I have been surprised and a little frustrated that there has been so little proper analysis of noise levels inside the cabin of Tesla Model 3s (and, of course, S and X) and how to manage them. I’ve seen plenty of ad-hoc solutions - extra door seals, application of damping materials - but no complete analysis of the sources of noise, how to measure them and how to deal with them.
I should say that I am no expert on this matter but I have home-built hi-fi speakers and acoustically insulated a very noisy Mini car when I was in my 20s. As an engineer by training and a consultant most of my life I also seek out rigorous assessments of situations on which to base decisions. So I’m hoping that this post will trigger someone with acoustic engineering experience to enhance, modify or naysay my lay analysis of car cabin noise. If nothing else, I hope this post focuses people’s thinking on how to reduce perceived noise levels in the cabin of an M3.
Based on my (limited) experience of soundproofing and damping I would divide methods to reduce noise in the cabin into three categories:
1) Soundproofing: I would define this as blocking sound paths from the outside to the inside of the car but there may be a better, more specific word for it.
Examples might include adding door seals and filling holes around windows or in car pillars.
2) Panel damping: This is the application of energy-absorbing sheet materials to car panels to prevent them “drumming” ie, vibrating in sympathy with, or being stimulated by, external vibrations from the wheels and motors. These sheet materials used to be gooey bituminous sheets but have now been replaced by more modern materials. It’s important to understand here that panels don’t need to be completely covered in such materials. In fact luxury car manufacturers will have worked out exactly how much to apply and where, to prevent 90% of drumming without adding lots of extra weight. You might only need a small square in the middle of a panel.
3) Sound absorption: This is dealing with what is left after you have blocked all direct sound routes into the car cabin and stopped car panels from drumming. Sound absorption is akin to air damping, that is absorbing acoustic vibrations and turning them into heat. Typical materials used to absorb acoustic energy are acoustic wadding and foam - and you need enough of it to make a difference! Thin sheets of sound-absorbing material will have negligible impact. Sound absorption can be divided into two further subcategories: (i) reducing the level of noise passing into the car cabin and (ii) deadening the inside of the cabin which not only reduces noise from the outside but also makes speech and other wanted sounds more intelligible.
Soundproofing and panel damping are probably the most talked about, but not necessarily the best understood, of these three categories - I am fairly sure that people will be applying far more damping material to panels than is actually necessary. But sound absorption doesn’t really seem to be discussed much and yet there seem to be several opportunities to introduce it into the car cabin. For example, I am amazed that there is no underfelt below the carpet in the car. The relatively thin carpet seems to have been laid directly on the (metal?) top of the battery compartment which you can “pop” up and down with your foot. Also, from pictures I’ve seen of a damping material being applied to the doors, there doesn’t appear to be any acoustic wadding inside the doors of the car. So there do seem to be opportunities to improve soundproofing of the cabin although it should be noted that it is more difficult to absorb low frequencies than high ones. My aim here, though, is not to offer specific solutions but to stimulate discussion and perhaps experimentation!
Which brings me to my final point on noise in the M3: how to measure it...
I’ve seen many statements of how many dBs of noise there were before and after attempts at noise reduction and, sometimes, disappointment that the level doesn’t seem to have changed very much. But dB - a logarithmic scale typically used for measuring electrical and acoustic power levels - is meaningless unless it is properly linked to the ear’s perception of sound. For example two sounds of equivalent acoustic power level (dBa), one at 30Hz and the other at 1kHz will have a completely different impact on the listener. In fact the 30Hz may be next to inaudible to many people, depending on the absolute sound level, whereas 1kHz is close to peak sensitivity of the ear so extremely audible. There are several different “curves” that can be applied to acoustic meters to allow for the ear’s varying sensitivity to different frequencies but these are averaging tools and don’t really help in proper acoustic analysis.
All this is therefore to say that rather than a dBa measurement of sound level in an M3 cabin before-and-after application of some form of acoustic insulation, what is actually needed is the
change in acoustic signature - that is, a before-and-after frequency analysis of the noise spectrum in the car and mapping onto how the ear will perceive any change in specific frequency levels. For example, strategically damping the floor of the cabin and placing underfelt under the carpet may well reduce medium to high frequency noise in the cabin but with a lesser effect on low frequencies. Since a lot of car tyre noise is low-frequency, a dBa meter may not show much change in sound pressure level, while there may be a perceptible difference to a human.
Sound signature analysis requires far more sophisticated equipment than cheap dBa meters bought off Amazon. I’m not suggesting we should rush out and buy it. But it would be great if there was someone in the Tesla community who had access to such equipment and might be willing to perform some tests!
Anyway, I’ve said my bit. I hope people looking for ways to reduce noise in their M3s find this simple analysis useful and I look forward to anyone adding a more rigorous assessment and and actual recommendations on the best way forward.