In Reply to: Re: test for transients posted by john curl on July 18, 2003 at 10:35:22:
Hi,Couple of problems to overcome have been holding this up...need
a little help.First is the input waveform. There are 7 voltage sources for
the 1,3,5,7,9,11,13th harmonic with amptitudes set to be equal
to the harmonics of a square wave. An additional voltage source
for the small sine wave. This works pretty good. The problem comes
when changing over to a triangle. To make a triangle (I'm at work
so I don't have any of this in front of me), you simply reverse the
3,7,11. When I do this, I get a very hi crest factor for the
triangle waveform. So hi, I had to shift the operating point of
the stage to accomodate it.What should the triangle peak be? Should it be the same peak level
as the square? If this is true, then I just need to adjust the
amptitudes.Second problem is leakage thru the FFT. This shows itself as
"skirts" around each harmonic or IM component at each frequency.
It's fairly easy to get a FFT to produce a hi-res (> 150 dB) range
with single tones, but this combo of frequencies in these waveforms
is not as FFT friendly. This causes the "skirts" of the "square wave"
harmonics to rise up and "bury" the interesting harmonics of the sine
wave. I'm still in the process of getting this setup.The frequencies I'm using are 2500Hz for the sine, and 525 Hz for
the square wave. Are these hi enough frequencies? As you go higher
in frequency, the need for smaller step sizes becomes greater. This
greatly increases the sim time.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: test for transients - mfc 07/18/0311:21:43 07/18/03 (13)
- Re: test for transients - john curl 15:42:14 07/18/03 (12)
- Re: test for transients - mfc 20:10:44 07/18/03 (11)
- Re: test for transients - john curl 22:34:11 07/18/03 (2)
- Re: test for transients - john curl 14:35:37 07/19/03 (0)
- Re: test for transients - mfc 22:53:22 07/18/03 (0)
- Re: test for transients - mfc 21:14:44 07/18/03 (7)
- Re: test for transients - Scott Frankland 21:56:29 07/18/03 (1)
- Re: test for transients - mfc 22:17:05 07/18/03 (0)
- Re: test for transients - 13th Duke of Wymbourne 21:29:55 07/18/03 (4)
- Re: test for transients - mfc 22:14:39 07/18/03 (3)
- Re: test for transients - 13th Duke of Wymbourne 21:05:22 07/19/03 (2)
- Re: test for transients - mfc 17:01:29 07/22/03 (0)
- Re: test for transients - john curl 15:09:36 07/21/03 (0)