Will this synthesizer work?

[j4cobgarby]

I've attached an image of the schematic for a synth I'm designing.
The basic principle is that I have three 555 timers, each generating a square wave whose frequency is controlled by the resistors R1, R9, and R13. Each timer's output can optionally be converted to a sin wave or left as a square wave based on a switch on the output of each timer. The three different timer outputs now are summed together at U2, but are weighted according to R5, 6, and 7. The resultant waveform is then sent through the speaker at the bottom. The whole thing is powered by a 9V DC jack.

I have two main questions:

• Are the 555 timers wired up correctly? This is my first time using them like this.
• Will there be a problem with audio impedance at the speaker? I have little understanding of impedance in this context.

Thank you in advance!

 

[BobK]

You cannot connect the discharge pin to Vcc, this is a short when discharge is low. There may be other problems.

And, no, you cannot drive a speaker directly from an op amp. You need an audio power amp.

Bob

 

[j4cobgarby]

You cannot connect the discharge pin to Vcc, this is a short when discharge is low. There may be other problems.

And, no, you cannot drive a speaker directly from an op amp. You need an audio power amp.

Bob

Why can I not connect discharge to Vcc? How else will I get a 50% duty cycle? Or.. Just a low value resistor instead of shorting it, perhaps?

 

[BobK]

Why can I not connect discharge to Vcc?

Because it shorts the power supply.

How else will I get a 50% duty cycle? Or.. Just a low value resistor instead of shorting it, perhaps?

You cannot get a 50% duty cycle with the standard astable circuit. Connect the pot to the output pin to get a 50% duty cycle.

Bob

 

[Audioguru]

Please look on the datasheet of a 555 how to make an oscillator.The rectangular output of the 555 sounds awful like a buzzer.
Your filters are so simple that they do nothing at low frequencies and reduce the levels at high frequencies before the waveform is near a sinewave.
You are using adjustable series resistors instead of using them wired as volume controls, and they should be log type, not linear type.
Your speaker is shorting the output to ground. A proper power amplifier would feed the speaker through a coupling capacitor to block the DC at the amplifier.

The opamp has positive feedback instead of negative feedback.

 

[Audioguru]

A Cmos 555 can produce a perfect squarewave with the simple circuit using the output pin.

 

[j4cobgarby]

Please look on the datasheet of a 555 how to make an oscillator.The rectangular output of the 555 sounds awful like a buzzer.
Your filters are so simple that they do nothing at low frequencies and reduce the levels at high frequencies before the waveform is near a sinewave.
You are using adjustable series resistors instead of using them wired as volume controls, and they should be log type, not linear type.
Your speaker is shorting the output to ground. A proper power amplifier would feed the speaker through a coupling capacitor to block the DC at the amplifier.

The opamp has positive feedback instead of negative feedback.

How, then, would you suggest converting the square waves to sine waves?

 

[Audioguru]

Years ago I converted a high frequency squarewave to an oversampled sinewave. Instead of the 16 stage shift register shown here, I used a simple 5 stage CD4018 counter with 5 resistors creating 10 steps. One of the counter outputs is the squarewave and the counter is fed from a 10 x the frequency variable clock oscillator. I also used a switched capacitor lowpass filter IC that had its frequency changed by the variable clock oscillator. The sinewave was good at all frequencies and its level never changed.

 

[BobK]

For a synthesizer, you really don’t want sine waves. They sound absollutely boring, not like any instrument because they have no harmonics. A softened square wave, i.e. with a low pass filter to round the edges, sounds vaguely like a woodwind. A sawtooth sounds vaguely like a bowed string.

Also, the envelope is critical to get sounds that are unstrument like.

Bob

 

[Audioguru]

Buzz, buzz, beep,beep is not a synthesizer.

 

[j4cobgarby]

For a synthesizer, you really don’t want sine waves. They sound absollutely boring, not like any instrument because they have no harmonics. A softened square wave, i.e. with a low pass filter to round the edges, sounds vaguely like a woodwind. A sawtooth sounds vaguely like a bowed string.

Also, the envelope is critical to get sounds that are unstrument like.

Bob

How would I make an envelope for the sound?

 

[BobK]

You would need a lot more circuitry than you currently have.

It all depends on what you are trying to achieve,

There is plenty written about analog synthesis out there.

Personally, if I wanted to make a replica of the old Moog or ARP type of synthesizer, I would do it digitally with a microcontroller.

Bob

 

[j4cobgarby]

You would need a lot more circuitry than you currently have.

It all depends on what you are trying to achieve,

There is plenty written about analog synthesis out there.

Personally, if I wanted to make a replica of the old Moog or ARP type of synthesizer, I would do it digitally with a microcontroller.

Bob

I was thinking of doing it with a microcontroller but this is more fun, in my opinion.

I believe I've come up with a final design (attached)

I changed the wiring of the 555s, and made the frequency be controlled by a piano-like input, instead of a potentiometer, and I've added an audio amplifier IC to power the speaker!

 

[BobK]

I think you better start simulating or breadboarding before you proceed.

Bob

 

[j4cobgarby]

I think you better start simulating or breadboarding before you proceed.

Bob

You're right.

Any tips on how to simulate something like this?

 

[Audioguru]

I do not know what you are doing because you do not know what you are doing.

 

[davenn]

I was thinking of doing it with a microcontroller but this is more fun, in my opinion.

it is, IF you know what you are doing

I believe I've come up with a final design (attached)

you cannot put a bunch of diodes in series like that and expect it to work

Any tips on how to simulate something like this?

LTSpice is a free simulation program and is readily available on the net

 

[Audioguru]

Here are the important coupling capacitors and negative feedback for the mixer.

 

[j4cobgarby]

I do not know what you are doing because you do not know what you are doing.

I found the 555 wiring from this website: https://www.electronics-tutorials.ws/waveforms/555_oscillator.html
The diodes are there so that nothing weird happens because I want all of the 555s to share the same resistor

 

[Audioguru]

I found the 555 wiring from this website: https://www.electronics-tutorials.ws/waveforms/555_oscillator.html
The diodes are there so that nothing weird happens because I want all of the 555s to share the same resistor

That website (in India?) does not work right now.
If the diodes allow the 555 ICs to work then they will almost play the same frequency. Why use three oscillators when one will work much better?

Use an audio mixer circuit to mix three oscillators together. Since they play almost the same frequency then the mixed sound will flutter as the signals add in phase then cancel when out of phase. If you want to play a musical chord with three frequencies then use a musical IC designed to do it.