Accelerometer + LabView + Physics question

[qusaimn]

Hello community, this is going to be a long question so please bear with me while I try to explain my issue. My question is a mix of understanding accelerometer sensors and the physics of how they work.

I have a rotating equipment that has two, 1axis, piezoelectric IEPE accelerometers (AC coupled so they cannot read static accelerations to my understanding) mounted in the radial and axial direction. The rotating equipment itself is at a 45degree angle.

Now, spinning at 60rpm, my centripetal acceleration is as below,

Here is my accelerometer set up in LabView

 

Here is the data from the radial accelerometer at constant speed. This is where my question comes in.

As I mentioned before I have an AC coupled piezoelectric accelerometer which cannot sense static accelerations. Now,

1) I do not expect 2g vibration in the radial direction but I am seeing  periodic ~+-2g in my data which makes me think am I seeing centripetal acceleration as calculated in the picture above. 

2) Now , since it is spinning its possible that  gravity is being sensed due to changing direction (gravity in negative direction to accelerometer at 0 degree and positive when the accelerometer is at 180degree)  ,but then I would expect it to be a periodic ~+-0.7g (because g*cos(45)) since the centripetal acceleration is constant at constant speed (lets say 60rpm) and my accelerometer should not be able to record it.

3) What I would expect seeing is just a roughly flat line with some vibrations in the equipment due to whatever reason, since the accelerometers can neither sense centripetal acceleration nor gravity.

Am I missing something in understanding piezoelectric accelerometers, is my LabView set up wrong, or is my understanding of the physics flawed? I have also attached the accelerometer manual for reference.

[Bob_Schor]

What is the orientation (with respect to gravity) of the axis of rotation?  If it is anything other than "pure vertical", you should expect to see the "rotating gravity vector" (relative to the accelerometer's axis) being transduced.

Bob Schor

[qusaimn]

Hello Bob. Thank you for the response.

So if it is sensing the changing gravity vector it should be reading between +- g*cos(45) correct? The axis of rotation is at 45 degrees w.r.t gravity.

[Bob_Schor]

---

@qusaimn wrote:

3) What I would expect seeing is just a roughly flat line with some vibrations in the equipment due to whatever reason, since the accelerometers can neither sense centripetal acceleration nor gravity.

---

If the accelerometer is rotating at a uniform velocity in such a way that its "axis of sensitivity" was not subject to a changing gravitational acceleration (such as taking place in "free fall" or when the axis is aligned with the "gravity vector", you should get a flat line.  Think about the two extremes -- pointing the sensitivity axis straight up or down (gravitational component unchanged -- always "up" or "down") and axis in a plane normal to gravity (so during one revolution it goes from "up" to "horizonal" to "down" to "horizontal").  So with an axis of 45°, you have component of Earth's 1g gravity vector (0.71 g) varying at the speed of rotation of the axis.

Draw yourself a picture ...

Bob Schor

[mcduff]

---

@Bob_Schor wrote:

---

---

I would be surprised/stunned if the actual accelerometer was rotating with the equipment.

Rotating equipment imparts vibrations to the surrounding; that is what your accelerometer is detecting.

There are some inconsistencies with what is written and what you have measured. You said your equipment is rotating at 60 RPM, or 1 Hz. The graph you have shown seems to have higher frequencies if the x-axis is time.

Take a FFT of you data. What do you see?  Do you see a vibration that matches the rotational frequency? Do you see overtones?