IR transmission power controlling

Maybe this is a big topic, I will explain my problem.

(Reworded)

If you want to adjust the distance IR communication can reach, you will face this problem: transmission power adjustment.

That means we need to control the current through the IR led - a tsal6100 for my example. As a software engineer what I can think of is dc-dc switch power solution. I found AL8861 and composed the circuit based on datasheet info. My question is: can this circuit work?

Why I asked this question? According to my understanding, for the nature of switch power, there are ripples on the output,usually few handreds mA, this will lead to the transmission distance vary too?

（Appended on 5/1)

Transistor is right,we should use analog dimming instead of pwm dimming.for the programmable resister is a little expensive,I planned to create my own one:

The unit of resist is K , try to get a series of voltages allocated from 0.3 to 2.5V .(the analog dimming range of Vset defined in datasheet of AL8861.)

（Appended 5/1)

If build base on LDO (ZLDO1117),this is right?

In order to get rid of influence of resist from CD4052B,I should use big resister (in Ks), is that ok?

• Since you'll need to modulate the signal anyway, you may be able to make the pulse width narrower and narrower while keeping the pulse frequency at the center of the receiver passband. – Chris Stratton Apr 27 at 5:13
• What are you actually trying to solve? If you want somebody 10 m away to be able to receive your signal, but a snooper 12 m away (for example) shouldn't be able to, that's not going to be practically possible. Also notice the absolute maximum peak forward current is specified as 200 mA, so not all of the curve you posted is usable. – The Photon Apr 27 at 5:34
• Of course,we try to adjust during it's available range.the max current is 200ma,then the range maybe is 100ma to 200ma,how to? – Xiao Apr 27 at 6:32
• You ALWAYS need to control the current through ANY sort of LED. The driver in your diagram does this by sensing the voltage across Rs and hence the current through the LED. I'm struggling to understand your actual question. – Finbarr Apr 27 at 7:16
• Regarding your update, you wouldn't mix PWM (full on-off) dimming with a modulated signal as you would generate spurious signals. You need to control the current limit by analog current limiter (a resistor might suffice) so that the only modulation on the LED is the code you are transmitting. – Transistor Apr 30 at 15:24

If you want to adjust the distance IR communication can reach, you will face this problem: transmission power adjustment.

Generally power is set at design stage to be adequate for the maximum transmission distance and is not adjusted in use. The defined range does not have a precise cut-off and, as anyone who has used a TV remote will know, factors such as beam spread, reflections from nearby surfaces, etc., are as important as distance.

That means we need to control the current through the IR led - a tsal6100 for my example.

LEDs (almost) always require current control due to their exponential relationship between current and the applied voltage.

As a software engineer what I can think of is dc-dc switch power solution. I found AL8861 and composed the circuit based on datasheet info. My question is: can this circuit work?

Figure 1. OP's proposed solution.

Your schematic has a couple of problems. Page 12 of the datasheet states

Open Circuit LEDs

The AL8861 has by default open LED protection. If the LEDs should become open circuit the AL8861 will stop oscillating; the ISENSE pin will rise to VIN and the LX pin will then fall to GND. No excessive voltages will be seen by the AL8861.

I expect that there will be a recovery time associated with this shutdown. The second problem is that there is an inductor in the circuit. Inductors don't like you switching off the current in an instant and will generate very high voltages ($$\ V = \frac {dI}{dt} \$$) when you try it.

Why I asked this question? According to my understanding, for the nature of switch power, there are ripples on the output, usually few hundreds mA, this will lead to the transmission distance vary too?

Figure 2. Current versus time.

Note that the period of the current waveform is about 1.7 μs giving a frequency of almost 600 kHz. From memory, most IR transmission systems (as in remote controls) use 38 kHz so variations in light intensity will be averaged out in the receiver as the carrier frequency is so much higher than the modulation frequency.

Other than that, it's a bit hard to say as you haven't said what you are transmitting, how far and at what frequency.

• I must thank you for 100 times:),you are so kind to spend so much time to explain. 1) In my case I have to reserve a method to adjust the output current in the future,I can't define the use case later. 2) you mean the tripple will not influence the transmission distance? 3) the led open issue,it looks like a deadly bug,any other choice? I use 38k pwm to transfer info,20meters at least. – Xiao May 1 at 2:12
• Actually, I'm trying to build a laserwar game system. Instead of laser I use IR like most of other people. The frequency will be 38/56k,the distance is a question,now,I am trying to get 20 meters transmission distance with spot diameter less than 50cm. But,if you play in house,20 meters is too far away, And if you want to define a sniper rifle ,20 meters is a little near. So, we'd better reserve a parameter to define this. – Xiao May 1 at 2:29
• Info transfered: 16 bit for 1 shot. – Xiao May 1 at 2:30
• All led driver come with open protection,all. Time to think about build from ground:( – Xiao May 1 at 7:14