Lowest part count 60Hz sine wave oscillator

Okay guys and gals,

Lowest part count 60Hz sine wave oscillator. Please show me your elegant solutions;

I have a +/-5V supply on the board. 60Hz MAINS is not on the board.

Lowest part count - a µP with built-in DAC.

You should be probably a bit more specific, e.g. about frequency and voltage stability and permitted distortion. The classicall analog solutiion, an OP RC oscillator with feedback stabilized amplitude. Many examples have been published at Edaboard.

FvM said:

Lowest part count - a µP with built-in DAC.

You should be probably a bit more specific, e.g. about frequency and voltage stability and permitted distortion. The classicall analog solutiion, an OP RC oscillator with feedback stabilized amplitude. Many examples have been published at Edaboard.

Click to expand...

60Hz +/- 0.05Hz
Stability +/-0.1%

I might power the board from a $10 ac wall adapter, seems like the simplest solution

How accurate do you want it to be? Frequency tolerance and distortion?

Maybe a Wien bridge oscillator properly calculated will generate your sine wave.

Other way is a dual or quad opamp IC (depending of the output impedance you wanted) and a schmitt trigger with integrator triangle wave generator at an amplitude of 0.6V. Then a resistor (aprox 4.7K, you can play with that value) connected to the buffered triangle wave and two diodes pointing in opposite directions between the resistor and GND. If you correctly generate the triangle wave you will see a very nice sine wave at 0.6V peak amplitude. Then you can buffer and amplfy it to your application.

Low part count, no MCU, analog accuracy, and the space depends of which package you decide to implement.

Hope you understood, just ask me if you have any question.

You cannot request lowest part count without specifying; tolerance, supply sensitivity, output impedance , signal level, % distortion, supply current, etc..

The only important criteria are; quality, availability and cost.

I might suggest a 4060 counter and low freq., low cost crystal and use either LPF on square wave for 50ppm accuracy for $0.25 in volume or use R ladder on counter outputs for a pseudo sinewave.

But maybe you saw this on another thread.


But you will go far to understand the best design is a good spec.
Anything less is short-sighted.

- One variac.

Externet said:

- One variac.

Click to expand...

Was that an SMT or THT variac ?

I'd vote for the PIC.

How about a finger injecting hum into a self biased CMOS inverter?

Tony :laugh: