How to measure cabin noise?

[tirwin]

I just got home from a ~800 mile trip in the '83. The car ran like a top the whole way.

One thing I wanted to try to do was to measure the cabin noise. I have an app for my phone (dB Meter Pro). At highway speeds -- 4000rpm in 5th -- I measured 72dB (average). At idle, I measured 62dB.

When I got home, I did a little research and the Internet tells me that some of the most quiet cars measure 72dB at ~60mph.

Huh? That doesn't make sense to me because my exhaust is LOUD. I wore ear plugs for the trip.

At first, I thought there is no way my measurements could be right. But then I read a little more and what I found is that the dB measurement doesn't really tell the whole story. In order to get a better picture of cabin noise, you need to understand the sound frequencies in use.

I'm just wondering -- what is the best way to measure cabin noise?

Anyone have any thoughts on the matter?

[cabmandone]

My thought would be that if you need to wear ear plugs while driving... it's pretty loud inside the car.

[Duc Hunter]

Well I use an iPhone app called decibel I think. I also try to have the phone up near ear level and teh picrophones (bottom of the phone) uncovered and pointed toward the back. Mine I think is louder than 72db.

[tirwin]

Nick,

Yea, it is very loud. Outside the car the exhaust measures mid-80s dB. Of course, the sound of the car is the #1 compliment I get.

What I’m starting to think about is if there is any way I can possibly have my cake and eat it too.

But I need a way to objectively measure cabin sound levels. Just wondering if my simple approach is fundamentally flawed.

Chris, I’ll try your suggestions and see if it makes a difference.

[AG81]

I have a close friend who works for Kenwood. Greg always said that frequency, to him, is the most annoying factor....

[Bill Douglas]

It was interesting. I wore my gun club earmuffs (class 5) in the 911 because the dogs start barking at full volume when I get near my sister's place. And I found it was quieter of course, but I could hear different 911 noises. All OK, just the exhaust and road/tire noises went away.

[Canada Kev]

Sound intensity (dB) and frequency is a funny and interesting thing. I'm by no means an expert, but some of the things I've learned in the telecommunications world might apply here.

When you are measuring dB with a simple meter, it likely is measuring at 1000 Hz or 1 kHz. This is a standard frequency that many things use as a reference point.

Frequency will affect how you will perceive the level of loudness to be. Our ears are at their most sensitive from about the 2 kHz to 6 kHz zone. For example, that 72 dB sound intensity measured in the quietest cars, if it is measured at 1kHz, will seem as loud as a 75 dB sound if its frequency is doubled to 2 kHz.

At about 3 kHz, our ears are usually about their most sensitive and that same sound, if its intensity hasn't changed, will seem to our ears to be almost 80 dB.

So with this in mind, if the sound, regardless of what it is from, can be modified so that its frequency can be modified to be outside of that 2 - 6 kHz zone range, the car will be perceived as being quieter. In addition, if any sound attenuating material installed in the car can be tuned to work in that same frequency range, it will allow less sound of that frequency enter the car, thus feeling quieter, even if a dB meter reading levels at 1 kHz says its not much lower.

Keep in mind that the power level of decibels is doubled after an increase of 3 dB. Thus the power difference between 72 dB and 75 dB is a factor of two. And then doubled again when you reach 80 dB. Then you can throw in phons and sones to really muck things up.

Like I said, it's all very interesting and weird. It's almost enough to drive one to drink. So now it must be beer o'clock...

[Duc Hunter]

On long road trips (like the 6 hour run down to DRT in Feb) I wear my Sony noise canceling headphones in my Rat Rod. It's not that loud, but 6 hours of freeway.....ugh. As mentioned above you hear different noises (my wheel bearings/tires, etc). It sure makes the trip very nice. So I am looking into a bluetooth headset setup for that car. Run noise canceling headphones with microphones to take calls and be like a rally intercom. Over kill for many, but fits my car.

[tirwin]

What you guys are saying seems to make a lot of sense.

I don't think my 62 dBa @ idle/ 72 dBa @ highway measurement was really reflective of what I was actually hearing -- it had to be louder. So most likely a problem with my method and the equipment. Those numbers don't really jive with this chart.



There's probably a lot of reasons why my simple test wasn't very accurate -- mic quality, mic at wrong position/direction, noise cancelling functionality in the phone, etc. But at least the process was the same. So while I don't think I can trust the accuracy of the actual numbers, I do think the relative magnitude was probably right.

Since the decibel scale is logarithmic, that means that the cabin noise at cruise (72 dBa) is 10x louder than at idle (62 dBa). I think the relative magnitude sounds about right.

I was wearing ear plugs that claim a 30dBa reduction in sound. It made a huge difference -- the trip was much more tolerable. On the way to Amelia I was listening to music and turned the volume up a little bit and it was perfect. I did not feel as "ear fatigued" as I have in the past. I should say that I would only do this for trips of several hours. Most of the time I roll the window down and enjoy the engine scream, but after a few hours at highway speeds I feel like my head has been in a drum.

On the way back, I was listening to some podcasts. I noticed that when the hosts were talking in studio I could hear fine. When people called in they sounded really muffled and it was hard to hear (because of crappy audio quality of telephony). I definitely felt more ear fatigue but still not as much as I would have without the ear plugs.

I like the idea of noise cancelling headphones. I have a set of in-ear that I use when flying. But I am wondering if there is a possible safety issue with that? Mine are so good, they work too well at times. I had a flight attendant sneak up on me the other day and I couldn't hear her at all. She was behind me and when she knelt down to get in my face I almost jumped out of my seat. But is that really any different than a quiet luxury car though? Need to think about that more.

Lots of good thoughts here. Need to ponder some more.

[tirwin]

This is really good stuff.

I think this brings up a really good point. What is the cost/benefit analysis to sound deadening a car?

The more I think about it -- and I am no expert at all -- the "sound field" of the car is what is important.

Here's my simple example. Let's say you have a luxury sedan. That car is going to have almost no engine and exhaust noise compared to a sports car. If the engine & exhaust sounds are the biggest contributions to cabin sound, then spending money on sound deadening is probably worthless (plus it adds weight!).

The first thought that comes to my mind is how do you figure out what factors (tire noise, road noise, engine, exhaust, etc.) are in which frequency bands?

My gut tells me that most sound deadening solutions are not going to do anything to help with a loud exhaust. Totally different problems.

Need more science-y stuff and a beer.

Originally Posted by Canada Kev

Sound intensity (dB) and frequency is a funny and interesting thing. I'm by no means an expert, but some of the things I've learned in the telecommunications world might apply here.

When you are measuring dB with a simple meter, it likely is measuring at 1000 Hz or 1 kHz. This is a standard frequency that many things use as a reference point.

Frequency will affect how you will perceive the level of loudness to be. Our ears are at their most sensitive from about the 2 kHz to 6 kHz zone. For example, that 72 dB sound intensity measured in the quietest cars, if it is measured at 1kHz, will seem as loud as a 75 dB sound if its frequency is doubled to 2 kHz.

At about 3 kHz, our ears are usually about their most sensitive and that same sound, if its intensity hasn't changed, will seem to our ears to be almost 80 dB.

So with this in mind, if the sound, regardless of what it is from, can be modified so that its frequency can be modified to be outside of that 2 - 6 kHz zone range, the car will be perceived as being quieter. In addition, if any sound attenuating material installed in the car can be tuned to work in that same frequency range, it will allow less sound of that frequency enter the car, thus feeling quieter, even if a dB meter reading levels at 1 kHz says its not much lower.

Keep in mind that the power level of decibels is doubled after an increase of 3 dB. Thus the power difference between 72 dB and 75 dB is a factor of two. And then doubled again when you reach 80 dB. Then you can throw in phons and sones to really muck things up.

Like I said, it's all very interesting and weird. It's almost enough to drive one to drink. So now it must be beer o'clock...

[tirwin]

One quick anecdote from the trip. There were 4 of us driving down. I was the lead car with my loud pipes. The other guys couldn't believe how many times people moved over out of the left lane for me. I think when left-lane hogs hear me coming, they move out of the way. Like the motorcycle guys say -- loud pipes save lives AND get you there faster!!!

[Coastr]

This is something I also have taken an interest in over the years because some cars are so much quieter and refined than others. In the quest for NVH improvements the N bit is only part of the equation - a bit like a tasty meal needing to look tasty and feel tasty as well as being tasty.

As we know from listening to an in-tune orchestra the frequencies and matching with other frequencies is important. We can all listen to our favourite music at high DB but a dog barking or baby crying at same DB is intolerable.

So as others have said it’s the frequency mix at which the volume happens that determines, as well as ‘feeling’ sound through our bodies with vibrations.

Some engineers devote their whole careers to the task so unsurprising its hard to get a good start. I’ve seen spy shots of cars rigged up to test NVH and even then I suspect there’s an expert in there taking ‘butt feel’ notes on getting the mix right. Modern intakes all have resonance chambers and things like that which are tunes to match the noise wth the revs and exhaust.

Still, I’ve always thought you could code up a NVH analyser from a phone on the passenger seat. Would be useful test driving new cars to see which is actually he most peaceful, as well as measure degradation over time.

[tirwin]

Coastr,

Thanks. Good thoughts.

This is what I am thinking. In a perfect world, you would measure the entire sound field. What frequencies are present and what are the relative intensities of each frequency?

It would then be interesting to compare the effective sound dampening range of various products and solutions.

For example, the sound frequency of a loud exhaust might be in a range that CCF, MLV, etc won’t do anything to help.

Also, I never seem to see anyone talk about what effect glass has on cabin sound. I can see where some products help with low frequency drone from road noise, but are they helpful with engine/exhaust noise? Wouldn’t higher frequencies just pass through the glass? Seems like there is a lot of marketing and little science info available.

I’m looking into a few ideas on how to measure this on the cheap. Will report back.

[Coastr]

I had a bmw where the original owner spec’d the double glazed glass. That was the single most effective thing at cutting out high frequency road noise - think tyre noise from other cars and wind. The difference was amazing. As I had to disaassemble the interior on that car several times (Ac evaporator dash out, Leaky sunroof carpet out, airbag diagnosis seats and console out) I got familiar with what goes into NVH damping. The carpets had a 30-40mm thick layer of high density foam underneath and the dash itself had a lot of sound deadening. They used grey foam to fill air gaps around the edge of the console and underneath the HVAC box. I assumed that this was to stop sound waves travelling from the holes in the car for the cabin air intakes and gear shifter hole. All basic stuff - but you could still get the straight 6 induction and exhaust note, which made it sound nice.

I think the key there is the engine air-moving noises are deisrable because they are at a ‘tuned’ frequency (signla/noise ratio low) whereas things like mechanical noise (injectors, timing chain) are more just noise. Then you get road noise from tyres which is just a wall of white noise.

So my very amateur understanding of this is the the foam blocks out the high frequency stuff and leaves the lower frequency and more tuned noise to come through. Which leaves the passenger and driver in silence except for the engine note. Going back to the orchestra example, you’re putting a filter between a screeching violin -and French horn and audience applause, but still have the tuba, bass and drums coming through.

So to start measuring you need seat time in a quiet car that still has a note (bmw do this, rolls do not) and measure the sound level of all the frequencies. That would take care of the sound measuring.

[horizontallyopp]

There are different weightings for the measurement of sound pressure level (SpL)
https://www.noisemeters.com/help/faq/frequency-weighting.asp

This can affect the reading you are getting and what the phone app is set to. Years ago, I picked up a sound pressure meter at Radio Shack as it was on sale. I have measured the level inside my various cars over the years and been surprised at how loud they are. My first was my old VW Vanagon Westfalia, which despite the fact that it never went fast, was still in the high 80s dB. In my old Porsche, like most, I wear ear plugs. My old Becker radio (even with newer speakers) is hard pressed to be heard over the wind/engine/exhaust noise.

I wear Etymotic high high fidelity earplugs, as they dampen the sound, but still let frequency ranges through so it doesn't "muffle" everything so you can still hear those odd sounds that the car loves to make.
https://www.etymotic.com/consumer/hearing-protection/er20.html

[911tracker85]

need to re-read this. how much sound deadening to put into my project has been a question. 78 930 heavily modified to handle track duty even though it will not spend much time on the track.

[tirwin]

Just getting back to this. Work had me busy for a few days.

Coastr, that is good info. I think we’re on the same page. My Mercedes is pretty quiet so I could use that as a comparison to baseline against. Just need to do some research on tools & methodology.

Keith, thanks for the pointer on the Etymotic ear plugs. I have not seen those before.

Bob, long story but I’ll try to be brief. I redid my interior maybe 5-6 years ago. I did the damplifier + CCF + MLV approach. I don’t think I did it properly in hindsight. I was trying to do enough to get some benefit but not so much to add too much weight.

Back then I had stock exhaust. The sound after the interior project was better than original but not what I had hoped to achieve. Later I went to a test pipe and 1-in/2-out muffler. The sound is awesome but it is loud. I might as well have no sound deadening — I don’t think it helps at all anymore.

I am now thinking about going to SSIs + 2-in/2-out muffler. I’m trying to figure out if I can find a decent compromise. There’s a lot of marketing around both exhaust and sound deadening. I’d like to see more hard data.

My basic premise here is to figure out the noise sources and whether spending money on sound deadening is helpful or a complete waste of time and money.