Docushare

To get to Moog's Docushare system:

http://docushare.moog.com

username: jfox2
password: regular password

Q factor

The q factor determines how underdamped a system is.

Q = 1/(2*zeta)

More can be found in:

http://en.wikipedia.org/wiki/Q_factor

EZ Analyst - On saving/exporting data

To do averaging in EZ Analyst,

1) Click on "continuous measurement" button.

2) Take the data. (settings are under configuration menu)
NOTE: if taking time histories, the #avgs specified is irrelavent. It will take data for as long as you tell it to.

3) Save data. Go to "save multiple data set".

4)Exporting data...this is done in two steps. ONE, you have to go to the export menu to enter in all your settings (what you want file to be named, what you want to export, etc.). TWO, you have to actually export the data. This is done under the "control" menu.

WHAT IT EXPORTS: For awhile I was unsure if EZ Analyst exported the AVERAGED FRF or just a single FRF from the last time window (because when exporting multiple data set data, EZ Analyst only saves time history data from last time window if you are doing a continuous measurement instead of recording time history). Despite this, EZ Analyst DOES export the AVERAGED FRF. I did a self check and found this to be the case!!

Checking out equipment in Atrax

Log in using regular user name and password.

Go to Check out/ return equipement.

Double click the CSA logo to allow editing of check out log.

If checking out, scan each item.

If returning, just enter the return date in the proper field.

I-deas not working

if I-deas loads up alright, but does not take data, it could be because the address scheme for the input modules is flipped (aka reading the last module first and first last). To change this, go to:

UGS\NXI5\tda\hpvxi-1432

This is the setup file for ideas. Scroll down to the "flip 1" option and change it.

Ideas
flip
not taking data

Good Modal Extraction

To get good modal extraction from tap tests and the like, You should have at least 5 (better to have 10) spectral lines within the modal bandwidth.

Modal Bandwidth: the area between the half power points of the mode

Half Power Points: take the peak of the mode, normalize it to 1, come down to .707 on either side. Whatever width that is is the modal bandwidth. (.707 because you are looking at an FRF, a linear scale. If you were looking at a PSD it would be 0.5)

Monday, June 29, 2009

TAP TESTS

To get good coherence/results:
1) Use Delrin hammer tip. This gets more energy in lower frequencies. Brass does not.
2) If you move the hammer around, this generates a signal from the load cell!! SO...KEEP THE LOAD CELL AS STILL AS POSSIBLE!
3) Tape the accel cable to the string holding the beam.... I'm not actually sure this makes much of a difference but it is cleaner that way (the cable isn't wobbling around after each hit).

Thursday, May 14, 2009

FRF matrix - a matrix containing data points from all the FRF measurements you've made on a structure. It would be a 3D matrix where each "slice" of the matrix is a different frequency bin (f1, f2, f3...). Each entry of that "slice" is the value of each FRF at each measurement location. Ideally, this should be a symmetric matrix since m(1,2) is ALWAYS equal to m(2,1), where m(1,2) is the response at location 2 due to excitation at location 1, and m(2,1) is the response at location 1 due to excitation at location 2.

Well conditioned matrix - a matrix whose inverse is insensitive to minor changes in values. Sometimes, when you have FRFs where some of the values are very large (100k G's e.g.) and others are very small (.1G, e.g.), the inverse is very sensitive to changes in those small numbers. Since in testing, those small measurements could be near the noise floor, their values could change arbitrarily, causing problems in the inverse calculation.

Deconvolution - is an algorithm-based process used to reverse the effects of convolution on recorded data. convolution is simply when you have an input that is multiplied by a transfer function to result in a different output. Deconvolution is simply when you have the output and transfer function and work backwards to get the input.

Thursday, March 19, 2009

When tapping a hole, you will get a lip of material around the hole. To stop this from happening, use a single flute countersink and bore out (you can do this by hand with aluminum) a small bit around your hole. You can do this post tapping if you want, but then you have to run the tap through again to make sure you did not screw up the thread.

The Single Flute Countersink set is next door above all the taps. 82deg for english, 90deg for metric.

Also, MANIFOLD...I never really knew what the definition was exactly. Maybe because it has multiple ones:

For basics, a manifold is "something having many different parts or features".

In Machinery, manifold means "a chamber having several outlets through which a liquid or gas is distributed or gathered".

Tuesday, March 17, 2009

What Brad is looking for when he asks for mass properties of a system so that he can design isolation system:

If isolators are already in place, give locations of isolators relative to origin.

Clearly specify the coordinate system.

Give him the inertial properties as aligned with the output coordinate system, along with mass, distance from origin to COG, etc.

This should all be included in one easy to reference document!

Friday, March 13, 2009

Parasolid and Step Files:

These are kernels that communicate and migrate 3D surface or solid data - NONPROPRIETARY information . They do not communicate things like features or extrusions, or material content!

If you want to see for yourself, just open them up in text editor! They are saved in an ASCII format. All the numbers in this file only communicate 3D locations.

Wednesday, March 4, 2009

OK...I need a way of remembering which is single-ended and which is differential.

Single-ended is when you have one wire and compare it to ground
Differential is when you have two wires and difference the two.

...now to remember this, only one of these has a "g" in it... sinGle-ended... and that's the one that references Ground.

ECTRON AMPLIFIERS (563H)
yes they are a pain to deal with.
To zero the bridge:
1. Short the input connection(pins B and C...pin layout can be found in schematic in back of manual).
2. Put gain to 1. Adjust RTO (referred-to-output) until output is zero.
3. Put gain at 1K. Adjust RTI (referred-to-input) until output is zero.
4. Put gain back to 1... if output is not zero, do steps 2 and 3 until it is.

Wednesday, February 25, 2009

I am learning about serial communication. Baud rate is the rate of transfer (bits per second). Data is transferred one bit at a time in "packets". A packet consists of a start bit, a stop bit, the data bits, and sometimes a parity bit. The receiving device and transmitting device must have the same settings (baud rate, # data bits, etc.) to successfully send data back and forth.

Friday, February 6, 2009

Learned all about transistors and "NPN open-collector ouputs"!

Transistors consist of a base, emitter, and collector. When no current is applied through the base to the emitter, no current flows from the collector to the emitter. However, when current does flow through the base to the emitter, a proportional amount of current flows from the collector to the emitter. The relationship is: Ib = B*Ic.
So... a little base current allows a much larger collector current to flow. (B approx = 100). Oftentimes, Ic can only go so high (depending on its source power). Thus, even if Ib is very high and via the relationship Ic should be huge, Ic will only go as high as it can based on the circuit its hooked up to.

NPN open-collector output:
Now, take the idea of a transistor and put it at the output of some circuit. More specifically, connect the circuit output up the the transistor "base", connect ground to the "emitter", and leave the collector open (hence open-collector output). Now, you have a connection (the collector) that is either a short circuit or an open circuit, depending on whether the base is receiving current from the circuit.
Use this feature to hook up any other circuit you want to the collector, without worrying about affecting the internal circuit!

NPN vs PNP: these are two types of transistors. They differ in that a NPN is typically closed unless a current is applied to "base", where as PNP is open unless a current is applied to "base"

Tuesdsay, February 3, 2009

Learned a lot about dB and octaves. First, for Power ratios, the definition for dB is 10log(P1/P0)...NOT 20!! That is only for amplitudes!! Also, when finding points for a PSD based off of a spec which specifies slopes such as dB/oct, make sure that you are using the right reference number for your dB calculations!!!! The forward point is always the reference point... that is your reference amplitude and your reference frequency. If you have a spec which says 6 dB/oct, then that means that it is 6 dB per octave slope FROM THAT POINT!....that point is P0... not P1.

Friday, January 16, 2009

I learned a lot today about data reduction. First off.... if there is drift in the accel data, it is most likely due to temperature effects. This I knew. What I didn't do though and needed to was HIGH PASS FILTER IT!!!! This gets rid of the drift and makes the data MEANINGFUL!!!

To make a highpass or lowpass filter, type in FDATool in matlab.. then choose what you want to create. Once you build your filter, go to Edit-->Convert to Singleton!! Then File-->Generate m file to save it. To use it, follow the code below:

filterlow = lowpassfilter_1000Hz; %load low pass filter
Blow = filterlow.Numerator;
Alow = filterlow.Denominator;
data_filtered = filtfilt(Blow,Alow,data); % apply the lowpass filter.

Also, when writing reports its important that I specify which DATA TYPE I am going to use... periodic? random? transient? This matters because it will determine whether I use regular auto spectrum, PSDs, or ESDs.