I am adapting a jfet buffer to suit my configuration. Please review the attached ltspice artifacts.
I am seeing some oscillations(?) after 100K Hz in the FFT output. What is it indicative of? The wiggles almost disappear if I connect the RC filter at the output (R11, C4), but I'm not sure if its required.
I am seeing some oscillations(?) after 100K Hz in the FFT output. What is it indicative of? The wiggles almost disappear if I connect the RC filter at the output (R11, C4), but I'm not sure if its required.
Attachments
Some of it may be real but you need to set the transient command max step size to {1e-4/f} or less (<=100n), ".tran 0 {10/f} 0 {1e-4/f}"; ".par f=1k"
You have the plotwinsize=0 directive which is the other thing required. I do an FFT of the source which should be about zero THD. If not, then something is wrong with the parameters.
You also have noise in the supply, which will show up in the FFT.
You have the plotwinsize=0 directive which is the other thing required. I do an FFT of the source which should be about zero THD. If not, then something is wrong with the parameters.
You also have noise in the supply, which will show up in the FFT.
Last edited:
Thanks steveu, your suggested changes did the trick. No wiggles now. Is that a universal rule of thumb for max step size?
Whatever works. If you get too crazy, the simulation runs slow. I use at least {1/f/4096}, usually {1e-4/f}. If the FFT of the sine source is good (0%), then it's good enough.
It's aliasing, or imaging actually. 1/step_size becomes the sample rate.
Last edited:
