LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Impact test in concrete beams with accelerometers

Hi,

 

I am still a bit struggling with the acceleration readings in an impact test of concrete samples. The beam has natural frequency (bending mode) of 545 Hz. I am using NI6361 card for voltage measurements. I am connecting a single ended 1-axis accelerometer which is already calibrated. Now, I would like to get the voltage measurements and post-process it. I have tried to scale the accelerometer readings as well.

 

I have tried the following techniques with the settings applied on the power supply which provides the current and excitation (Attached document).

 

1) Measure the voltage (+/- 10 V corresponds to +/- 10,000 g) and post-process the data to obtain the values in m/s^2.

 

2) Scale the voltage to acceleration directly using MAX.

 

3) Convert the voltage to acceleration using an in place element structure (Despite the fact that this is not suitable for my application, as I have other sensors)

 

The accelerometer readings seem to be very sensitive. I would like to ask what is the most appropriate way to set it up correctly or if I am doing something wrong.

 

Regards,

 

0 Kudos
Message 1 of 29
(4,304 Views)

Here are the results I got for a single touch in the accelerometer, very unusual as the capacity is +/-10,000 g. Y-axis is in g and x-axis in seconds. 

undefined

0 Kudos
Message 2 of 29
(4,222 Views)

@StathPol wrote:

Here are the results I got for a single touch in the accelerometer, very unusual as the capacity is +/-10,000 g. Y-axis is in g and x-axis in seconds. 

undefined


For what it's worth, it didn't go over the max max which is 12k.  You might be doing it correctly.  If you "touched" it with your finger, it's definitely NOT right.  If you tapped with an allen wrench or something hard, it could generate quite a high acceleration for a fraction of a second.

Bill
CLD
(Mid-Level minion.)
My support system ensures that I don't look totally incompetent.
Proud to say that I've progressed beyond knowing just enough to be dangerous. I now know enough to know that I have no clue about anything at all.
Humble author of the CLAD Nugget.
0 Kudos
Message 3 of 29
(4,191 Views)

I appreciate and many thanks for your reply, as I am stuck on this problem for a couple of months now :(.

 

The 1-axis IEPE accelerometer sensor goes to power supply, NI6361 and then to PXI. We know that the charge produce by this piezoelectric accelerometer is in the millivolt range. The accelerometer is connected with a charge amplifier to decrease the noise. The accelerometer requires an external current to be supplied and the excitation comes from the coupler (power supply).

 

I have done a small test today by hitting the table where the accelerometer was fixed on a bracket, before I go to the large scale application. I set the properties on the power supply as follows:

 

Sensitivity of the accelerometer = 0.532 mV/g

Range 10 V (g) = 10,000

Gain = 1.93 (automatically adjusted)

Low Pass Filter (-3 dB) = Bypass

HP Filter (-5%) = 0.5 Hz

Current (mA) / Bias (V) = 4 / 11

Units = g

Volts Full Scale Output = +/- 10 Volts

 

I have tried to record the voltage readings, and then, divide the voltage with the sensitivity of the sensor to get the g force:

 

undefined

The 1st option worked, whereas the second did not work. The graph I got by reducing the gain is shown below. 

 

undefined

 

Since the properties are all set in the power supply (coupler), do we need to specify these properties in the VI as indicated in the 2nd Vi as well ? or just create an analog input voltage task as indicated in the 1st VI?

 

Many thanks for your reply. 

Regards.

 

 

 

 

 

 

0 Kudos
Message 4 of 29
(4,164 Views)

Have you tried to catch the signal with an O-scope?  That would definitely determine if your hardware is set up correctly.

Bill
CLD
(Mid-Level minion.)
My support system ensures that I don't look totally incompetent.
Proud to say that I've progressed beyond knowing just enough to be dangerous. I now know enough to know that I have no clue about anything at all.
Humble author of the CLAD Nugget.
0 Kudos
Message 5 of 29
(4,132 Views)

Keep in mind: The sensor will report what it feels. That migth or migth not be the signal you are looking for!

See the sensor as a simple 1DOF system (seismic mass, piezo as a spring with small damping)  resonace @ 100kHz according to your datasheet

AND

See the coupling  of the sensor as another spring with the mass of the sensor ....

usually you want this spring to be as stiff as possible. see application notes on sensor mounting techniques

 

If have acess to the university LabVIEW distribution have a look at the sound and vibration suite you maybe can use and try the shipped exsamples!

 

 

 

 

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 6 of 29
(4,113 Views)

Thanks a lot for your replies.

 

I have used an O-scope on the hardware, and it works fine. Only in the hardware though.

 

Henrik_Volkers...I do have access yes..Thanks for letting me know for the examples..I will have a look at them.


Many thanks.

0 Kudos
Message 7 of 29
(4,102 Views)

@StathPol wrote:

Thanks a lot for your replies.

 

I have used an O-scope on the hardware, and it works fine. Only in the hardware though.

 

Henrik_Volkers...I do have access yes..Thanks for letting me know for the examples..I will have a look at them.


Many thanks.


That step was very important to do because it confirms that the hardware is working correctly.  You can't know it's a software issue until you've ruled out a hardware issue.  🙂

 

Now you can go ahead and make tweaks to your software until the results align.

Bill
CLD
(Mid-Level minion.)
My support system ensures that I don't look totally incompetent.
Proud to say that I've progressed beyond knowing just enough to be dangerous. I now know enough to know that I have no clue about anything at all.
Humble author of the CLAD Nugget.
0 Kudos
Message 8 of 29
(4,077 Views)

Henrik_Volkers and Bill, 

 

I highly appreciate your replies in my problem and many thanks for this. For the hardware checks, an O-scope in conjunction with a signal conditioning device was used to check the hardware. It is fine as it was set to provide a 5 Volts signal.

 

I have done a small test by holding the accelerometer downwards (and dropping it in a sponge) where it is assumed to give 1g because of the gravity; however the result I got was 300 g. I firstly got the voltage signal and divided it with the sensitivity of the sensor. I set the sensitivity of the sensor and the voltage output +/- 10,000 g on the power supply and then the gain is automatically adjusted. So, 

 

Gain = Full scale output / (Sensitivity*Range)

 

= 10 Volts / (0.000532 V/g * 10000) = 1.87

 

I also have attached the settings I used for the power supply. In the specifications, it says that a minimum of 2 mA is recommended for general sensor operation and I used 4 mA as for IEPE sensors, current can be between 2-20 mA. 

 

I also have attached the VI I am using the acquisition with the data transferred to TDMS file.

 

Thanks for your kindly replies.  

0 Kudos
Message 9 of 29
(4,060 Views)

undefined

0 Kudos
Message 10 of 29
(4,057 Views)