Optical wireless audio transmitter

Dehaka

Joined Jul 27, 2017
4
Good day

We are designing an audio system capable of sending audio through a transmitter using Frequency modulation and receiving it at a
min of 1.5m away. We where instructed to choose between white,red or infra red LEDs. We had to keep in mind the wavelength of the audio signal to be transmitted (between 10kHz and 50kHz) also the sensitivity to the different types of LEDs that the phototransitor will pick up.But this week we have to design our LED transmision circuit and phototransistor receiver circuit and we have to choose our carrier frequency(32.768/40/75/77.5 kHz) by doing a simple experiment. We are instructed as follows :
1)How will you drive the LED? 2) Will the circuit work both in daylight and in complete darkness?  3)How will you maximise the range? 4) Based on your carrier frequency and bandwidth calculations in part 1 above, how important are the rise and fall times of the phototransistor? Can you do something to improve them?  5) Determine the required currents and voltages to get the components into their desired operating regions.

We really would like to have some insight to these questions before we submit our own report, to check if we go about this the rigth way.

Greetings

#12

Joined Nov 30, 2010
18,223
First, this looks like school work. Second, parts of this problem have already been solved in the remote controls for your TV.
Look to the obvious before you re-invent the wheel.
This assignment is an exercise in what's already been done. That's why it looks like Homework.

AnalogKid

Joined Aug 1, 2013
9,498
1. None of those carrier frequencies will work. See FM modulation 101.
2. Edit - removed.
3. Technically, those carrier frequencies can be made to work. However, it increases the complexity of the transmitter and receiver.

ak

Last edited:

Dehaka

Joined Jul 27, 2017
4
Good Evening
Yes this is a project we need to build, or design. It is so we can learn more about how this field of work. Don't get me wrong I'm not asking for the answers I just want someone to point me in the right direction where to look for answers or what to look for specific. We were giving different frequency cristals which will be use to build an on board oscillator to generate the carrier frequency, we can create a frequency multiplier off those frequencies as well. But like I said they just gave us the project and said build it. I started by testing the effect of different lights on the phototransistor and then how the effect changes if I use different carrier frequencies. But I don't really understand how we are going to build the transmitter or how the modulating scheme(FM) works or infact how to build the circuit. So if anyone can maybe point me to material i can use to help myself would be kind.

Greetings

AnalogKid

Joined Aug 1, 2013
9,498
Where are you located?

What kind of school is it?

What class requires that you build an FM transmitter without knowing how FM works?

ak

Dehaka

Joined Jul 27, 2017
4
Good morning

I'm currently studying E&E at stellenboch university 3de year, South Africa, and we did a brief over view of FM in system and signals 315, but that did not go so well.

Greetings

AnalogKid

Joined Aug 1, 2013
9,498
we did a brief over view of FM in system and signals 315, but that did not go so well.
Then you're in for a rough time. I'm not picking on you or anything like that, but FM is not simple, and audio quality FM is really not simple.

To start with, look into the Nyquist frequency and the Shannon Sampling Theorem. FM can be thought of as a sampled data system. At any moment, the instantaneous value of the (audio) input voltage is translated to the instantaneous value of the carrier frequency. At the receiver, the instantaneous value of the carrier frequency is translated back into the instantaneous value of the output voltage. The higher the carrier frequency is relative to the modulating (audio) signal bandwidth, the easier it is for the receiver to *discriminate* the frequency changes accurately.

Your carrier frequencies all are less than twice the maximum audio frequency, breaking rule #1 of communications theory. As I said above, there are ways around this, but they make everything more complex. For 50 kHz audio (why so high?), what about a carrier somewhere in the 250 kHz - 500 kHz range? Low enough to see with a low cost scope, low enough that super-critical RF design and assembly techniques are not required, but high enough that the recovered signal won't suck.

ak

#12

Joined Nov 30, 2010
18,223
For 50 kHz audio (why so high?),
There's one I didn't lock on. Who said audio is 50KHz?
The last time I checked, humans can not detect above 20KHz, and 16KHz is generally sufficient.
Are you working for bats or dogs audio range?

crutschow

Joined Mar 14, 2008
28,215
Building a crystal controlled VCO for the FM transmitter is far from a trivial task .
It rather sounds like whoever generated the requirements has never done this himself.

DickCappels

Joined Aug 21, 2008
7,989
If they severly restrict the bandwidth the bandwidth to say, 2.5 kHz they can get some intelligible speech across to the receiver.

AK, They are not to use lasers, but LEDs.

They could make a crystal controlled FM transmitter with a CD4046 and a crystal reference, but crystal control is not a requirement but something we inferred by the frequency mentioned. It might be that they can get by just using a two transistor multivibrator and modulate it by modulating the voltage to the base resistors.

(some text removed for clarity)
1) How will you drive the LED?
2) Will the circuit work both in daylight and in complete darkness?
3) How will you maximise the range?
4) Based on your carrier frequency and bandwidth calculations in part 1 above, how important are the rise and fall times of the phototransistor? Can you do something to improve them?
5) Determine the required currents and voltages to get the components into their desired operating regions.
I suggest you start with the available photodetectors and work through to the speaker, then having done that work out the receiving optics and the transmitter LEDs and optics. After that you can answer the question about how to drive the LEDs and all of the other details.

As #12 said in post #2Take some time to read about infrared remote controls and in particular how the IR receiver ICs work.

AnalogKid

Joined Aug 1, 2013
9,498
AK, They are not to use lasers, but LEDs.
I yield to your interpretation. I read it as 1.5 *miles* minimum. I edited my response.

ak

MrAl

Joined Jun 17, 2014
8,556
Hi,

Some questions came to mind when i read the first post here.

1. Do you really need 50kHz for audio or or can it be limited to 20kHz?
2. Can you use a pin diode instead of a photo transistor? Photo transistors are usually much slower. Pin diodes can go very high.
3. If you dont need super duper frequency stability, you can ever use an LM339 VCO to generate the FM signal.
4. To get best range, use at least one lens, and for the receiver use a light tube made from cardboard and painted flat black in the inside with the receiving device at the far end. The tube should be at least three inches long if possible, and should be thiick enough to not let any light in from angles larger than maybe about 10 degrees from the focal axis.

DickCappels

Joined Aug 21, 2008
7,989
I read it as 1.5 *miles* minimum. I edited my response. ak
Thank you for mentioning it, until I just went back and reread it I also interpreted it as 1.5 miles which would be a immense challenge with LEDs. 1.5 meters -they have a chance.

Dehaka

Joined Jul 27, 2017
4
I have attached the circuit we are going to use. Yes it is 1.5m that we need transfer the audio across using the LED's. I made a mistake the audio we want to transmit is between 10kHz and 20kHz. As you can se in the circuit diagram we have the transmitter using Infrared LED and the receiver using a Phototransistor(bpw77na). Then on the receiver side we are building the circuit in 4 parts, the phototransistor circuit, the high-pass filter(which is the combination of the Cc capacitor and the Rf resistor), the Vvg (which is the virtual ground, we just build this to get our ouput of the phototransistor circuit to swing around 2.5Volts) and last the FM-demodulator(which we are using a PLL(phase-lock-loop) combined with a VCO and a low-pass filter).

I would just like to know how would I go about designing the Rbe and Rc resistor for the phototransistor circuit?
And how must i design my High-pass-filter(Cc and Rf). What I have done is the following:

We are using the transmitter transistor as a switch, so the phototransistor is either going to be on or saturated. Now I don't know how to calculated Rc or Rbe, because what is the Ic current of beta of the phototransistor? I have looked in the datasheet but can't find it. Second I did try taking the max Ic curent of the phototransistor and using the formula : Vcc - Ic(max)Rc - Vce(sat) = 0 : I get Rc = 47Ω but when i tested it with my demo circuit using infrared LED I only got like 5.1mA for Ic. I don't know if my calculation are right or if I must use a larger resistor value?

Second what experiment can I do to determine how large my Rbe should be? I build the circuit without the Rbe and found that in ambient light there is a voltage drop of 0.05V across the 47Ω Rc resistor which means that Ic = 1,06mA. Now normaly I would devide this by beta to get Ib(current in the base) and do a KVL to get the value of the Rbe resistor.

If anyone can help in anyway I would appreciate it.

Attachments

• 533.1 KB Views: 5