NI-DAQmx: Impact Test real-imagine plot

[Alfeos]

Hello.

I am a student of mechanical engineering and I am not familiar with signal proccessing. I have to use LABView for a measurement experiment. In that project I have to use an impact hammer and an accelerometer to test surfaces of some specific meterials. My teacher tell me that I have to take the real-imagine plot of undamaged pieces as a model so I can use them at future for model. My problem is that the code I had write or the one that I use form NI example (NI-DAQmx: Impact Test) returns me a messy plot no matter how many hits I made. My teacher said me that as many hits I do the plot will be more clear but he cannot figure out what goes wrong either cause he is not familiar with labview neiher.

I wrote two different codes one using ready VI's and one that made the mathematic as my theory book say.

Please help me cause I am in a dead-end and can't move on with that.

I can post images from my codes or how I use the NI-DAQmx: Impact Test.

Thank you!

[preston_johnson]

Thank you for your post. 

1st - are you getting good time series data from the hammer and from the response accelerometer?  In the time series plot you should see the impact and the decay.  Are you using a data acquisition board which provides the ICP power to the hammer and the accelerometer?

[Alfeos]

I got the following measurement setup. Accelerometer with it's own amplifier and power (comes as set) , I used 2 kind of hammers 1 made by an other student with its own amplifier, and one that we bought and I connect it on an amplifier also. Then I connect them (hammer accelerometer) on BNC2110 board and then on the data acquitition card.

The time  signals  looks to be good. But I got the same messy plot if I use a simulated sine signal for each channel.

[TV_Enable]

Hi,

Thank you for your quick reply.  Could you please attach screenshots of your Real vs. Imaginary plot and the time waveforms acquired during the DAQmx step.  If you could attach your code, that would be great too.  Thanks!

Regards,

Todd V.

[Alfeos]

Thanks for yours intresting.

I have made two codes the first one is made with traditional and I have tried to make the mathematics by my own.

The second one is based on the example (SVXMPL Impact Test) with some differentials.

I have used mx very few times and I experienced major difficulties with the example on how I can use it. I cannot measure anything with that. I guess there is an issue about the way I choose the channels.I set ch 0-1 (there i plug hammer- accelerometer). Maybe some help on that can solve my problem faster ofcourse I will prefer mine own to work if that is possible. The graphs I need to got is force - accelerometer time and frequency, nyquist real imagine and coherence (smth that the example got). But I don't know if the 2 mathematic way's are the same (you can see it at the code2).

I will be very grateful with every kind of help

thank you

[TV_Enable]

Hi,

There are a few things that I noticed.  First, you are using Traditional DAQ.  I would strongly encourage you to use DAQmx.  It is simpler and has far fewer bugs than Traditional DAQ.  Second, I would go into Measurement & Automation Explorer and create an acceleration task.  To do this, right-click on Data Neighborhood and select Create New.  Then choose NI-DAQmx task and choose an analog input acceleration task.  See if this will allow you to get the necessary data.  If so, you can generate LabVIEW code from this task.  Finally, in your AI Config and AI Start VIs should not be inside the loop, they should be before the loop begins.  The AI Clear and merger error VIs should be outside the loop as well, except they should be after the loop is over.  Essentially, you are setting up, starting, and closing your task each time the loop iterates.  This is very processor intensive and will drastically slow down your loop.  Please try the above steps and let me know if they work for you.  Please let me know if you need any assistance doing the above tasks.  Have a great day!

Regards,

Todd V.

[Albert.Geven]

Hi Todd

Just curious, what is an accelleration task, and what does it accellerate?

Where can I find written info on this.

[mallorim]

Hi Albert,

An accelerometer is a type of sensor that is used to measure acceleration and output a corresponding voltage. Using Data Acquisition (DAQ) hardware, you can acquire the voltage from the accelerometer. However, most accelerometers will need  some form of current excitation, typically IEPE Excitation which is a widely used standard. You can view these sites for a discussion of Accelerometer Principles and Types of Accelerometers.

Some DAQ cards are custom built to not only acquire the voltage but to supply the excitation current at the same time, specifically Dynamic Signal Acquisition (DSA) cards. An acceleration task is an instance of a DAQmx input task specially created for measuring accelerometers. You can programmatically configure the task in the LabVIEW or create the task in Measurement and Automation Explorer (MAX), as seen in this Knowledge Base. You will notice in the screenshot below, when configuring an acceleration task in MAX, you can supply values for the accelerometer's sensitivity, units, and the amount of current needed by the accelerometer. All of these values can be found in an accelerometer's specifications document. By entering these values, an acceleration task will allow you take the voltage measurements and scale them into g's or whatever unit your accelerometer measures.  This way you do not have to create a custom scale yourself.

Hope this explanation helps.

Regards,

Mallori M.

[rayen]

I am newbie in labviewprogramming. Then I have enquiries about my project which is using labview. I amusing impact test with NI DAQ 6251, impact hammer B&K 8200 andaccelerometer B&K 4371 then attach it in a structure. I have done somenecessary setting and it works. I can get some graphical result for magnitude (dbchecked) phase, coherence and others are works too.

But, it is very important to meto get the value of the force when I hit the structure with the hammer. I shouldget this force value, to make the amplitude respond from accelerometer relativelycompared by the force.  Can any one helpme what should I do with – especially – the labview program, so that it canresult in the force data. Then, how to get the value of the amplitude per force(m/N, in Y axis) versus the frequency (in x-axis) in a graph/chart. I really needyour help for my research. 

[Ali M]

Hello Rayen,

One thing to take note of is that the NI 6251 does not have simultaneous sampling; therefore there will be a small interchannel delay between taking a sample from one channel and the next. To achieve simultaneous sampling, the following NI devices are recommended: S Series, DSA boards, 5101, 5112, SCXI-1140, SC-2040 and some X series boards. In most cases, your 3rd party hardware Manual/Specification sheet would indicate a Sensitivity spec or conversion factor which will describe what value the voltage produced by your device(s) corresponds to. You can use this data and perform mathematical calculations in LabVIEW to obtain the force data. Also, using the Sound & Vibration Toolkit (SVT) you can obtain the frequency response by performing an FFT analysis from your accelerometer data. In order to take further considerations on sampling your data, I have included a llink below which describes how Interval Scanning works using the M Series card (NI 6251).

1. What Is the Difference Between Interval Scanning and Round Robin Scanning?

http://digital.ni.com/public.nsf/allkb/6E0238CC5D6441CD8025667C0065D939?OpenDocument

Thank you,

Regards,

Ali M

Applications Engineer

National Instruments