Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Interpreting Modal analysis results with excitation frequencies

Status
Not open for further replies.

Domainpower

Mechanical
Sep 13, 2012
20
Dear all ,

I have done modal analysis of a turbine using Finite element analysis and as output i have obtained a set of natural frequencies.

Now regarding excitation frequency , i have this turbine rotating at a fixed rpm while the water jet will be acting on it at different positions as shown in the attached picture.
If the rpm of machine is 400 and has 6 water jet excitation's , for the outer blades the excitation frequency will be = (400/60)*(6) = 40Hz.
Hence the fundamental excitation frequency for outer blade becomes = 40 Hz
and its higher harmonics will be = 80,120,160Hz.........

Now the question is to evaluate dynamic amplification of stress due to resonance , do i need to include all the natural frequencies where i am close to the excitation frequencies
OR
Depending upon the type of the excitation that is 6 water jet in this case that some natural frequencies even close to the excitaion frequencies will not be able to excite the system or May be a mode shape with a particular nodal diameter can have a greater importance.

Please guide.



 
Replies continue below

Recommended for you

Your situation is similar to a 6 cylinder in line engine. There is a temptation to assume that axial or lateral modes may not be important, and that forcing can only occur at multiples of the total excitation frequency. In practice you will see variations in the excitation from each nozzle, so you will get all harmonics from 7 Hz upwards.


Obviously if the excitation from each nozzle is not exactly equal then lateral modes can be excited.

Sorry, there are no shortcuts.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Thanks GregLocock for your reply.

few clarifications:

Just for my understanding for a moment if i ignore my axial and lateral modes and focus only on the bending dominated modes.

I will separate a set of bending dominated natural frequencies from the full output of modal analysis , ranging from global bending (Lowest frequency ) to very local bending(highest frequency).

Then my question is (considering only bending modes) : Is it possible to ignore some bending modes on the basis of things like it is excited by 6 jets in one revolution or had it been one water jet per revolution.

I am attaching one relevant paper which was published for this product , please refer to page 4, it seem to link nodal diameters with the number of excitation jets to find out the dominating bending frequency while ignoring the rest.

It will be great if you could as well refer me some paper or book which can help me better understand this.



 
 http://files.engineering.com/getfile.aspx?folder=ba9b5edd-ac85-4540-8d37-206643759004&file=Reference.pdf
It appears the paper takes an approach equivalent to 6 jets per revolution (40hz as lowest excitation frequency). Greg's point is valid if there is a significant difference among jets (could be velocity, perhaps difference in angle).

If the force from 5 of the jets have identical magnitude 10 and the 6th jet has magnitude 11, then this force set could be represented as a superposition of excitation of 40hz at magnitude 10, plus 6 hz at magnitude 1 (with proper consideration of phase when add the resulting responses).



=====================================
(2B)+(2B)' ?
 
can anybody give some insight into the above posted question:
considering force from all 6 jets is same

[highlight #EF2929](considering only bending modes) : Is it possible to ignore some bending modes on the basis of things like it is excited by 6 jets in one revolution or had it been one water jet per revolution.

I am attaching one relevant paper which was published for this product , please refer to page 4, it seem to link nodal diameters with the number of excitation jets to find out the dominating bending frequency while ignoring the rest.

It will be great if you could as well refer me some paper or book which can help me better understand this.[/highlight]
 
I can't because my experience tells me that ignoring variations in excitation is a damn fool assumption. Hopefully(?) you'll find someone with less experience to tell you otherwise.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Thanks Greglocock for your reply,I am new to vibrations and will go with your experience.

Now when i take all the modes into consideration,
what are the potential points i look into my Campbell diagram (from FEA) in addition to resonance frequencies?

In other words i am not sure what impact the magnitude of excitation force (also the number of excitation per revolution), the nodal diameter of the mode shapes and the harmonics of the excitation frequency have.





 
The elephants in the room are the damping of each mode, and the degree of nozzle to nozzle variation, which are inputs to your dynamic model which FEA cannot help you with.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Dear GregLocock , From the testing team i got the input that maximum variation for nozzle to nozzle excitation can't go beyond 1%.

For damping ratio values we have done hammer test on the bucket to get the individual values for each natural frequency.

Now i have to use these damping ratio values to estimate the amplification of the stress.

But before proceeding i wanted to understand what impact the magnitude of excitation force (also the number of excitation per revolution), the nodal diameter of the mode shapes and the harmonics of the excitation frequency have
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor