I don't have any before and after decibel readings for your car. The fact is' date=' I don't have reliable numbers for any vehicle. That's not for lack of trying...
I've concluded that the only reliable way to measure before and after is to put the vehicle on rollers, in a climate controlled space. This isn't something I'll be able to do anytime soon. In the meantime I am experimenting with a stationary vehicle, inside, with recorded noise, outside of the car. [/quote']
you stated he is the "best in the industry in R&D" yet he admits to not having even the most basic of spaces required for acoustical testing - a functional lab. acoustics labs are a basic requirement for any objective testing. the lack of a certified lab means you are just making statements and observations. automotive manufacturers have acoustic labs they use for testing vehicle noise - hence my statement that he is not the "best in the industry in R&D". and those labs do cost tens of millions to construct - that's not an exaggeration. FWIW, even having the vehicle on rollers wouldn't fully quantify the effects of each different road surface on each tire with the myriad of wear patterns that contribute to noise. if you want to lower road noise, step 1 is to buy better/quieter tires and maintain them properly - but some road surfaces will be louder than others.
no matter what you do to the firewall, roof, door, and floors, the windows and door seals will be flanking paths that will quickly become the dominant source.
using a loudspeaker source in a garage is nothing different than what an average hobbyist can do. simply place mics on the inside in several locations, in several locations on the outside, record all simultaneously and then compare all while taking room/cabin modes and arrival times into consideration. we use these setups at work to evaluate rooms and noise in commercial buildings. when publishing the results as "data" you need to assign a tolerance in the accuracy - otherwise it's useless numbers. the accuracy is called into question because simple testing doesn't isolate each flanking path nor does it quantify the contribution of each flanking path. flanking paths are important and very difficult to quantify - logarithmic contributions are complex.
if you add accelerometers to the panel to do a modal analysis of the surface then you'd be closer to evaluate the benefits of deadener. but you'd need take the same car with multiple doors, install all doors the same (and quantify how they were installed equally), then test each product on each test door while replicating the test each time.
if you wanted accurate equipment (Class 1) then you'd be looking at $20-10k per analyzer and another $20-30k for an accurate noise source - so maybe $100k could get you enough equipment to start testing. Look to companies like Larson Davis, PCB, B&K, etc. who supply legit acoustic labs.
To quantify each flanking path you'd need to build an acoustics lab capable of testing the transmission loss of each component and each system. The toughest part is controlling the noise level on the measurement side to be at or below NCB15 (the high end of a proper measurement lab). the component testing lab itself would be the size of a decent home.
i certainly don't have the time or money to devote to that testing, nor would the sales of product offset the time/money spent. therefore; all that is a waste of time to just sell deadener. automotive manufacturers can justify these test environments.
1. buy deadener, whatever brand.
2. clean the metal surfaces to remove grease/grime
3. apply per mfr instructions
4. knock on the panel and listen for ringing
repeat 3-4 and stop adding when you are happy with how solid it sounds. my personal goal is to sound more like a dense wood.
then i add water-resistant sound absorption in the door to eliminate the reverberant chamber that all doors are.
some products will require more but they cost less. some products require less but they cost more. a few products fall in the "high value" category - meaning they balance cost and performance. i think that is what everyone is truly interested in.
it's a subjective process because most only care about cost and will make any excuse or justification why they don't need to spend more money. a few people are genuinely interested in which product has the most "value" but that is also difficult to nail down.
simply put, there are too many variables to consider and thus there is no equation that can quantify what is "best".
again, i have nothing against SDS products and have nothing negative to say about them. I wish Don the best in his efforts and attempts to quantify what is the right balance of product cost vs. benefit.
every door is different so there is no magical ratio to all vehicles and all products.