Hobbes26

I meant that as a comment to Don's comment about the naysayers that refute the tests that he's done as being non-applicable to the real-world conditions of a car.

LOL... I guess that's why we're having this discussion then! If the results are questioned that only helps us to figure out how to overcome the limitations of the measurements. Although that does make for more work, seeing as you'll probably end up remeasuring everything again.. haha So I assume that you're not aware of any ISO/ANSI/ASTM/etc. standard that has been developed for measuring what we want to measure... I haven't been able to find anything. Perhaps there is something in the european standards (hopefully they'd be in english)? Anyways, as far as variation in samples of sheet metal goes, what frequencies are most affected? At low frequencies where we can excite single modes, is there a lot of variation in those resonance frequencies? What variation can we allow in our measurements? General questions: What frequencies are we most interested in? Lower / Middle / Higher / Full Bandwidth of Human Hearing? Perhaps we can cancel out the higher frequencies, seeing as those will be most easily damped, but we probably need to find an upper limit. And what frequencies are generated in the car - vibration from the road, noise from the exhaust, wind noise, etc... What shape and size of panel (and gauge/thickness) ? We can 'easily' calculate (and model) theoretically the response of a square panel given the dimensions, stiffness and density of the material. At least to make sure we're measuring what we expect. Although I'm not sure how to calculate it when we have the damping material applied. I think we've established that we want to measure both Transmission Loss as well as Damping. I haven't seen the output of a vibrometer - are you able to measure in 2 dimensions? Is there an easy way to show the difference between the damped and undamped measurements and compare them (for the layman (or Lehman ))? And most importantly, can we abstract this to the real world? There will have to be disclaimers regarding stuff such as application of the material onto a clean surface, something on the location of the material in the car, how the car is a complex structure, etc.

I know that - I'm just not sure (haven't looked into it much yet) of the frequency dependency of the string/bungee and what we want to have. Man this is getting complicated... lol.

The book is Beranek, Noise and Vibration I think (it's at work), bright green cover. I didn't think of using sound power for the transmission loss measurement, as i'm only familiar with the method of using two reverb rooms, etc.. But I think TL would be one of the things that we should measure in these tests. A bungee material, at first thought, might damp the vibrations too much...?

?? damping and transmission loss in the same measurement setup? I'm not sure whether we want clamped edges... Not sure what the sound power measurement would be used for... but it would be cool, lol. You have access to an intensity probe? The book i've been looking into suggests merely hanging the sheet metal by a string that is 'long enough' to not damp the sheet any and not allow vibrations to travel up it to the support. Of course, hanging it by a string would require drilling a hole in the panel. What magnitude of differences in sheet metal are you talking about? Will it really be of that much difference to affect the measurements?

Wow, didn't know someone that had access to a scanning vibrometer. That would definitely come in handy. As for sheet metal consistency, I didn't realize that it would be off by that much - I was just assuming that it would be pretty consistent and all we'd have to standardize on is the dimensions and gauge. If we can measure it with and without the damping material added, would that be sufficient to determine the damping ability of the material? The source could possibly be a loudspeaker with a certain bandwidth, sitting a distance away from the sample. I don't see why a reverberant room would help the measurement, since that would mean you're putting the sample in a diffuse field. Unless you're talking about measuring the TL of the sample, with the window. The impact hammer approach would be much better, though setting that up in a repeatable manner seems to be difficult. Justification of the test and abstracting that to the real life situations shouldn't be necessary. Those who dismiss the tests without a valid concern shouldn't even be bothered with. You don't need to be a scientist to understand how a comparative test like this works, and how the results should be interpreted. Don, As far as the heat testing goes, perhaps an infrared thermometer would be useful?

The enclosure he has is hardly sealed airtight... but in other cases, if you DO have a sealed enclosure, it would be a leak....