Low distortion sinewave oscillator without big capacitor.

That (ish) was a bit big. H2,3 are circa -180 and -190 dBc. MAGA

Make America Go Away

You do seem to want to sound like John Larkin.

And if you don't notice that it's there you won't realise that you have skipped it.

Sample and holds necessarily put switching spikes into the power supply rails. Decoupling the relevant bits from the rails can minimise the consequences, but if you are aiming for harmonics 120dB below the fundamental, you have to decouple a lot of routes rather carefully. Somebody like you who doesn't like ferrite beads and ferrite chips wouldn't enjoy that.

Telling myself that I should have got to bed hours ago. Now that I'm retired I can afford to be occasionally silly (if not all that often).

The phrase "insult the the intelligence" doesn't imply any actual insult

- you just run the risk of boring them. You do need to read more widely.

I wasn't able to reply directly to Bill Soman's post 10th April 6:33 AM due to error 411 header too long.

However I just wanted to address the following:

"Sample and holds necessarily put switching spikes into the power supply rails. Decoupling the relevant bits from the rails can minimise the consequences, but if you are aiming for harmonics 120dB below the fundamental, you have to decouple a lot of routes rather carefully."

"Somebody like you who doesn't like ferrite beads and ferrite chips wouldn't enjoy that."

You really are a complete idiot Bill.

Ed, DO NOT FEED THE TROLL

Joe

Yeah you're right. I am done with Bill.

Not a well justified assertion.

I don't give him the flattery he feels he deserves.

How about using a DAC driven from a uC, or even some clever bit of hardware?

A uP with an ADC could measure the sine amplitude and an MDAC could tweak loop gain. But you may as well just synthesize the sine wave with a 20-bit DAC.

If you could somehow measure sub-PPM distortion, tweak that down in the DAC data.