Acoustic Analysis of Chassis

[Ravikiran027]

Can anyone help me to perform acoustic analysis of CHASSIS.

Thanks

[Ravikiran027]

Can anyone help me to perform acoustic analysis of CHASSIS.

Thanks

[Mbirt]

http://tuningforkswholesale.com/images/tuning_fork_set.jpg

[Ravikiran027]

Thanks for your reply Sir...

Could you please help with the procedure to perform acoustic analysis in ANSYS using elements fluid 30 and fluid 130.

Thanks in advance.

[GianVioli]

I sugggest do some research on your own first, this forum isnt for hand outs

[Ravikiran027]

Thanks for your reply sir.

I have gone through many technical papers from SAE, IMECH, Science direct and others. Also studied the Ansys Help Documentation (also examples on Acoustic Analysis). However, I did not get idea of where to create the acoustic element in the FSAE Chassis model. By studying the acoustic analysis example in ansys, I understand that i have to choose an element or create an element (fluid30) exhibiting the acoustic properties and over that an absorbing element (fluid130) needs to be created.

I am confused of where to or how to create the acoustic elements. please suggest me.

Thanks in advance.

Ravi

[Drew Price]

I am confused why you want to perform this type of analysis?

[Kirk Feldkamp]

I'm going to go out on a limb and guess that "acoustic analysis" has somehow been mis-translated into "resonant frequency" or something like that. Ravi, can you explain exactly what you're trying to figure out from the analysis you're trying to do?

[Price]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I am confused why you want to perform this type of analysis? </div></BLOCKQUOTE>

Obviously he wants to tune the chassis. It seems like most people in the US go for A 440, but we have had good luck with A 442. A little higher up the scale, but the extra 2 Hz really seems to make a difference in the transient response.

http://fsae.com/groupee_common/emoticons/icon_razz.gif

[swong46]

My guess is to study the load paths and the tension/compression forces of each tube when the chassis is torsion tested?

Edit: Nvm hah

[BillCobb]

Look for information on "normal mode analysis". Its rocket science. Having nodes at attachment points is probably what your job is all about.

(Unless your job is all about "lead", in which case your job is all about turning lead into gold). Use the PB2Au transform in Matlab to get that job done bestly.

[Ravikiran027]

Hi,

Thanks for each and everyone for your concern.

Actually Its my Masters Project Title. I choosed FEM and Ansys as my interisting topic and University allotted me with the title "Acoustic Analysis of a Student Car using FEM". I have to perform acoustic analysis and find out the plots of vibrations and noise levels at different frequencies.

Regarding the acoustic analysis... there are two examples under the fluids topic tutorials with the exact name "Acoustics" (chapeter 15) and explained briefly about acoustic analysis.

Any suggestions or guidance appreciated.

Thanks
Ravi

[Adambomb]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Ravikiran027:
Hi,

Thanks for each and everyone for your concern.

Actually Its my Masters Project Title. I choosed FEM and Ansys as my interisting topic and University allotted me with the title "Acoustic Analysis of a Student Car using FEM". I have to perform acoustic analysis and find out the plots of vibrations and noise levels at different frequencies.

Regarding the acoustic analysis... there are two examples under the fluids topic tutorials with the exact name "Acoustics" (chapeter 15) and explained briefly about acoustic analysis.

Any suggestions or guidance appreciated.

Thanks
Ravi </div></BLOCKQUOTE>

Just to make sure there isn't something lost in translation here, you decided that you wanted to do your Masters research on FEM using Ansys, and the university just gave you a title for your project??? As a fellow graduate student I can tell you that does not sound like a promising method for doing research. Surely there has to be someone there who can actually tell you something about what exactly you are trying to find out, some sort of goals or something. Just deciding that you're going to do a random FEM study on an isolated component removed from its natural environment sounds like something that is nearly guaranteed to not yield usable results.

I think BillCobb has the best match of application to the analysis you want to do. I made the mistake of bringing up the topic during an Optimum G seminar, and within about 15 seconds Claude Roulle had convinced me that it was something I didn't want to go near. Not that it was a bad study, but that it was very very involved; not the kind of thing you could get any kind of results from by doing an FEM on a single component. You would also really want to be intimately intimately familiar with vibrations as well as vehicle dynamics in general.

[AxelRipper]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Price:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I am confused why you want to perform this type of analysis? </div></BLOCKQUOTE>

Obviously he wants to tune the chassis. It seems like most people in the US go for A 440, but we have had good luck with A 442. A little higher up the scale, but the extra 2 Hz really seems to make a difference in the transient response.

http://fsae.com/groupee_common/emoticons/icon_razz.gif </div></BLOCKQUOTE>

You know what they say: sharp is fast!

[Zac]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by AxelRipper:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Price:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I am confused why you want to perform this type of analysis? </div></BLOCKQUOTE>

Obviously he wants to tune the chassis. It seems like most people in the US go for A 440, but we have had good luck with A 442. A little higher up the scale, but the extra 2 Hz really seems to make a difference in the transient response.

http://fsae.com/groupee_common/emoticons/icon_razz.gif </div></BLOCKQUOTE>

You know what they say: sharp is fast! </div></BLOCKQUOTE>

Truth. You definitely don't want to be flat.

[Drew Price]

You are asking for help and guidance with your masters thesis... through an internet forum.... populated by undergrads.

Interesting.



I do not know of anyone who has done a mechanical vibrations analysis of their frame, let alone components actually subjected to oscillating / vibratory loads... like engine mounts. If you break a mount in racing, one of two things has happened:

1) The mount was too weak in the first place / something was designed stupid (bolt too small, wrong grade fastener, no damping component, not enough thread engagement, not enough clamping force, etc)

2) Something got fucked up in assembly (too tight, too loose, too many super-mechanics)



(hehe.... I said vibratory)

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I do not know of anyone who has done a mechanical vibrations analysis of their frame, let alone components actually subjected to oscillating / vibratory loads... like engine mounts. </div></BLOCKQUOTE>

Every OEM auto company in the world performs this sort of analysis.

[Nick Renold]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I do not know of any Formula SAE Team who has done a mechanical vibrations analysis of their frame, let alone components actually subjected to oscillating / vibratory loads... like engine mounts. </div></BLOCKQUOTE>
I think this is what Drew meant.

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Nick Renold:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drew Price:
I do not know of any Formula SAE Team who has done a mechanical vibrations analysis of their frame, let alone components actually subjected to oscillating / vibratory loads... like engine mounts. </div></BLOCKQUOTE>
I think this is what Drew meant. </div></BLOCKQUOTE>

I have seen a few teams in design semis that have modeled and physically validated the first torsional and bending modes of their vehicle structures.

[Wesley]

I wonder what the hoop wave analysis would look like in a steel tube frame....

[Rajesh]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Ravikiran027:
Thanks for your reply Sir...

Could you please help with the procedure to perform acoustic analysis in ANSYS using elements fluid 30 and fluid 130.

Thanks in advance. </div></BLOCKQUOTE>

[Big Bird]

I must have missed this one when it came up.

We did a full normal modes analysis on our last full spaceframe FSAE car in 2003. Think a little laterally and you can understand why. What two properties are required to calculate a natural frequency, and in what relationship to each other? And are we interested in these properties when designing a race car?

[Big Bird]

OK, no-one is biting so I'll explain myself. Firstly, I could barely give a rat's about the vibrations in the car itself, unless the chassis is so poor it turns into its own oscillating spring.

We generally want our chassis to be:
1. Stiff - although if we used stiffness as our sole design criteria the optimal solution for the chassis would be a 2m x 1m x 1m billet of chromoly
2. Lightweight - although if minimal mass was our sole criteria we can easily end up with a flexi-flyer.
The goals of low mass and high stiffness directly conflict with each other.

Natural frequency is proportional to stiffness divided by mass (k/m). So you can use it as a metric to measure your "structural efficiency", or how effectively how you are obtaining stiffness without adding in too much mass. The higher your natural frequency, the better.

Whats more, the animations that pop out of a normal modes (natural frequency) analysis are brilliant for assessing your chassis design. They will show if your chassis is least effective in a twist mode, or a bending mode, or a "breathing" mode (where a part of the car expands and contracts like your lungs breathing). Whats more, the animations show you what part of the car is "weakest", as it is often obvious which part of the chassis is flexing most for each of the modes. So you can focus in on the weakest area, revise design, re-analyze, etc etc.

So normal modes analysis can be a useful tool to add to your chassis design skillset. You just have to realize that the value in the exercise is not necessarily about the vibrations themselves.

Hope that helps.

[ed_pratt]

Ah Geoff!

You're giving away our valuable secrets!

In reality I would guess(hope?!) loads of teams are doing this. As you've said, it's a really neat way of analysing the "structural efficiency" of the design and the modal animations are pretty intuitive - they say a picture is worth a thousand words, how much for a video?

Ed