Simple LED modulation driver c

Thread Starter

wilfr3d

Joined Nov 28, 2017
5
Hi I am doing a project on High speed modulation of LED's and I was just wondering if anyone has a very simple circuit design for the this experiment.
I saw another circuit online (attached below) and I made it in the lab but it seems to be giving me wrong results. I am changing the frequency on the function generator gradually from 1Hz all the way up to 25 MHz. Channel one of the oscilloscope was attached to the input signal from the FG and channel two of the oscilloscope was attached to one leg of the LED to measure the output. Every time I changed the frequency on the function generator, I was measuring the phase shift between channel 1 and channel 2 of the oscilloscope and also their amplitudes. Also I noticed that, after around 800KHz on the function generator, the results just doesn't make sense at all. Any help would be appreciated of a very simple and basic circuit to modulate an LED. Thank you

upload_2018-4-20_20-39-42.png
I have shown this circuit to my supervisor and he said its too complicated for just modulating LED's and he would prefer a much simpler circuit.
 

Attachments

MisterBill2

Joined Jan 23, 2018
3,884
What sort of modulation is the instructor asking for? Amplitude modulation, or pulse-width modulation? OR possibly just plain OFF/ON modulation? I do wonder about the emitter circuit wit ten ohms in series with 3000 Mfd. The resistor defeats the purpose of the capacitor. Likewise on the input. But aside from that it is hard to imagine a simpler circuit.
 

Thread Starter

wilfr3d

Joined Nov 28, 2017
5
Thanks so much for the reply. I believe he is after the Amplitude Modulation because he wants me to find out about the voltage drop across the led during modulation at different frequencies from the function generator. He suggested to use the oscilloscope...channel 1 attached to one leg of the LED and channel 2 attached to the other leg of the LED, then subtract the 2 waveforms using the oscilloscope's math function to get the waveform of voltage drop across the LED. The next step will be to attached a different channel of the oscilloscope to the input signal (on the circuit above, it will be on the left hand side of the 150 ohm resistor) to get the input waveform and compare it with the waveform across the LED in terms of phase shift and to see the difference in the amplitudes. I have to do this for each time I change the frequency on the function generator until I reach a point to where the LED will not work at high frequencies. The supervisor is not very sure about my circuit since I am getting strange results so he suggested that I need to find a much simpler circuit than the one above.

upload_2018-4-21_0-8-18.png

I have found this circuit online for driving an LED which is simpler but I am not sure whether it will work or not
 

MisterBill2

Joined Jan 23, 2018
3,884
The voltage drop across an LED does change a small amount as the current changes quite a bit. There is another discussion about exactly that fact. Attempting to measure the voltage across the LED will certainly be subject to a lot of noise because the changes in voltage are very small compared to the changes in current. I suggest putting the LED in series with the the emitter , perhaps leaving the 220 ohm resistor in the collector circuit. That would allow measuring the LED voltage with the input common grounded, giving less noise.
ALSO, try using the AC coupling mode on the scope input. That will avoid the much larger common mode voltage, which can be confusing.
 

Thread Starter

wilfr3d

Joined Nov 28, 2017
5
Once again thanks so much for the reply. I will create the circuit as you suggested on Monday and do the measurements again. Like you said about small changes in voltage across the LED as the frequency is changed on the FG, when I used the multimeter to measure the voltage across the LED at difference frequencies using the circuit in the question above, the difference between the voltages across the LED was very small. Its only that my supervisor said he wants to see the voltage across the LED in terms of a waveform from the oscilloscope not just the value using the multimeter. I will keep you updated on Monday. I have swapped the LED to the emitter side of the transistor like you suggested as shown below. Thank you

upload_2018-4-21_10-9-6.png
 

Thread Starter

wilfr3d

Joined Nov 28, 2017
5
Also, one quick question please. Does driving an LED to turn it on means the LED is being modulated? Or the whole process of LED modulation is based on changing the frequency of the function generator which is connected to the input of the circuit. Thanks
 

MrAl

Joined Jun 17, 2014
6,501
Also, one quick question please. Does driving an LED to turn it on means the LED is being modulated? Or the whole process of LED modulation is based on changing the frequency of the function generator which is connected to the input of the circuit. Thanks
Hi,

Turning the LED on and off would be amplitude modulation (AM).
Varying the frequency would be called frequency modulation (FM).

In both cases the LED brightness would be modulated, it's just a different type of modulation for each.

FM works better than AM when there will be natural causes for changes in the amplitude that are strong enough to change the signal significantly.
 

crutschow

Joined Mar 14, 2008
23,366
LEDs often use digital pulse modulation (fully on and off) for transmitting data since its easier to do and more accurate than amplitude modulation.
 

Thread Starter

wilfr3d

Joined Nov 28, 2017
5
Hi thank you all for the help. Just a quick update about my project. I made the simple circuit above in the laboratory and started modulating the LED changing the frequency from the function generator starting from 1Hz going all the way up to 25MHz. During modulation at difference frequencies, the oscilloscope was measuring the amplitude voltage of the input signal waveform from the function generator and the amplitude voltage of the output waveform from the LED. Also, the oscilloscope was measuring the phase shift between the input waveform and output waveform.

The second step was to calculate the amplitude voltage ratio between the input and output waveform and I used the formula below;
Amplitude ratio = output amplitude/input amplitude

Afterwards I had to change the amplitude ratio into dB using the formula below
20 log (amplitude ratio)
the log is in base 10

The last step was to plot the frequency response (dB) against the frequency and from this graph, and correspondence frequency from the graph when dB is -3dB is the cut off frequency of the LED.
upload_2018-4-29_15-9-6.png

For some reasons, all my dB values were below -3dB so from my graph, it doesn't seem like I have a cut off frequency which is impossible. I am going to attach my table of values and my graph below. All I need to find out now is the cut off frequency of the LED during modulation. Thank you all again for your help and I really appreciate it.

upload_2018-4-29_15-1-41.png

upload_2018-4-29_15-2-13.png
 

MrAl

Joined Jun 17, 2014
6,501
Hello,

Just wondering, what are you measuring to get the LED amplitude?
Is it the voltage across the LED? If so, what is that telling you?

Not sure what you are doing exactly but it seems that you would want to measure the light output of the LED not the voltage across it. Alternately maybe the current through the LED with the assumption that it correlates directly to LED light output. Maybe both actually to get the change in LED efficacy.
Not entirely sure what you are trying to do though.
 
Top