![](/images/new-backgrounds/1097476/9747619x1.webp)
Phase Accumulator Spectrum
An alternative way to observe the quality of the waveforms synthesised by the phase accumulator is to look at the frequency spectrum. I was lucky enough to have a
3rd
12.5MHz Harmonic Fundamental
12.4125MHz Fundamental
As we know the 12.5MHz signal is |
|
| ||||
nominally clean in terms of the synthesis |
|
| ||||
process and this is reflected by a distinct | 12.5MHz |
| ||||
12.5MHz component which is >45dB above | Fundamental |
| ||||
the noise floor. The zoomed plot shows how |
|
| ||||
the fundamental covers only a narrow |
|
| ||||
bandwidth keeping in mind that the |
|
| ||||
resolution of the FFT is only ~50KHz. |
|
| ||||
Note the third harmonic of a square wave is |
|
| ||||
also very distinct at 37.5MHz and is even |
|
| ||||
close to being the theoretical 9.54dB down |
|
| ||||
from the fundamental (1/3 amplitude). |
|
| ||||
|
|
|
|
|
|
|
| 5MHz/division |
| 1MHz/division |
|
|
|
| 10dB/division |
| 10dB/division |
|
|
|
|
|
|
|
| 12.4125MHz |
|
When we move to the less than perfect |
| |||||
Fundamental | 12.4125±0.74MHz | |||||
12.4125MHz waveform the effects of that |
|
| ||||
5ns of cycle jitter and jumping between |
|
|
12.5MHz and 11.76MHz waveforms is obviously having an impact on the spectrum. Although the fundamental frequency component is good, if a little wider in bandwidth, there are significant modulation effects leading to a family of spectral components and raised noise floor. Hint –
Frequency Generator for the