Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Moving table transient analysis 1

Status
Not open for further replies.

elogesh

Mechanical
Joined
May 10, 2002
Messages
187
Location
IN
Hi,

For one of our customers, we are modifying existing structural unit as a part of value engineering.The Structural unit is used for scanning.

In side of the unit consists of movment of table in X and Y direction.

We did static analysis of the structural unit for 4 diferent extreme positions of the XY table using FE calculations.

Then customer requested dynamic analysis of the structural unit. This is important, because after the movment XY table to one position to reaching another position, scanning starts. But it takes some time for the system to settle down, so that vibration due to XY table movement dies out.(Reaching the end position and start of the scanning). Customer has given following parameters in this regard,

Max velocity in XY direction = 16 in/sec

Acceleration = 0.72 G.

How to proceed with the analysis? We need to determine, the time required for the vibrations to decay out of the order of 0.001 in.

Is it possible to keep the table at one location and apply the specified velocity and acceleration inputs in to the table simulating the dynamic characteristics?

Or otherwise, I need to move the table with respect to time (Moving load analysis), to evaulate the dynamic characteristics?

Regards,
Logesh.E

 
The crucial physical property is damping, which will probably be non linear.

I suspect you will find that the decay time constant is similar to the existing product, so use the physical measurements from the existing machine to guide you.

I think you could probably get away with a stationary model, but finding the right force signal to apply might be difficult.

Tools I would use, in order of effort and usefulness:
lots of real world measurements,
SDOF model,
MBD analysis (ie ADAMS or Nastran 4D),
stationary dynamic model,
MBD with flexural elements

Since the damping is the most important characteristic and the hardest to model I doubt you will make significant progress without the real world tests.

Is this supposed to be a quick check, or a full blown analysis?



Cheers

Greg Locock

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

Using the scanner itself can you scan an "x" edge, a "y" edge to see if they are at "different" locations w.r.t time.
If they are then the table has not settled down yet.

Empirical vibration measurement? - attach strain gages to the platform.
 
Hi greg,

Thanks for your comments.

I will definitely take due importance about your comments on damping.

But unfortunately, customer doesn't have the test data. He knows the system takes several orders of miliseconds to settle down.He has to scan a whole lot bunch of components inside the unit. For every component scanning, table moves morethan 10 times. In this sceanrio, the settling time becomes a important parameter for the overall productivity.

1) Before start of the project itself,we emphasized to the customer,that like friction, damping also depends upon the physical reality and simulation is not going to help with out having some data about the damping characterictics. But the customer has some difficulties in carrying out the tests immediately.In this scenario, how can I help the customer, before the measuments are being taken?

But I will keep emphasizing to customer about the testing to determine the damping levels.

2) "I think you could probably get away with a stationary model, but finding the right force signal to apply might be difficult"

We don't have MBD software, hence probably we want to proceed with SDOF model or stationary dynamic model.

In this scenario, how to proceed further. Do you have any recommendations for feeding the velocity and acceleration values in to the FE model for the determination of the response of the system?

3) It has to be quick check.

Regards,
Logesh.E






 
Can you get them to send you at trace of the output of the machine as it settles down? That will at least give you an idea of the exponential decay rate.

I'd go back to a SDOF model and look at what you need to change to get a given peak delfection after a certain time.

The answers (off the top of my head) are few, probably the easiest is to reduce the impulsiveness of the force, ie why not apply some elctrical damping to the drive signal - a soft stop algorithm?

Or, increase the mass of the measuring head of the instrument, if it is compliantly mounted.

Or, increase the stiffness if it is a stiffness defined system.

Or increase the damping - use cast iron for the metal parts in place of steel, consider appropriate use of damping elsewhere.

Alternatively add a harmonic damper. This would be very cheap and easy in mass production, rather a pain to do for a single unit. The maths is easy, the practicalities less so. A dashpot is a nice thing to use for protoypes as it does not need accurate tuning.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top