Let’s try the JMP SE community (cc’d) to see what help we can find there.
Jeff Perkinson
JMP Customer Care Manager
SAS Institute Inc.
SAS Campus Drive • Cary, North Carolina 27513 • USA
Ph: 919-531-6521
E-mail: Jeff.Perkinson@jmp.com<>Jeff.Perkinson@jmp.com>
Share. Ask. Learn. Connect. - community.jmp.com<https://community.jmp.com>
Connect with me in the JMP User Community<https://community.jmp.com/people/jeff.perkinson>.
On Oct 8, 2015, at 10:10 PM, Bill Worley <Bill.Worley@jmp.com<>Bill.Worley@jmp.com>> wrote:
Hey Guys,
Please take a look at the request below from one our customers. I no experience with FFT and JMP.
Thanks,
Bill
From: Putnam, Thomas R. (Tom)
Sent: Thursday, October 08, 2015 5:02 PM
To: Bill Worley <Bill.Worley@jmp.com<>Bill.Worley@jmp.com>>
Subject: FFT function
Bill,
I hope you can help me with attempting to analyze the attached data using FFT.
If you run the scripts, you can perhaps see there are two modes.
This is a round part that has some ovality. I was hoping to use FFT to quantify this in some parametric manner so that I could apply to other parts for comparison.
I see that FFT requires a matrix, but I’m unclear about the output and interpretation.
Help??!!
Tom Putnam
tom.putnam@timkensteel.com<>tom.putnam@timkensteel.com>
Business Process Improvement Technologist
Desk: 330-471-6626 Cell: 330-323-8345
________________________________
This message and any attachments are intended for the individual or entity named above. If you are not the intended recipient, please do not forward, copy, print or disclose this communication to others; also please notify the sender by replying to this message, and then delete it from your system. TimkenSteel Corporation
Message was edited by: Jeff Perkinson
Attachment removed. Bill says the data are proprietary.
Nate --
You have to zoom into the left graph to find out how many periods are at the frequency of the peak.
The data set is sampled at 2000 Hz.
t = 1/f.
t = 12.585 periods * .0005s (1 sample) = .0062925.
.0062925 = 1/f. f = 158.9 Hz.
Resurrecting this thread a bit.
Just had a similar question come from a customer. Anybody have a good explination as to why Matlab's FFT and JMP's spectral density would come up with different values?
It might be easy to interpret results by cos and sin functions, not by Euler's expression or complex numbers.
The linked PDF file explains how to transform FFT's results to coefficients for sin and cos functions.
The linked script shows the interpretation in terms of sin and cos functions.
https://community.jmp.com/docs/DOC-9294
https://community.jmp.com/docs/DOC-9295
Thanks Dan. I will the add-in with our customer.
billw@jmp -- this Add-In from Bill Fish may be something you'd be interested in forwarding to your customer: JMP Addins for Frequency Analysis and Dual Seasonality Time Series Analysis https://community.jmp.com/docs/DOC-6994
Byron,
Thanks for your reply. Maybe a little naive here, but is this okay to share with a customer?
Best,
Bill
Hi Bill,
besides using the Scripting functionality to run an fft with MATLAB I believe Dan V. has some example on how to do this, but I do not remember what exactly The first I guess was this blog post: How to use JMP and MATLAB together
Regards,
Martin
