Tuned Vibration Dampers- Optimal parameters and general discussion 1

[FeX32]

Gents,

I have a general question for the experts here that have tuned or designed engine/crankshaft tuned vibration dampers before. (preferably the kind on the C/S nose)
I have been reading through some of the info on the forum here in this regard and am wondering if someone has some good insight about a comment in this old thread (

http://www.eng-tips.com/viewthread.cfm?qid=260497).

In it about halfway down ivymike says

"I quit reading it after the author claimed that his dampener both absorbed less power and also provided greater dampening. I don't believe the two results are mutually compatible; greater dampening would also imply greater energy absorption.
"
seven years ago when I last did TV analysis on a regular basis, that result was not uncommon. The most effective damper commonly did not dissipate the most heat - they often dissipated less heat than those which were less effective at reducing vibrations. It's a bit fuzzy now, but I think that result was common with elastomeric or viscoelastic dampers, and not so much with viscous ones.

This comment about the TVD that were the most effective being the ones that didn't necessarily dissipate the most energy is counter-intuitive to an extent. I have done a decent amount of theoretical linear TVD work before and can know from that there is an optimal damping that reduces the effectiveness at the frequency of interest but broadens its working range so that it doesn't create 2 ineffective regions below and above this frequency of interest.
Maybe he is really just referring to the damping and not energy - is it possible to look at this from strictly an energy point of view?

Any comments are appreciated.
Thank you.
Fe

 

[GregLocock]

Imagine a simple 2dof system. If the masses are very roughly equal then the attenuation at the initial resonance is excellent, and does not rely on damping. Away from that speed, yes damping is important.

ie it all depends on what you mean.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

Thanks Greg. I know what you mean - basically I am trying to understand more about what detail to look for in TVD development and how important damping is - from an overall performance perspective.
Is there a key analysis you would typically conduct?

 

[GregLocock]

To be honest I just order a selection of geometries and rubbers the supplier recommends for the given crankshaft modes we don't like (bending and torsion), and then either measure TVs or select them by ear. I don't remember actually measuring anything for bending but I suppose crank block acceleration would do. Bending I usually tune by ear.






Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

Thank you Greg.
I didn't realize that these rubber TVD's are also utilized for the bending mode as well. When tuning by ear - do you think this could also be done by noise analysis?
Also, I assume you are testing on engine dyno?
And do you think there is a method to verify performance without having an engine to test on?
Like if you were targeting a certain mode and frequency you could test the TVD separately from the engine to correlate potential performance?
Thanks again.

 

[GregLocock]

"When tuning by ear - do you think this could also be done by noise analysis?" You should see a change in noise or vibration at the bending frequency. Normally i'm a fan of objective analysis but in the case of crankshaft bending modes the right tune stands out.

"Also, I assume you are testing on engine dyno?" No,in car usually.

"And do you think there is a method to verify performance without having an engine to test on? Like if you were targeting a certain mode and frequency you could test the TVD separately from the engine to correlate potential performance? "

In the more general case of tuning tuned absorbers we found that hammer testing the absorber bolted to a large lump of concrete was a remarkably poor indicator of its performance in the car.

I don't know why.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

Thanks Greg.
"In the more general case of tuning tuned absorbers we found that hammer testing the absorber bolted to a large lump of concrete was a remarkably poor indicator of its performance in the car."

What if you excite the TVD torsionally and measure response to correlate with damping - would this be a way to correlate?

And how important is the bending mode in these? Does the supplier tune to this mode? Are the frequencies close to the torsional mode?

 

[GregLocock]

The bending mode is not an objective problem (usually), it just causes poor sound quality. Not every car gets a bending absorber, I'd be a little surprised if most cars did not have a TV damper. You can get away without one in some cases, but they are there for crankshaft fatigue primarily, as well as noise quality. I think your proposed torsional test is the best hope for a rig test, but if you are going to all that trouble why not just stick it on the engine?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

Thank you again.
"Not every car gets a bending absorber,"
So are there absorbers available that are only for this mode, or are torsional dampers tuned to both?
"your proposed torsional test is the best hope for a rig test, but if you are going to all that trouble why not just stick it on the engine?"
I understand what you mean here; but if you don't have the engine to test with (say it's not built yet) then there must be a method to propose an optimal TVD - while only having a simulation with proposed crank geometry.
What do you guys usually use for simulations of this nature?

 

[GregLocock]

In the eighties we used a separate little harmonic damper for bending. It bolted onto the crank nose inside the pulley. However in the nineties Ford Australia started using a design of TV damper in which the rubber was profiled in side view, which meant that for a given TV tune you could vary the bending tune. Very clever.

I don't do NVH any more, I don't know what they use.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GregLocock]

I've just employed a poorly trained CAD guy to draw a TV damper. Changing the nominated radius to infinity or even negative alters the bending frequency of the damper while having little effect on the torsional frequency.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

That's some great info Greg. Thanks again.

That is a clever idea using the curvature to effect the bending mode.
So I can conclude that if we tune the torsional mode out without considering the bending mode it can be a bad thing?
I would assume so.

Do you remember what kind of damping factors you were using? I assume the damping of these devices are just as if not more important than adjusting the natural frequency. Is this generally correct?

 

[GregLocock]

I'd guess we were around 60-80 Shore A, whatever damping that gives. I wouldn't worry about the bending mode, nobody used to. On an I6 it is audible, but not an I4, generally. On the LT5 Corvette engine, which had a crank nose mounted gear type oil pump, the bending mode was strong enough to smash the pump up, so that got a bending damper.

Yes tuning for frequency is much more important than the damping. In the more general case of tuned absorbers too much damping is often a bad thing.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

Thanks Greg.

"In the more general case of tuned absorbers too much damping is often a bad thing."
Can you elaborate? If there is lots of damping it guarantees that the TVD doesn't create another 2 resonance conditions that have high amplitude as well.

What about viscous dampers? They are not tuned for frequency but have high damping across the board?

 

[GregLocock]

I've never sat down and worked it out, but if you have a lot of damping the damper mass can't move as much so it can't absorb as much energy in the rubber, is a handwavy explanation. A few minutes in excel would prove it one way or the other. Viscous dampers are a whole separate thing, although I suspect the same applies in some respects. I haven't tuned one.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FeX32]

hmm. Thanks again Greg.
The viscous dampers seem to just add damping across the board and are maybe only on engines that are highly dynamic. Not sure how true this is nowadays.

 

[FeX32]

Hi Greg,

Another question comes to mind when looking at TVDs.

If I put a certain amount of energy (from combustion) into the crankshaft and it vibrates torsionally at a certain speed there will be a certain energy associated with this vibration. If we now put a TVD (not has less vibration at this speed), does the total energy change in the system? That is, does reducing the vibration translate to a fuel efficiency gain?