Previous in Forum: Installing an Operating System without a CD   Next in Forum: Free - secure- remote access that works great.
Close
Close
Close
8 comments
Associate

Join Date: Apr 2009
Posts: 28

Universal File Format

02/12/2010 10:56 AM

Hello,

I couldn't decide where to post this, so it goes into general discussion...

I have recently received the results of a vibration test conducted in a university laboratory; and the results data (autospectrum and frequency response data) are given in Universal File Format. I haven't worked with the Universal File Format before, so I'm a little confused about how to handle the data. Here are the inputs of my question:

* The files are in "Universal Dataset Number 58". I have recieved the corresponding key file and understood most of it. Ordinate is real and single precision; the abscisse is evenly spaced with increments of 4.0Hz. As far as I understood, the data in the columns are ordered like this:

Y1 Y2 Y3 Y4 Y5 Y6

Y7 Y8 Y9 Y10 Y11 Y12

Y13 ... ...

* The test was a random vibration test ranging from 5 to 2000 Hz for a total of 19,18 Grms.

The problem, or what I fail to understand is that, the file has 6401 values as data (total of 1067 lines with 6 columns), and the file says it starts from 0Hz with increments of 4.0Hz at every point. So this takes the end frequency to 25604Hz! I'm considering if the beginning and end points of the data are 0 and 2000 Hz as in the test, but there are more data points due to sampling (or resolution). Could this be possible? Seems unlikely because this is not time domain, but then again, I'm not very educated on the subject.

I'd also love any help that could make me plot these files on Excel. I can also use Matlab but the company has limited number of licenses and I generally can't get to use Matlab.

I hope I've stated my question clear enough. Thanks in advance for all the help!

Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Anonymous Poster
#1

Re: Help regarding universal file format

02/12/2010 12:16 PM

I always got plots and have never gotten data in universal file format.

Here are a few ideas. None of them stand out as being good ideas. They are just the best ideas I have at the moment.

The math that analyzes the data probably works on powers of two. Is the spacing really a "nice" 4.000000000 Hz? This could affect the number of samples.

Most items are shaken for a set amount of time on the X axis, then redone on the Y axis, and finally done on the Z axis. Could you have 3 sets of data in the file?

I always want data for the table and the item. Tables sometimes shake side to side and this can be very nasty if something is binding. Thus, if I don't see good table plots then I don't even look at the item under test. Could there be data for more than one accelerometer?

Were the accelerometers single axis or three axis?

Did you ask the people at the university how to use the data? I have had many hours of shake time at several different military labs. I always got paper plots. I never had anyone offer Universal File Format. Wikipedia doesn't seem to know about Univeral File Format. Don't be afraid to ask the university. This is not a "dumb question" from someone new to shaking. I am interested in the answer and I have broken many thousands of dollars of products and fixtures on shaker tables.

Please let us know what you find out.

Register to Reply
Guru
Hobbies - Musician - Engineering Fields - Chemical Engineering - New Member Engineering Fields - Control Engineering - New Member Engineering Fields - Instrumentation Engineering - New Member

Join Date: Jan 2007
Location: Moses Lake, WA, USA, Thulcandra - The Silent Planet (C.S. Lewis)
Posts: 4216
Good Answers: 194
#2

Re: Universal File Format

02/12/2010 10:55 PM

Hi archmagus,

I'm sending you a private mail with my normal email address. Send me the file and I will help you sort it out.

Mike

__________________
"Reason is not automatic. Those who deny it cannot be conquered by it. Do not count on them. Leave them alone." - Ayn Rand
Register to Reply
Associate

Join Date: Jun 2006
Posts: 38
Good Answers: 1
#3

Re: Universal File Format

02/12/2010 11:15 PM

Hello,

Something that might be useful to know, for some of us anyway, is what the file extension your data is stored in. That may provide a clue as to what application would best work for your conversion attempts. For example, is it ".doc", ".xls" ".txt", ".odt", ".rtf" or something else?

__________________
rega451
Register to Reply
Guru
Hobbies - HAM Radio - New Member United Kingdom - Big Ben - New Member Fans of Old Computers - Altair 8800 - New Member Canada - Member - New Member

Join Date: Jun 2006
Location: Toronto
Posts: 3968
Good Answers: 119
#4

Re: Universal File Format

02/13/2010 4:16 AM
__________________
Per Ardua Ad Astra
Register to Reply
Associate

Join Date: Apr 2009
Posts: 28
#5
In reply to #1

Re: Help regarding universal file format

02/16/2010 2:27 AM

Thanks for the reply. Here are the situation from the perspective of testing in more detail:

The tests was done on all three axes seperately, and a total of 6 accelerometers (some single axis, some three axis) were used. Despite that, the raw data I have recieved is in lots of small files; and each axis has its own data files. Moreover, each axis of each accelerometer has its own data file for each test. To sum up; I'm pretty sure that a single file that I have at hand represents:

A single axis of a single accelerometer for a single test run at a single axis.

And I have the control accelerometer data too, fixed on the shaker plate. So I can confirm if the data is healthy, just if I can plot it (grrrr :) )

As for the problem with the university; it's a relatively new lab and they don't know how to process the raw data either, they just get the plot from the software bundled with the data acquisition system. So even if I ask them, it won't be of much help.

It's confusing though, you name something "universal file format" and nobody knows about it. They should have given it a more specific name :)

Thanks for he help! I'll post here when I get to solve the darn problem...

Register to Reply
Associate

Join Date: Apr 2009
Posts: 28
#6
In reply to #3

Re: Universal File Format

02/16/2010 2:30 AM

The extension of the file is "*.uff" which is the abbretiation of "universal file format" :) I've already found the guide files that says which line provides which information, so I basically have the map. What I don't get it is that; the number of data points involved is too much when considered the step size specified in the files...

And when I plot what I have as I understand (all the data points plotted one after another) what I get is far more different than what the university gave me as a sample plot of one my my files. So I'm definitely missing something somewhere, I just don't know where :)

Register to Reply
Guru

Join Date: Feb 2008
Location: East Tenneesse, USA
Posts: 685
Good Answers: 45
#7
In reply to #4

Re: Universal File Format

02/16/2010 7:08 AM

Hello aurizon, I took your links and looked around. I found a link to

Universal File Formats for Modal Analysis Testing

Down toward the bottom of the page are two links one is pdf that looks like it gives some good info on the UFF formats in testing.

Since I never have used UFF, this page and links seem to me might explain the format differences in the difference versions.

Just trying to add my 2cents. I may even not even be in the right ball park.

Charles

__________________
Metal is the material, The forge is life, The anvil and hammer bring character and soul.
Register to Reply
Associate

Join Date: Apr 2009
Posts: 28
#8

Re: Universal File Format

02/25/2010 4:11 AM

Hey everyone, I have finally found my solution. And such a simple one...

The data indeed IS spaced with 4Hz intervals as the file specified. And since it has 6401 data points, it DOES display the data for up to 25604Hz. I just have to select the frequency of interest to display. In this case, the first 500 data points that is up to 2000Hz. And it didn't make any sense when I plotted the entire spectrum, becasue I couldn't distinguish the shape of 0-2000Hz interval for after 2000 the entire spectrum was zero! Thus I saw an initial peak and a zero afterwards when I plotted the entire spectrum.

I hope my struggle helps everyone to gain some information on Universal File Formats. And thanks for all the help! :)

Register to Reply
Register to Reply 8 comments
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Copy to Clipboard

Users who posted comments:

Anonymous Poster (1); archmagus (3); aurizon (1); metalSmiths (1); Mikerho (1); Rega451 (1)

Previous in Forum: Installing an Operating System without a CD   Next in Forum: Free - secure- remote access that works great.

Advertisement