home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.engr:2122 sci.engr.mech:411 sci.research:1153 sci.math.num-analysis:2976
- Newsgroups: sci.engr,sci.engr.mech,sci.research,sci.math.num-analysis
- Path: sparky!uunet!utcsri!torn!newshost.uwo.ca!gfx.engga.uwo.ca!perwaiz
- From: perwaiz@gfx.engga.uwo.ca (Jim Perwaiz)
- Subject: VORTEX SHEDDING GFREQ. _ FFT (SUMMARY)
- Organization: University of Western Ontario, London
- Date: Fri, 9 Oct 1992 18:32:20 GMT
- Message-ID: <1992Oct9.183220.21929@julian.uwo.ca>
- Sender: news@julian.uwo.ca (USENET News System)
- Nntp-Posting-Host: cad7.gfx.engga.uwo.ca
- Lines: 31
-
- Here is the summary of the responses on how to get a
- better FFT plot.
-
- 1. Wind tunnel should be stable.
- 2. Use windowing.
- 3. Put the hot-wire out of the wake of the cylinder.
- 4. Use and improve analog anti-aliasing filters.
- 5. Sample at at least 10 times the expected frequency.
- 6. A good reference on windowing: " On the use of windows for
- harmonic analysis with the discrete fourier tranform",
- Proc. of the IEEE, vol 66, no 1, jan 1978.
-
- 7. Instead of the FFT try the "max. entropy spectra"
- method - Numerical Recipes has a write-up.
-
- 8. My own: read the ILS software manual again !!!
-
-
- I hope this meets the requests of some people.
-
-
- Once again, my THANKS to everyone for their great help
- and interest. My apologies for posting this summary so late.
-
- Good bye
-
- --
- -----------------------------------------------------------------
- Jamshed Perwaiz |perwaiz@engrg.uwo.ca
- Department of Mechanical Engineering |perwaiz@gfx.engga.uwo.ca
- The University of Western Ontario |
-