ripple/noise and faulty corrections acquiring with PXIE-4496 from impact hammer

[Henrik_Volkers]

First: What do you want to measure?

What are the specs of your hammer (incl. sensor)?

Usually you measure something on the target (anvil 😉 ) too.

The FRF of the 4496 IEPE input for the low frequency is coarsly a 33nF, 10M high pass that coudn't be turned off and should take about 5.5s to settle (it's all in the manual 😉 )

[LAPadron]

I am doing dynamic identifications of beam elements, so on the beam I have several IEPE accelerometers with which I measure  the response to the impact. With impact and accelerations, I build FRFs for different points.

The hammer is a Bruel and Kjaer 8206-001 IEPE impact hammer. 11.4 mV/N, <100 ohms output impedance,

thanks for the 5.5s tip!!

[johnsold]

1. What does the response of a beam accelerometer very close to the impact point look like? Preferably a massive, stiff beam.

2. What does the hammer response look like if you turn it upside down and hit the underside of the beam, striking upward?

Lynn

[LAPadron]

The response of a concrete stiff beam looks, I think, nice (picture below + detail yet below), gets damped and goes back to zero. As for the hammer upside down, I haven't done it. I will do it next week and post. Both hammer and accelerometers are IEPE.

[johnsold]

This makes me think that there is nothing fundamentally wrong with your setup.  Something either mechanically or electrically is slowly resetting to the queiscent state.  It just takes much longer than you are expecting.

Lynn

[LAPadron]

Just as promised, here is the response of the hammer upside down. Sampling frequency is also different (down from 50000 to 25000 Hz). Again, the oscillations are only a 0.2% of the impact magnitude, so I guess no real problem. Thanks.

[johnsold · Accepted solution]

As you say, the signals which you cannot explain are very small.  Unless it becomes a real problem for your analysis, try not to worry about it too much.

Lynn