Active Vibration Control

[sweetday]

Hi all.
I am relative new in vibration control. I was given a task to reduce (eliminate-ideal case) vibration that happened in grinding machine.
I used acceleration and force sensors with help modal analyzer to measure it.
It seems like a dumb question but how can I get the displacement data from acceleration. If I integrate it directly, the result is "nothing".
Can anyone help me out from it? It would be great then.

sweetday

 

[GregLocock]

There must be something wrong with either your acclerration signal or your integration process. You need to double integrate, by the way.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sweetday]

Yeah, i mean double integrate.
In simulation it is working, but how bout putting it in reality? I have no idea on it.
Any answers that might help?
Thank you.

sweetday

 

[sweetday]

Another question but still related.
To get the model, I did it in ADAMS and estimated it with the system identification toolbox from Matlab.
To compensate force coming from the process, which generates the vibration, my control strategy is to give force but opposite.
My another silly question is with this additional force fed into the system, would it change the result from system identification toolbox?
Thanks again for any reply. I really appreciate it.

sweetday

 

[GregLocock]

double integrating accelerometer signals is a useful technique in the real world, so long as you watch out for the many real instrumentation type problems.

Sorry, I haven't used the matlab toolkit. I saw it demonstrated, looks great. One day...

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

Applying an equal and opposite force is a nice idea.

In practice, there are some associated challenges:

- Can you buy or make an actuator that responds fast enough?

- What 'immovable' object do you attach it to, and how does that object respond to the applied force?

I.e., the system you have to model may be bigger than you thought it was.



Mike Halloran
Pembroke Pines, FL, USA

 

[svanels]

Some questions:

How long has this problem been going on?
Whas the equipment recently overhauled or relocated?
Are there missing hold-down bolts or other "redundant" parts?
Did somebody changed the operating speed?
A dynamic absorber could help, but you need to find the vibrating frequency.

 

[sweetday]

Thanks to all.

Mike, Yes, reality is much bitter than simulation
Your second challenge is actually my last question.

Svanels, the equipment I think is just fine.
The vibrating frequencies are 70Hz, 120Hz and "others"

sweetday