ICs to generate a sinusoidal waveform with frequencies 1 Hz - 10 kHz

Hello Ray,

Building a fixed frequency sine is fairly simple. However that uses op amps opposed to a "sine generator" chip. You asked for a 10,000:1 range which is the opposite of fixed. The same 10,000:1 ratio makes freq to voltage difficult. 

I would suggest a sine data table, but I am not aware of a hardware way to clock in to analog out. Our MSP processors can do table to ADC in software. 

Hi Ray,

I think you mean a switched capacitor filter like the MF10-N? Something like this?

https://ee-paper.com/three-kinds-of-sine-wave-generator-circuit-diagram-sharing/

https://www.ti.com/product/MF10-N?keyMatch=MF10-N&tisearch=search-everything&usecase=GPN

Kai

Hi Ron,

Good Day. Please see the response of our customer to your reply. Thank you very much.

Hi, Thanks for the quick and useful reply. I see it is hard to generate a variable frequency with a ratio of 10000:1 all in analog circuits . Should be easier if the frequencies to be generated are something like 950 Hz to 1050 Hz? Anyway can you recommend me a TI IC to produce a fixed frequency sine?
And lastly...i did not have in my mind the switched capacitor filter...Do you mean for this " an IC that converts a period of a square wave into a proprotional voltage"? I am sorry but i can't figure out how to use it

Best Regards,

Ray Vincent

Hi Ray,

the switched capacitor filter is a filter which has the corner frequency controlled by the frequency of a square wave signal. By this you filter out the harmonics of another square wave so that a pure sine wave results.

Kai

Ray,

This app note covers various fixed frequency analog sine wave generators. https://www.ti.com/lit/an/sloa060/sloa060.pdf

Is the goal simply to turn the input square wave into a sine wave? I suspect not.

Hi Ray,

Ray Vincent Aranzaso said:

Should be easier if the frequencies to be generated are something like 950 Hz to 1050 Hz? Anyway can you recommend me a TI IC to produce a fixed frequency sine?

so, you no longer need to generate a sine in the range between 1Hz and 10kHz?

If you must generate a fixed frequency sine, what characteristics must it have? Very constant frequency? Or low distortion? Or...? I ask, because there are zillions of ways to generate a sine.

The more details you provide, the less we need to guess

Kai

Hi All,

Good day. Please see below the response of our customer to your reply. Thank you very much.

--> "the switched capacitor filter is a filter which has the corner frequency controlled by the frequency of a square wave signal. By this you filter out the harmonics of another square wave so that a pure sine wave results."
--> I think this is not what I am looking for: In this case the sinusoidal output would come from a filtered (notch filter with MF10) square wave and I do not want a bad THD in the sinusoidal output.

-->"Is the goal simply to turn the input square wave into a sine wave? I suspect not"
--> the goal is to output a sinusoidal waveform whose frequency is the same of an input square wave. The square wave has a variable frequency therefore the sinusoidal generator IC must have a variable frequency (around 1 kHz...e.g.:sometimes 900Hz, sometimes 1.1 kHz and so on). The most important thing is the synchronization of the two waveforms. I was also looking to PLL circuit since they are designed for similar purpose...

Best Regards,

Ray Vincent

Ray,

If the PLL makes a clean enough sinewave then PLL will work provided that the input frequency doesn't change quickly. The PLL loop may lag or overshoot (resulting in phase error). You could also Low Pass Filter the PLL sine to clean that signal. 

For best response time, just use the incoming square wave (if 50% DC) and add many stages of low pass filtering with a cut off frequency between the highest input frequency and the lowest input frequency 3rd harmonic.  You will need a lot LPF stages set close to 3rd harmonic to remove the odd harmonics and keep phase shift of the fundamental sine under control. If 'sync' doesn't have to include phase alignment then the filtering job will be much easier (fewer stages set to lower frequency cutoff).

Hi Ray,

a PLL shows jitter, the higher the faster the settling time is programmend.

I would also go for a steep low pass filter removing the harmonics.

Kai

Like this, for instance:

ray_sine.TSC

Kai