Dynamic Absorber

[fire911]

We have a 20,000lb vertical pump/motor with vibes over 0.38 ips in the horizontal direction from running at the natural frequency of the system. We tried adding mass and were successful to a point, but needed more then we thought prudent. We are discussing dynamic absorbers and I was wondering how to guesstimate the weight I need on the absorber which will have minimal "whip". I know how to calculate the support length and I realize I can split the weight into a couple of Absorbers to reduce the weight on a single support. Also...any good references I should look at to design a dynamic absorber? Thanks for the help.

stay safe;
paul

 

[electricpete]

There used to be a great article on building a dynamic absorber at

http://www.entekird.com/eact/eactpapers/eactpapers.htm

Rockwell has taken over Entek and I can't find the "Enteract" conference articles anymore.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[ebarba]

I think you might want to check out Den Hartog's book on vibration, published by Dover... it's practical, sound and straightforward, despite it's an old text book.

I designed various dynamic absorbers for cases up to 25000 lbs and as a rule of thumb you could use an absorber mass of 5 to 10% total mass. With a 3% added mass, it starts to be a bit difficult to tune the absorber, but it's still possible to get good results.

Good luck!

 

[fire911]

Thanks for the responses.

I did find the Entek article at

http://domino.automation.rockwell.c...ntekWebST.nsf/files/Fox00.pdf/$file/Fox00.pdf

but there appears to be mistakes in the formula (unit analysis shows a discrepency).

The rule of thumb was just what I was looking for, thats going to be alot of weight. I will have to see if I can find the Hartog text.

stay safe;
paul

 

[electricpete]

I talked to the author of that paper (Randy Fox).

On page 7 there is a formula
W2 = stuff - otherstuff

The otherstuff has a factor of L in the numerator. If you change that L to L^4, the formula should be correct.

I am impressed that you found the file. How did you find it? Is there an address where all the Enteract files are available?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[fire911]

Thanks for the formula correction. That makes more sense to me, I have made it a habit to check the units of formulas and this stood out.

I found the Entek article using a Google search for "Dynamic Absorbers". I am sorry to say I do not know where the rest of the articles went.

stay safe;
paul

 

[ebarba]

Dear Paul,

less added mass means a more difficult tuning... you could also use 1% mass, but that would mean you should devise a means of fine tuning the mass position (i.e. a screw to precisely position the resonant masses). Remember that, from the tuning position, you will have two resonances really near: one to the far side of the rod and the other to the near... a difference in postion of just 1/32" could change the state from total vibration absorption to resonance!

The difficulty increases when using more than one added mass. In this case you should tune the first one, then fasten it (prevent the mass from oscillating) then proceed with the next one, fasten it and so on.

On the basis of experience I strongly encourage you not to go under the 3% ratio if you do not want to have seriuos headaches during tuning... 5% if you do not have precision vibration measurement equipment.