Title: Re: IR LED without resistor nano
Post by: CrossRoads on Nov 26, 2016, 10:45 pm
There is nothing limiting the output current of an Arduino pin. If the user does not take steps to limit the current to 20mA continuous, and limits the time when 40mA is drawn, there is risk of blowing the pin. It's that simple.
Title: Re: IR LED without resistor nano
Post by: tinman13kup on Nov 26, 2016, 11:21 pm
The pin on the Arduino can put out X amount of current. You are trying to TAKE X+X from it. The arduino essentially sees it as a short circuit.

What do you suppose would happen if you try to power your whole house with a cellphone battery? The answer isn't "nothing". Yes, nothing in the house would operate, but the battery would essentially be short circuited and probably blow up.

Your devices cannot exceed the current output of the supply, plain and simple
Title: Re: IR LED without resistor nano
Post by: Delta_G on Nov 26, 2016, 11:57 pm
Im not sure I quite understand how it is i risk frying either if the led is rated 50ma but the ardiuno only provides 40ma.
You misunderstand.  Nothing limits the current out of the Arduino pin.  That is the purpose of the resistor.  Without the resistor the Arduino will provide whatever current you ask it to.  You could try to draw ten amps through it.  But anything more than 40mA will burn it up.  Your LED wants to draw 50mA.  So you're risking burning up the pin on the Arduino.  Arduino is going to try to deliver 50mA and in the process will burn itself up. 

We don't suggest the resistors because we are in the resistor business.  You really do need it.  Lots of really smart electronics folks around here.  We're not bullshitting you.  You really do need that resistor. 
Title: Re: IR LED without resistor nano
Post by: INTP on Nov 27, 2016, 03:44 am
You need a resistor, period.
You need a transistor to switch the IRLED when giving it a better power source. You'd see some improvement powering the IRLED with the 5V pin's 200mA with an output pin controlling the transistor.
You'd see the best range giving that IRLED 800-1000mA with a proper power supply and if you keep your signals at <1% duty cycle. Read the datasheet.
Title: Re: IR LED without resistor nano
Post by: markyj on Nov 27, 2016, 11:27 am
So, the led takes 50ma regardless, the arduino is only safely capable of providing 40ma and the resistor ensures that only 40ma is given to the led? Is that correct?
If so, using another pin on the nano could i provide an extra 10ma to the led safely too? Also, that said should i be calculating the resistor value at 40ma and not 50ma? As that is the value i calculated the resistor at, but i think after this new info i may have calculated wrong too.
Also, im not saying i dont believe anyones answers im just trying to wrap it around my head and asking about other possibilities i won't be using a second power supply so either i will just have to deal with low range or try alternative setups maybe the transistor idea if i can somehow split the power between 2 pins to provide the max power to the led.
Title: Re: IR LED without resistor nano
Post by: Smajdalf on Nov 27, 2016, 12:59 pm
Measure the current through the LED (voltage dropped over current limiting resistor). You may find the current is less than you think and safe current (through lesser resistor) may be enough.
You could use transistor to power the LED (Google using transistor as a switch).
Internal resistance of pins is around 50 ohm: it is safe to power a LED directly when Arduino is powered from low voltage (3.3 V) or LEDs with high forward drop (blue or white). But not IrLED from 5V.
If you don't have/want to use transistor you may connect more pins together to provide more current. When calculating current you need to take internal resistance of pins into account.
Title: Re: IR LED without resistor nano
Post by: outsider on Nov 27, 2016, 02:17 pm
Something like this?
(http://forum.arduino.cc/index.php?action=dlattach;topic=437954.0;attach=188580)
A 2N2222 might be better, it can do 800 mA.
Title: Re: IR LED without resistor nano
Post by: INTP on Nov 27, 2016, 02:58 pm
So, the led takes 50ma regardless, the arduino is only safely capable of providing 40ma and the resistor ensures that only 40ma is given to the led? Is that correct?
If so, using another pin on the nano could i provide an extra 10ma to the led safely too? Also, that said should i be calculating the resistor value at 40ma and not 50ma? As that is the value i calculated the resistor at, but i think after this new info i may have calculated wrong too.
Also, im not saying i dont believe anyones answers im just trying to wrap it around my head and asking about other possibilities i won't be using a second power supply so either i will just have to deal with low range or try alternative setups maybe the transistor idea if i can somehow split the power between 2 pins to provide the max power to the led.
I'm glad you're not one of those professed newbies that refuse to listen to wise advice even though it really seems that way.
How will you be powering the Arduino? Battery? USB? Wall plug?
As I said in my post you didn't read, you can use the 5V pin of the Arduino as your higher current source.
And as I said in my post that you didn't read, the IRLED handles current up to 1000mA depending on duty cycle. Had you read my post, you wouldn't come out saying that the IRLED 'takes 50mA regardless'.
And as others have tried to spell out for you, Arduino pins should only have 40mA pulled from them max, and it is your job to limit what is pulled, THERE IS NOTHING ON THE ARDUINO SIDE THAT ENFORCES THE 40mA LIMIT. Get that through your head.
No resistor is the same as connecting both ends of a battery to each other. While the battery has a listed max draw, shorting it out because you're fiercely dense means the battery will draw until it destroys itself. Nothing enforces the listed max, that's your job.
Title: Re: IR LED without resistor nano
Post by: MrAl on Nov 27, 2016, 03:52 pm
I am trying to send IR signals using an arduino nano. I have got it all working, it sends signals fine, the device picks up the signals fine, my only issue is range.
I had calculated the resistor I needed through a calculator site for it and everything works great, just the range is the issue. When I removed the resistor from the circuit the range is much better but I'm worried about having that as a permament solution as everywhere tells me I need to use one but then I looked at some of the outputs of the nano and checked against the LED and found that the nano doesn't even produce the ma's of the max ma on the LED. So I was hoping someone could confirm if I can use the LED without a resistor. Im using a nano, it also has a radio reciever module on it too and an IR LED with the following stats:
- Forward Current: 50 mA
- Forward Voltage: 1.2V
- Max Forward Voltage: 1.6V
I was hoping that the fact the duration of the LED being on would be milliseconds occasionally might be ok enough to run without a resistor too. Could someone please let me know on their thoughts of this or if there are any better LEDs to use for this with better range.
Thanks
Hi,

A couple things...

First, the 50ma is probably an average current rating and that is what the LED *should* be run at not what it does run at no matter how you power it.  If you overpower it, you burn out the LED.  So you can drive it at 200ma but that would burn it out so you dont want to do that.  That is what the resistor is for, to prevent an over current level.

Second, since that is probably an average rating (check data sheet) you can probably pulse it much higher like 100ma at 50 percent duty cycle and thus with a standard IR pattern you can get much farther communication distance.

Also, since the characteristic voltage is around 1.2v and you are using a 5v supply, the current is about 3.8/Rs where Rs is your series resistor.  Thus a 100 ohm resistor gives you about 38ma, but then again the output pin probably does not put out the full 5v at that current so you have a little safety margin already.

Next, to drive the LED safely from a 5v Arduino, use 2 i/o pins set up as outputs instead of just 1.  Use two 100 ohm resistors, one for each pin, to drive the LED assuming you will pulse it at about 50 percent duty cycle or less.  Alternately you can use 3 i/o pins set up as outputs and use 110 to 120 ohm resistors, one for each pin and that will get you up near 80 or 90ma peak current.

If you want to stick with 50ma anyway, use 2 pins and one 150 ohm resistor on each of the two pins.  That will mean you'll get about 25ma from each pin for a total of about 50ma.

None of these designs require a transistor circuit but they do require that you have at least one more i/o pin to spare that you can use for the second output pin.  They also require that you set both output pins to the same state one right after the other using digitalWrite() with no other instructions in between, or use a direct register i/o software technique to set both output pins at the same time.

Of course your power supply has to be able to handle whatever peak current you decide to use in addition to the normal running current of the board.
Title: Re: IR LED without resistor nano
Post by: tinman13kup on Nov 27, 2016, 04:00 pm
So, the led takes 50ma regardless, the arduino is only safely capable of providing 40ma and the resistor ensures that only 40ma is given to the led? Is that correct?

No.
Led's are somewhat different from other components when it comes to current. A led on it's own will just act as a short and pull increasing current until it destroys itself. That's going to happen pretty quick depending on voltage/current available. Take your IR led and hook both leads to a car battery (12V, 500A) and you won't even get to see a flash before it pops.

Leds require additional means to regulate the current to the specifications of the led. The current (If) for any particular led tells you what is the SAFEST max current that can go through that led without damaging the led. The easiest way is to use a resistor for low power leds, based upon the voltage, and led Vf and If. High power leds (>1W) require different means to limit the current via current regulating supplies or chips.

The power supply (in your case a pin) are capable of safely supplying "X" Volts and "Y" Amps . Most supplies are voltage regulated, which means it will try to maintain the voltage regardless of current. The volts are fixed. A 5V 200mA supply will read 5V regardless if you are using 1mA or 100mA. As you try to use more current than it is designed to supply, the volts will start to drop as current increases, there is considerable heating of connections/ traces/ components until something basically has a meltdown. On a physical power supply, there are fuses that prevent damage/fire. In ICs, some chips have internal protections, but those are usually dedicated power controlling chips. The pins you are using go directly to the ATmega chip. There are no such protections there. There is nothing to prevent more than 40mA from going through that pin, other than the USER (that's you) NOT TRYING TO PULL MORE THAN 40mA.  Using components that regularly use the max ratings of a supply is also a bad idea in the long run if you expect things to last for any period of time.
Title: Re: IR LED without resistor nano
Post by: MarkT on Nov 27, 2016, 10:08 pm
1) The absolute maximum rating for an Arduino pin (ATmega boards) is 40mA.  You'd never run the pin _at_ its absolute maximum if you want a reliable circuit, you run it comfortably _below_, say 25mA or less.

2) The LED has a continuous rating for current and a pulse rating.  The pulse rating is what the device
can stand without immediate damage, the continuous rating is what won't cause it to overheat and
be thermally damaged for continouous on operation.  If you are pulsing the LED you'd go to the right
graph in the datasheet to see what maximum current you can run for the duty-cycle you use, or estimate
it from the average current (which has to be equal (or better still less) than the continuous rating).

3) Arduino pins as OUTPUTs have about 30 ohms of internal resistance in the driver transistors for the pin,
you can factor that in to your external resistor calculations.  This internal resistance depends a bit on the
supply voltage - its more for a 3.3V ATmega board.

4) There is no pulse rating for Arduino pins given.  They may or may not be able to handle > 40mA for
short periods, so until Atmel say otherwise you make the conservative assumption that the pulse
rating is also 40mA.

Actually 40mA is much higher than most logic chips can drive, 5mA is more normal, so don't be lulled
into assuming all logic signals can drive an LED at 20mA - in fact most can't (and modern high-brightness
LEDS are way too bright at 20mA anyway).

For IR remote you need as much IR as possible which is why the IR emitters are driven in pulses at
high current, and to do this a transistor to boost current is needed, such as 2N2222 or a MOSFET.
Title: Re: IR LED without resistor nano
Post by: markyj on Nov 28, 2016, 02:22 am
Ok after reading the further replies i gathered that the 5v pin provides 200ma and that i can use this pin to power the led, then use a transistor to switch the led. And of course add the resistor to the led.
My question i have is you have said that i can run 200ma through the led, but im told it is 50ma.
Here is the led i have (should really have added this in my first post)
https://www.modmypi.com/electronics/leds/super-bright-5mm-ir-led-940nm/?search=ir%20led
I also looked at the datasheet (download on link) and saw that dc forward current stated 50ma and then saw that peak forward current (1/100 pwm) was 1.2A.
So from what i've been told and read on the data sheet is the led max current via pwm is 1.2A (not going to run at that level anyway) and the 5v pin can provide a max of 200ma safely (well rated but half should be better as not to run at max). I have tried to find a definitive answer to this and have read that the 5v pin provides 200ma, 500ma and 900ma but didn't find anything officially stating it. Also i will be powering via usb using a standard usb plug (the sort used for phone chargers, not fast charge ones).
Just want to thank everyone for their help, knowledge and patience with me on this. Cheers.
Title: Re: IR LED without resistor nano
Post by: INTP on Nov 28, 2016, 03:00 am
What are the specs on your USB power source? There is no 'standard'. Actually describe yours.
Because, while you could use the Arduino's 5V pin to pull 200mA from, you will get better distance if you power the IRLED with your power source (instead of through the Arduino) which will allow pulling 1A or whatever the source has available.
If you're using some kind of phone charger, look to where it says on the sticker something like "OUTPUT DC5V ~~ 1000mA".

Power source goes separately to Arduino and IRLED, Arduino controls transistor to pulse IRLED @ <1% duty cycle. Make sure your transistor collector current is at least whatever mA your phone charger is.
Title: Re: IR LED without resistor nano
Post by: CrossRoads on Nov 28, 2016, 03:00 am
If you run that LED over 20mA continuously, it will heat up and ultimately fail prematurely.
Make sure to read the notes, like this one for 1.2A current:
1. 1/100 Duty Cycle, 10μs Pulse Width.

USB input power on Uno has 500mA rated fuse, so there's one limit.
Barrel jack has 1A rated reverse polarity protection diode, and 5V current comes from a 800mA rated regulator, which will overheat and shutdown well before that if more than 7.5V is brought in.
Title: Re: IR LED without resistor nano
Post by: markyj on Nov 28, 2016, 03:46 am
I will have to get back to you on the power supply then as i am not able to look right now.
Am i correct in thinking that 1.2A is max for running the led via pwm?
I am also thinking for my next test then is to try the led on 5v pin limited to 100ma to see the range on that. If it is fine then i think i might go with that.
Using a nano is there a pin to pull directly from usb or would that require some rewiring? Is it the vin pin that i can pull directly from usb as i read i could do that somewhere.
Title: Re: IR LED without resistor nano
Post by: Wawa on Nov 28, 2016, 04:36 am
If you want range without the high current issues, use a second IR LED.

Use the diagram from post#10.
Use two IR LEDs and a 22ohm (not 82ohm) resistor in series, powered from the 5volt pin of the Nano.
Note that the 5volt pin of a Nano is 4.6volt when powered via the USB connector.

You have linked to narrow beam LEDs (10 degrees).
Your remote will be directional.
Leo..
Title: Re: IR LED without resistor nano
Post by: CrossRoads on Nov 28, 2016, 05:09 am
1.2A, but only on for 10uS.
PWM of 1, i.e. analogWrite(pwmPin, 1) yields a 3.9mS or 3900uS wide pulse.
You're gonna fry that LED.
Run it at 20mA.
Title: Re: IR LED without resistor nano
Post by: Wawa on Nov 28, 2016, 05:23 am
This is used for a remote.
I suppose it is (or should be) pulsed 50%, so a peak LED current of <=100mA is acceptable for a 50mA LED.
I picked a 22 ohm CL resistor for two LEDs in series on a ~4.6volt supply for a reason.
Leo..
Title: Re: IR LED without resistor nano
Post by: MrAl on Nov 28, 2016, 05:40 am
Ok after reading the further replies i gathered that the 5v pin provides 200ma and that i can use this pin to power the led, then use a transistor to switch the led. And of course add the resistor to the led.
My question i have is you have said that i can run 200ma through the led, but im told it is 50ma.
Here is the led i have (should really have added this in my first post)
https://www.modmypi.com/electronics/leds/super-bright-5mm-ir-led-940nm/?search=ir%20led
I also looked at the datasheet (download on link) and saw that dc forward current stated 50ma and then saw that peak forward current (1/100 pwm) was 1.2A.
So from what i've been told and read on the data sheet is the led max current via pwm is 1.2A (not going to run at that level anyway) and the 5v pin can provide a max of 200ma safely (well rated but half should be better as not to run at max). I have tried to find a definitive answer to this and have read that the 5v pin provides 200ma, 500ma and 900ma but didn't find anything officially stating it. Also i will be powering via usb using a standard usb plug (the sort used for phone chargers, not fast charge ones).
Just want to thank everyone for their help, knowledge and patience with me on this. Cheers.
Hi,

You dont seem to be reading all the posts here.  You can get away with just using 2 i/o pins and no transistor.
Title: Re: IR LED without resistor nano
Post by: Wawa on Nov 28, 2016, 06:45 am
You can get away with just using 2 i/o pins and no transistor.
Bad practice though.
Leo..
Title: Re: IR LED without resistor nano
Post by: markyj on Nov 28, 2016, 07:25 am
Hi,

You dont seem to be reading all the posts here.  You can get away with just using 2 i/o pins and no transistor.
I did read that, the only issue with that is the library uses only pin 3 to do the switching, it's not an issue for me to use a transistor though i could still go that route for the powered circuit. I will see what works that is safe and go from there.

This is used for a remote.
I suppose it is (or should be) pulsed 50%, so a peak LED current of <=100mA is acceptable for a 50mA LED.
I picked a 22 ohm CL resistor for two LEDs in series on a ~4.6volt supply for a reason.
Leo..
Yes a remote sort of, more of an ir blaster so it will be static. I have 2 more leds that i can try with too then though wont be able to test for a couple days so will need to get back to you on that idea.
Title: Re: IR LED without resistor nano
Post by: MrAl on Nov 28, 2016, 09:30 am
I did read that, the only issue with that is the library uses only pin 3 to do the switching, it's not an issue for me to use a transistor though i could still go that route for the powered circuit. I will see what works that is safe and go from there.

Yes a remote sort of, more of an ir blaster so it will be static. I have 2 more leds that i can try with too then though wont be able to test for a couple days so will need to get back to you on that idea.
Hi again,

Oh ok well that is fine if you are happy with using the transistor circuit shown in this thread, but be aware that if you are using a library for your LED right now and that lib is running the LED straight from the i/o pin and the remote works well with the device(s), then it may not work with the transistor circuit from this thread so far.   It depends on how you have the LED connected right now without the transistor.

If you have it connected so that a low on the output pin causes the LED to turn on, then the transistor circuit will not work as is.
If you have it connected so that a high on the output causes the LED to turn on, then the transistor circuit will work as is.


So if you have the LED connected with resistor to the pin and LED cathode to ground, then it will work with the transistor circuit shown in this thread.
On the other hand if you have it connected so that the resistor is connected to +5v and LED cathode to the i/o pin, then you'll have the change the transistor circuit so that it will supply the same signal to the LED, and that means putting the LED and it's resistor in the emitter circuit of the transistor rather than the collector circuit.  Not a big deal really except for a little more voltage drop (0.7v).

Also, note that changing any library to work with 2 pins instead of just 1 is not going to be a very big mod most likely.  You just have to find where the pin 3 is being turned on and off and include another digital write instruction and possibly change the timing slightly.  Not sure if you are into that kind of mod or not but i'm sure someone here can help.
On the other hand, a transistor on the output like that is probably safer for the Arduino chip in case anything gets wired wrong or something else goes wrong.  The transistor might blow but not the 328 chip.
Title: Re: IR LED without resistor nano
Post by: Wawa on Nov 28, 2016, 10:48 am
Yes a remote sort of, more of an ir blaster so it will be static.
Same difference.
The Arduino produces a 36-38kHz square wave carrier, so the LED is 50% of the time on and 50% of the time off.
With 50% duty cycle, the IR LED can handle 2x the continuous rated current, 100mA in this case.
By using two LEDs, you have twice the "IR light" ( not twice the range, more like 1.4x).
A second "2-LED + resistor" string can be added to the same transistor for "more light".
Or add two wide-angle LEDs.
Use a ~330ohm base resistor for two LED strings.
Leo..

Title: Re: IR LED without resistor nano
Post by: INTP on Nov 28, 2016, 02:23 pm
If you want range without the high current issues, use a second IR LED.
Is that really how physics works? Two equally dim LEDs being visible from further than one equally dim LED? This would be news to me if their distance adds somehow.
Title: Re: IR LED without resistor nano
Post by: MarkT on Nov 28, 2016, 02:38 pm
I think the point was to wire them in series...  5V is enough for 3 IR LEDs I think.
Title: Re: IR LED without resistor nano
Post by: CrossRoads on Nov 28, 2016, 03:24 pm

Yes, 3 in series from 5V would work.
Title: Re: IR LED without resistor nano
Post by: runaway_pancake on Nov 28, 2016, 04:02 pm
"Two equally dim LEDs being visible from further than one equally dim LED?"

Yes.
Example: 1 LED all by itself at X lumens vs. a billboard-full of those same LEDs each doing the same X lumens
Title: Re: IR LED without resistor nano
Post by: 6v6gt on Nov 28, 2016, 04:05 pm
The OP might get some additional advice if he said exactly what he is doing and also what IR receiver he is using. There may be better alternatives, like for  long range a narrow viewing angle is better so the beam is concentrated.
Here for example is a 3 degree / 100mA IR diode http://uk.farnell.com/osram/sfh4550/led-ir-5mm-850nm/dp/1573495 (http://uk.farnell.com/osram/sfh4550/led-ir-5mm-850nm/dp/1573495)

There may also be more sensitive receivers available for his application.
Title: Re: IR LED without resistor nano
Post by: Grumpy_Mike on Nov 28, 2016, 05:26 pm
Is that really how physics works? Two equally dim LEDs being visible from further than one equally dim LED? This would be news to me if their distance adds somehow.
Read all about it. Electromagnetic radiation is additive. Read all about it.

Yes they really do add up. It is only when it comes to resolving the individual components of the radiation do you need the sensitivity to observe the EM flux from one.
Title: Re: IR LED without resistor nano
Post by: DrAzzy on Nov 28, 2016, 05:36 pm
Is that really how physics works? Two equally dim LEDs being visible from further than one equally dim LED? This would be news to me if their distance adds somehow.
Other matters aside, LED efficiency drops as the current increases - people using large LED grow lights, for example, typically run the LEDs well below their maximum power rating to reduce the power bill. So you get more light out of two LEDs at 20mA each than one identical led at 40mA. 
Title: Re: IR LED without resistor nano
Post by: pwillard on Nov 28, 2016, 07:53 pm
reminds me... More LEDS = brighter (https://www.youtube.com/watch?v=-JVqRy0sWWY)
Title: Re: IR LED without resistor nano
Post by: MrAl on Nov 29, 2016, 01:55 pm
Is that really how physics works? Two equally dim LEDs being visible from further than one equally dim LED? This would be news to me if their distance adds somehow.
Hi,

Well that may not be a fact, but it is very possible.  The other variable involved here is the DIRECTION.

For example, we have one LED facing north and one facing south.  Do their emissions add? Certainly not.  Now rotate the south one so it points east, do their emissions add now?  Still no.  Now rotate that same one a little more north, do their emissions add now?  A little, depending on the flux pattern of each LED.  If one LED points north and one north east, then the flux patterns only overlap in the direction between north and north east, which would be very roughly where the hour hand points to 1:30 on the a standard 12 hour clock.

But of course we point them both in the 'same' direction when we want to use two for an IR remote control right?  It still depends on the pattern to some degree, but we estimate the total flux to be the sum of the two which makes it seem twice as strong.  Will it reach a receiver at twice the distance with the same level of light?  It will probably be close although not exact, so maybe a little less than twice the distance or if we get lucky we do get twice, but i think it's a good estimate that we get twice the distance.

The remote i use most these days has three LEDs and it is made by Sony.  It can also read codes from other remotes (very handy).

Title: Re: IR LED without resistor nano
Post by: Grumpy_Mike on Nov 29, 2016, 02:42 pm
Quote
Well that may not be a fact,
YES IT IS.  >:(

Quote
Will it reach a receiver at twice the distance with the same level of light?
Having a higher intensity of light will not go further than the lowest intensity of light you can get which is a photon.

There is no limit on how far a photon can travel if it is not blocked.

But the brighter the light the more photons arrive at given distance and so the easier it is to detect. Light doesn't just go so far and then stop.
Title: Re: IR LED without resistor nano
Post by: MrAl on Nov 30, 2016, 02:06 pm
YES IT IS.  >:(
Having a higher intensity of light will not go further than the lowest intensity of light you can get which is a photon.

There is no limit on how far a photon can travel if it is not blocked.

But the brighter the light the more photons arrive at given distance and so the easier it is to detect. Light doesn't just go so far and then stop.
NO IT IS NOT :-)

You should have read the rest of the post.

The light direction has to be taken into account, as well as the phase if they are of the same exact wavelength.
I gave a simple example where the two light sources were facing away from each other, there's no way their light can add.
Also, there are experiments that depend on the phase difference even though they are both pointing in the same direction.

Two facing in the same direction would look something like this:
)~~~~~~~  --->
)~~~~~~~  --->

and for our purposes they may add, but these two can NOT add:
<---  ~~~~~~(     )~~~~~~  --->

because they are facing entirely different directions.

The actual addition would come from looking at their radiation pattern and how those patterns overlap.  If both have a bright spot in the middle but they are pointing in slightly different directions, then there is a distance D where only their outer rim patterns overlap so the full intensity will not be had.

Two LEDs can be seen from a farther distance than one because there are more photons in the same area with two instead of one.  That adds to the POWER found at the receiver end, which of course can activate a sensor better.  It does not matter that both LED photons go the same distance, it is that when they do get there there are more of them to act on the sensor and thus provide more of a signal.  This means sometimes we say "the light goes farther" even though the photons dont, just that the influence is stronger.
In other words, the remote works at a greater distance from the sensor.
Title: Re: IR LED without resistor nano
Post by: Grumpy_Mike on Nov 30, 2016, 03:42 pm
Oh dear we are at it again MrAL are we. You have a track record of not understanding things, but thinking you do.

Quote
You should have read the rest of the post.
I did, painful as it was. I am glad I never taught you physics, you would have failed.

Quote
The light direction has to be taken into account, as well as the phase if they are of the same exact wavelength.
No. The phase is only important if the light sources are coherent, in an LED they are not. And the direction is irrelevant. Just assume an isotropic radiator and an emission intensity adjusted to compensate.
Quote
I gave a simple example where the two light sources were facing away from each other, there's no way their light can add.
Yep and a pretty stupid example it was. You can not get any light source that does not have some radiation in some direction, only in your unrealistically simplistic world can this happen, in real life it does not. But there was no point in this example anyway it did not advance the argument.

Basically it all comes down to the derivation of the inverse square law and the area on the sphere of radiation corresponding to the area of the sensing element receiving enough photons to get a sufficiently large signal to noise ratio to allow reliable detection.

Quote
and for our purposes they may add, but these two can NOT add:
<---  ~~~~~~(     )~~~~~~  --->

because they are facing entirely different directions.
That would be true if there was such a thing as a light source that only radiated in one direction. They do not exist.

Quote
This means sometimes we say "the light goes farther"
Yes you say it, but it is total bollocks from a physics point of view, which is what was being expressed in the question I responded to, which was:-
Quote
Is that really how physics works? Two equally dim LEDs being visible from further than one equally dim LED? This would be news to me if their distance adds somehow.
I hope you agree that the answer is yes they do add, and if you define "the light goes further" in a totally none physics way of meaning "the distance limit of sensing".
Title: Re: IR LED without resistor nano
Post by: Smajdalf on Nov 30, 2016, 07:17 pm
I am glad I never taught you physics, you would have failed.
I am glad you never taught me physics and poor children if you teach any. You can define the vague statement "light of two sources add under any conditions" in such way that it is true. But in any reasonable and useful sense it is not true. You say there is no such thing as directional source of light. What about a LED on a meter thick concrete? I guess it will take many years until the first photon from the LED passes the barrier. You want to say two LEDs separated with the concrete adds their intensity in any sense? What about LEDs in other galaxies? But we do not need to go so far ofc. It is common feature that LEDs produce relevant amount of light in limited angle. It is needed to take the different spatial intensity into account when estimating total light produced in given direction just as MrAl said.
Title: Re: IR LED without resistor nano
Post by: Grumpy_Mike on Nov 30, 2016, 07:27 pm
Quote
What about a LED on a meter thick concrete?
Read reply #36, I said
Quote
There is no limit on how far a photon can travel if it is not blocked.
Get the end bit - NOT BLOCKED

Quote
I am glad you never taught me physics and poor children if you teach any.
I never taught Physics to children, just grownups.
Title: Re: IR LED without resistor nano
Post by: ad2049q on Nov 30, 2016, 09:03 pm
BAD idea running an IR emitter direct from a 5V device like the Arduino.  The "natural" junction voltage of an IR emitter is 1 to 2 Volts when driven hard, so it would try to drop the other half of the Voltage in heating up an arduino digital out.  I prefer the approach of outsider, though 2 or 3 volts to drive the IR is likely to be better than 5.

Choose a transistor Q1 which is more than capable of running the IR at its maximum continuous rated current (for example 5mA), divide by the gain of Q1 (for example 120) so you want about 0.04 mA into the base when arduino digital out is ON (=5V).  If in doubt, 100kOhm from digital out to base is often sensible, though for some IR devices you'd need a different value.

Next measure or look up the voltage across the IR at its max.  If in doubt expect 1.5 V (it could be a little different).  How much resistance should go above Q1 to "use up" the remainder of the external supply voltage ?  Suppose that you have 3.0 Volts external supply.  1.5V / 5 mA = 300 Ohms in place of the 68 in the circuit of outsider.

You now have an external circuit to drive an IR emitter at its maximum (say 5mA) while the base resistor (say 100kOhm) gets +5V. You can do physical testing with just that on pinboard without particularly needing the arduino.  The datasheet of the IR tells you the highest current that you may use it at reliably, and you have briefly got something to work by overdriving it.  So you want more brightness to arrive at the detector.  Possible solutions include :
- duplicate the IR transmitter to get more brightness.  You could have a dozen of them with the kind of external driver of outsider.
- put a lens in front of the IR transmitter to send more of its output in the direction of the receiver [warning! might not be eyesafe]
- put a plastic optical fibre or comparable light guide in between transmitter and receiver.

Be exceedingly cautious with eye safety if thinking of using IR devices more powerful than 1mA and if in any doubt discuss with your supervisor.
Title: Re: IR LED without resistor nano
Post by: Wawa on Nov 30, 2016, 11:31 pm
Every IR (TV) remote I know pushes 100mA or more through the IR LED(s).

The old Philips ones used four LEDs in series on a 9volt battery.
12meters was easy, and if you aimed the remote carefully, it could do 60meters (done that).
Leo..
Title: Re: IR LED without resistor nano
Post by: MrAl on Dec 02, 2016, 08:47 am
Oh dear we are at it again MrAL are we. You have a track record of not understanding things, but thinking you do.
I did, painful as it was. I am glad I never taught you physics, you would have failed.
No. The phase is only important if the light sources are coherent, in an LED they are not. And the direction is irrelevant. Just assume an isotropic radiator and an emission intensity adjusted to compensate.Yep and a pretty stupid example it was. You can not get any light source that does not have some radiation in some direction, only in your unrealistically simplistic world can this happen, in real life it does not. But there was no point in this example anyway it did not advance the argument.

Basically it all comes down to the derivation of the inverse square law and the area on the sphere of radiation corresponding to the area of the sensing element receiving enough photons to get a sufficiently large signal to noise ratio to allow reliable detection.
That would be true if there was such a thing as a light source that only radiated in one direction. They do not exist.
Yes you say it, but it is total bollocks from a physics point of view, which is what was being expressed in the question I responded to, which was:- I hope you agree that the answer is yes they do add, and if you define "the light goes further" in a totally none physics way of meaning "the distance limit of sensing".

Hello again,


What you are doing is going general when i go specific, then going specific when i go general, and in that way you can say that everything i say is wrong.

So i wont waste my time with you unless you would want to argue (in a debating manner without your mocking tone) one point at a time.

Since the main point is one of distance and the number of LEDs and how they add, i'll reply with that in mind.

Facts:
1.  Light adds based on the number of sources and and the direction of each source.  An extreme example is two LEDs that point 180 degrees away from each other.  As we rotate one LED toward the same direction of the other LED, the light would start to add more and more, creating "more light" at the receiver.
2.  As the light adds once we get the direction optimal, the distance the light appears to travel (called the "Throw" in flashlight terminology) increases, and thus the distance the remote can be used at increases.
3.  The distance light appears to travel is based on the human eye response and there is even a name for it, the "Throw".  The 'Throw" in our case is the distance the remote can successfully operate the receiver reliably.  Yes, photons travel forever unless something blocks them, but the behavior based on the human common experience morphs this into a more tangible definition that every human can relate to without counting EVERY SINGLE photon.  It's the same thing humans do with every experience we have outside of the laboratory: a de facto statistical interpretation of a large number of quantum events.  Like it or not, that's the way it is.