IR LED Powering Circuit - Design Viability Check Needed

[milobindi]

I have been tasked with designing and building a device that will use a .4V Peak to Peak amplitude voltage source to power 5 IR LED's in parallel. I have attached my design schematic, which includes a low pass filter amplifier with a cutoff frequency of 120Hz and a gain of 101 (though the opamp rails will saturate at +/- 9V), a full wave bridge rectifier made of 4 diodes, a smoothing capacitor, and a current limiting resistor. My question is: will the design pictured deliver the proper voltage and current to the 5 IR LED load? Each IR LED needs between 20 and 100 mV of current and drops aprox. 2V. Do I need a resistor before the rectifier? Is my smoothing capacitor the proper value? I it is a new concept to me. Thank you so much for your help, I really appreciate it.

 

[Nigel Goodwin]

The entire premise seems confusing?, I've no idea exactly what you're trying to do, or why?.

However, you can't just stick LED's in parallel - one will take all the current.

 

[alec_t]

Welcome to ETO!

I have been tasked with designing and building a device that will use a .4V Peak to Peak amplitude voltage source to power 5 IR LED's in parallel.

In your circuit the LEDs are being powered by the 9V supply for the opamp; not by the 0.4V input. Are you sure you have understood the task correctly?

 

[milobindi]

The entire premise seems confusing?, I've no idea exactly what you're trying to do, or why?.

However, you can't just stick LED's in parallel - one will take all the current.

Thank you! I just did some research on that and there were suggestions to put a resistor in front of each individual LED instead, would that work? I was thinking a 56 ohm resistor for each. I really appreciate your help. The premise is that we have a camera for motion tracking that only detects the motion tracker markers and IR light. We also have a device that emits the .4V AC voltage when on, and we need a circuit that will use that output from the device to light up the LEDs when the device is in use so that there is a sort of IR indicator light for the camera when we take videos. There is no other way for use to know when exactly during our recordings the device was on. I hope that helps to clarify a bit.

 

[milobindi]

Welcome to ETO!

In your circuit the LEDs are being powered by the 9V supply for the opamp; not by the 0.4V input. Are you sure you have understood the task correctly?

Not quite, the 9V batteries will power the positive and negative opamp rails. The circuit will only function when the device with the .4V AC input is on because it requires an AC source for the capacitors. Therefore the LED's will only light up when the .4V AC voltage input exists, which is what we want since we are trying to use this circuit as a sort of IR indicator light for our IR sensitive video recording device.

 

[JimB]

Ignoring any other problems with the circuit, you need to remove the "earth" connection which I have highlighted with a red circle.
It is effectively shorting out one of the diodes, and is not required in this circuit configuration.

JimB

 

[AnalogKid]

From your description in post #5, the input signal is *controlling* the LEDs, not powering them. This is an important distinction. Knowing which parts of your system are power supplies and which parts are not is critical to explaining what you want and understanding how things work.

Also, the LEDs need 100 mA, not mV. Amperes and Volts are different. And, at 100 mA per LED you will need a 0.5 A poser source and a medium-sized power transistor to drive the LEDs. The vast majority of opamps cannot make a 0.5 A output current.

Is it a requirement that the LEDs flicker at the same frequency as the input signal, or can they be on steady-state? If the latter, since the input is an on-off signal, all you really need is a comparator as a missing-pulse detector driving a power transistor. The resulting circuit can run on just the +9 V supply if you want.

ak

 

[alec_t]

The non-inverting input of the opamp needs an appropriate bias circuit. Does the AC input signal generator provide a DC bias current path?
Even at 20mA per LED, most opamps would be unable to provide the 100mA total current.
Is your 9V supply mains-derived or a PP3 battery? The latter would be flat in next to no time.

 

[ronsimpson]

Using the +/-9V supply
Choose voltage comparitor good for more than 20V supply and can drive the LED current. 20mA or 100mA that is the question.
LEDs in series to save power. Five, 2V LEDs use up 10V. Current set by R7. In this example about 20mA.
Use LT1018, I used the 1017 in SPICE.
There are two points; +0.1V and -0.1V. Change the 1K resistor to get the voltage you want. Now set to have the LEDs on for any signal above or below 0.1V.
LR1018 is dual 8 pin IC. One half watches for voltage above 0.1 and the other watches for voltage below -0.1V. The outputs are ORed together.
LED current is the same for all LEDs. They share the same current.

 

[ronsimpson]

Input is 50hz +/-300mV signal. ID2 is the LED current at 19mA.
If the input was +/- 99mV then there will be no LED current.

 

[ronsimpson]

There is a question on input filter.
I used a 10k resistor and 0.1uf to get a smaller capacitor.
-3db point at 160hz approx.

 

[audioguru]

The forward voltage of an IR LED is a lot less than 2V, probably 1.2V. Then your 330 ohms current-limiting resistor will use much more battery power than the five series IR LEDs.

 

[ronsimpson]

The forward voltage of an IR LED is a lot less than 2V, probably 1.2V.

From post #1." about 2V"
I agree that 2V is high.
If five LEDs can be powered from the 9V supply then the whole circuit could run from 0 and +9v only. No -9v required.

 

[AnalogKid]

Or a series-parallel arrangement of 2- and 3-LEDs (with different current limiting resistors). If reaction time is not critical, you can do it with a single comparator detecting only the positive peaks.

ak

 

[ronsimpson]

If reaction time is not critical, you can do it with a single comparator detecting only the positive peaks.

You know there is going to be a complaint about the LED not being on 100% of the time.
A dual comparator or a single in one 8 pin IC .... about the same complexity.

Running 1.2 to 1.6V LEDs in parallel ..... 100mA....this limits what IC can be used. pulls too much power.
Parallel/series or just series reduces the current greatly.

 

[AnalogKid]

The LED is not on 100% of the time with a dual comparator, either; it just flickers twice as fast. In either case, a capacitor from <out> to GND solves it.

ak