The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Confusing results from dft
https://designers-guide.org/forum/YaBB.pl?num=1422053425

Message started by sos on Jan 23^rd, 2015, 2:50pm

Title: Confusing results from dft
Post by sos on Jan 23^rd, 2015, 2:50pm

I started being confused by the results simulating my circuit, so I simulated something very simple. Sin wave source driving two resistors in series, each of the resistors being 2k. The output is taken at the midpoint of the two R's (i.e. source amplitude / 2).

I am simulating transient noise.
Stop time 2ms
Noise fmax 1M
Noise fmin 1k
Noise seed 1
Noise scale 1
Noise update step

I am trying to attach an image, but it shows no sign of actually attaching. If it comes through, use that rather than the text below.
Both waveforms are plotting dft(Vt("/out") 10u 2m 4096 "Rectangular" 1 1). The top waveform has the source set to 5uV peak, the bottom one has the source at 1uV peak. The top result shows the source at 5uV; since this is attenuated by 2 at the measurement point, the measurement must be in peak to peak (the measurement shows 5uV). Same sort of thing on the lower plot when the source is 1uV peak, the plot shows 1uV peak to peak.

My big concern is the lower limit for the signals on the plots. In the upper plot, all the bin voltages start at -0.5uV and run up to about 0.15-0.2uV for the noise floor and the signal is as described above. In the lower plot, the bin voltages start at -0.1uV and run up to about the same level, 0.15 to 0.2uV.

Why in the world don't the noise bin voltages start at zero?

If you measure the noise, on the upper plot it starts at -0.5 and goes up to about 0.2, implying 0.7uVpp. The lower plot ranges from -0.1 to 0.2, only 0.3uVpp, implying different noise levels. When I apply dB20 to the dft, the two plots are identical for noise, there is no noise difference.

Thank you,
Steve

Title: Re: Confusing results from dft
Post by sos on Jan 23^rd, 2015, 3:34pm

Correction: the results are shown in peak. In the top plot, the source is set to 5uV peak, the signal plot shows 2.5uV at the divide by 2 point, so it is also in peak V, not peak-peak.

Title: Re: Confusing results from dft
Post by weber8722 on Feb 27^th, 2015, 10:24am

Hi,

I think it is a VIVA plotting problem. The bins should start at 0V and rise up, but they start at y-axis minimum. Best export the data or look to them as table.
Is this the problem? I think this little bug will be fixed soon.

Bye Stephan

Title: Re: Confusing results from dft
Post by RobG on Mar 28^th, 2015, 7:38am

Isn't the dft result a complex number? Perhaps it is just plotting the real part of the dft. Look for warnings.

I always take the abs() of a dft by habit.