Hi, Folks:

I'm reading some papers on DCO.

One question I got is on white dithering of DCO. Basically due to limited frequency resolution, there is high quantization noise without sigma-delta modulation. White dithering is often used in this case.

1. I'm not sure that white dithering is using finer resolution, i.e. smaller capacitor, for random dithering or not. For example, if dithering word length is 4-bit, can I say that dithering LSB resolution is 1/16 of normal control bits?

2. If it is using smaller capacitor, why not directly use that in normal control bits, to lower the quantization noise?

Thanks,
Neoflash

The idea is to have an infinitely small capacitance. The trick is to do that through time averaging of a larger cap. Of course this causes quantization noise and the smaller the dithered cap the lower the quantization noise.

Emad