LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Acceleration to displacement problems

Hello, 

 

I have collected data from a single axis accelerometer, i now need to analyse this data for a project but i am having some difficulties. This is on a tight time scale so any help is greatly appreciated!

 

The data is from a high velocity impact experiment where a small lead round has been shot from an air rifle into a composite material. I need to gather the velocity and displacement graphs from the acceleration data provided. I was initially thinking that i would produce graphs that looked very similar to that shown in the picture attached (from another post). However, the results i am producing look nothing like this and im finding it hard to explain them. 

 

Can any give advice as to whether what i have done is correct or incorrect? and how to fix it if needed?

 

Thanks in advance. 

 

 

0 Kudos
Message 1 of 14
(3,694 Views)

Hi hamish,

 

a quite simplistic approach:

check.png

No idea why you calculated the PSD…

Best regards,
GerdW


using LV2016/2019/2021 on Win10/11+cRIO, TestStand2016/2019
0 Kudos
Message 2 of 14
(3,679 Views)

Thanks Gerd, 

 

Could you please send me the flies for what you created? 

 

0 Kudos
Message 3 of 14
(3,668 Views)

Hi hamish,

 

unfortunately I didn't save the VI after I made changes.

But the block diagram is fairly simple, so you can recreate it easily!

Best regards,
GerdW


using LV2016/2019/2021 on Win10/11+cRIO, TestStand2016/2019
0 Kudos
Message 4 of 14
(3,653 Views)

Some remarks:

~10kSPS is to slow for such high bandwidth impacts .. 100kSPS or more would be needed.

I do shock calibrations in the range up to 100k m/s² and use samplerates >= 1MSPS

OK, pusle durations could be as low as 40µS

 

What type accelerometer did you use ? Signal conditioner? DAQ?

How is the accelerometer mouted?

 

And if you want to do something meaningful with the data .. why you use g (gram? 😉 )  instead of SI units (m/s²) to convert it to velocity and displacement?

dat.png  

 

IN short: That data is useless unless you only want to detect the impact time with a resolution of about 0.1ms

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


0 Kudos
Message 5 of 14
(3,651 Views)

Hi Henrik, 

 

The hardware I'm using is: 

 

Accelerometer - https://www.bksv.com/en/products/transducers/vibration/Vibration-transducers/accelerometers/8339

NI cDAQ 9185 

NI 9230 sound and vibration module.

 

Accelerometer was screwed into the composite material, roughly 50mm away from impact area. 

 

When I tired to up my sampling rate above 12K I received errors saying that the hardware had maxed out sadly.

 

I will take your advice and change the units to m/s^2. 

 

What are your thoughts on what the velocity and displacements graphs should look like?

 

I realised that I attached the wrong VI in my original post, I have attached a new one that I plan to use, it incorporates the SVL integrators, do you have any further advice? 

 

Thanks!

Download All
0 Kudos
Message 6 of 14
(3,645 Views)

Your biggest problem is the high bandwidth of the shock and the accelerometer output.

The (first) resonance of the sensor is at about 130 kHz and the sensor will easely output frequencies up to that frequency.

Even if the signal passes an internal (anti aliasing) lowpass filter, the result is not what you want to look at.

Do you have a scope?  To get an idea of the signal you can use the scope in AC mode and 1Meg input impedance and hook it with a T connector at the 9230 input. In that way the 9230 will supply the CCLD/IEPE sensor.

Single shot with 1V trigger positive slope ... 

 

EDIT: Or use an other signal conditioner if available. If you put a T at the input, you should measure a bias voltage of about 11V , and the AC signal sitting on it. 

 

Depending on the material of your composit you maybe have an build in mechanical lowpass filter... but you need to check that.

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


Message 7 of 14
(3,629 Views)

The mounting surface needs to be flat and (best) polished (typical roughtness <1µm) and you must use some grease or oil between the monting surfaces. Mounting torque according to spec 1.8 Nm (not less! )

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


Message 8 of 14
(3,619 Views)

Thanks for the advice but its too late for that now as we have already conducted the testing of our samples. The objective of the experiment was to investigate the impact and shock performance of the composite materials. 

 

I was planning on relating the peak acceleration values and frequency responses to the damage seen in the composite. I.E the more ductile composites should have a larger peak acceleration but a lower frequency than the stiff materials. 

 

What else do you think we can do with our data? Hopefully there's something 

 

 

Thanks

0 Kudos
Message 9 of 14
(3,614 Views)

Peak value seems the only thing at a fisrt glance.

You can try a peak search with 3 or 4 points width , that matches a square function on the peak over 3 resp 4 sample ..  results in different values

peaks.png

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


0 Kudos
Message 10 of 14
(3,604 Views)