How to tune an exhaust for noise..?
Discussion
I compete in tarmac rallies in a BMW Compact fitted with a E36 M3 race engine. I run a custom made, twin box exhaust and the system has always run at 97-98db. At the last couple of events the noise level has increased and we have only just scraped by at 100db. The silencers are 'plated' rather than using traditional baffles so shouldn't get louder with use. Anyway, spectators always comment that the car is very quiet compared to rivals' on stage. It seems to be loud at rest at 4500 revs (exactly where it is measured). I know that car manufacturers tune exhausts to be quiet at certain revs to get by noise regulations, any idea how to do this? I'd be quite happy for the car to be noisier but it needs to stay under 100db when tested... I'm also keen to avoid any more restriction in the system.
It sounds like you have a really inconvenient resonance or boom right around the noise test point. Ask the scrutineers to give you a reading at, say, 3500 and/or 5500rpm just to see how that is, it would be interesting to see how sharp a peak it is at 4500rpm. If spectators are commenting that it's quieter than competitors, must be a serious effect to put it close to noise limits when tested, I'm surprised you can't hear it when driving. My guess is you have a third engine order problem, something around 225Hz.
My experience with exhausts is relatively limited, but resonances like this are caused by a vibrating mass. All mass systems have particular frequencies where they vibrate more violently - it could be the air mass flowing through the exhaust or the exhaust piping itself, but I imagine it's more likely to be the airflow that's resonating. Changing the diameter of the exhaust pipe slightly might shift the resonance away from your noise test point without having a huge effect on the "loudness" as perceived by a spectator.
The resonance might be in the back boxes you mentioned, if you have any others you can fit it'd be interesting to see what they do. If it's still bad with a different back box, presumably it's occuring somewhere further up the system. If this fixes it, you know your issue and you need to alter the flow in the back box, eg. adding or removing some baffles.
In the presumably less likely event that it's the pipework itself resonating, rather than the airflow, adding or removing mass to it would shift the resonance.
My experience with exhausts is relatively limited, but resonances like this are caused by a vibrating mass. All mass systems have particular frequencies where they vibrate more violently - it could be the air mass flowing through the exhaust or the exhaust piping itself, but I imagine it's more likely to be the airflow that's resonating. Changing the diameter of the exhaust pipe slightly might shift the resonance away from your noise test point without having a huge effect on the "loudness" as perceived by a spectator.
The resonance might be in the back boxes you mentioned, if you have any others you can fit it'd be interesting to see what they do. If it's still bad with a different back box, presumably it's occuring somewhere further up the system. If this fixes it, you know your issue and you need to alter the flow in the back box, eg. adding or removing some baffles.
In the presumably less likely event that it's the pipework itself resonating, rather than the airflow, adding or removing mass to it would shift the resonance.
Hmm, interesting thanks. It hadn't occurred to me that resonance could be the issue and that removing restrictions may improve things... Chatting to my mechanic tonight it dawned on me that this problem has only arisen since I fitted a standard engine in place of the full race one which is being rebuilt. It may be that the different cams etc on the race one overcome the issue - although I had thought being more powerful it would exacerbate the problem. In the meantime I need to find a short term fix - do you think fitting a removable 90 degree downward facing bend on the tail pipe to direct the exhaust at the ground would work?
Well, I'm assuming that it's particularly loud at the noise test point and quieter elsewhere, as your spectators' comments suggest, but we can't be sure. That's why I asked if you could get test results either side of the 4500rpm point. Perhaps you could get a very very rough subjective idea by simply listening to it yourself beside the tailpipe at 3500, 4500 and 5500rpm?
If it is a lot louder at 4500rpm than elsewhere, then you have a resonance somewhere. In this case don't worry about "fewer baffles = louder" etc, yes, it might get a bit louder across the range but if it eliminates this resonance then your noise test will drop significantly!
Putting a 90° bend in the tailpipe is unlikely to help, as I think there must be some big resonance elsewhere which will still make it out, but as I say my experience with exhausts isn't that great so it could be worth a try. Do you have alternative back boxes you could fit?
It happening only on one, less powerful engine makes it sound a lot like an inconvenient effect rather than it just being "too loud"! Are the exhaust manifolds changed?
If it is a lot louder at 4500rpm than elsewhere, then you have a resonance somewhere. In this case don't worry about "fewer baffles = louder" etc, yes, it might get a bit louder across the range but if it eliminates this resonance then your noise test will drop significantly!
Putting a 90° bend in the tailpipe is unlikely to help, as I think there must be some big resonance elsewhere which will still make it out, but as I say my experience with exhausts isn't that great so it could be worth a try. Do you have alternative back boxes you could fit?
It happening only on one, less powerful engine makes it sound a lot like an inconvenient effect rather than it just being "too loud"! Are the exhaust manifolds changed?
A digital sound meter need not cost a bomb. I doubt if a cheap one will be accurate enough to rely on near the limit for your class, but should allow comparison at various revs or with mods to the pipes.
See: http://www.ebay.co.uk/sch/i.html?_sacat=0&_nkw...
John
See: http://www.ebay.co.uk/sch/i.html?_sacat=0&_nkw...
John
I thought about smartphone meters, but I've never seen any tests to see how they compare to a proper calibrated SPL meter. My concern is that it's not meant to sample a) this sort of frequency range or b) this sort of level, so I'm not convinced it'll be able to capture and A-weight properly. Might be the opposite of help if it fails to capture the lower frequencies that are dominating.
I might try mine out alongside a scrutineer's meter when I next get a chance, though, it'd be interesting to see!
I might try mine out alongside a scrutineer's meter when I next get a chance, though, it'd be interesting to see!
Accurate to +/- 2dBA, which is quite a lot, as the decibel scale is logarithmic.
This study concluded that they may be useful, but cannot be definitive recorders of noise level.
http://blogs.cdc.gov/niosh-science-blog/2014/04/09...
John
This study concluded that they may be useful, but cannot be definitive recorders of noise level.
http://blogs.cdc.gov/niosh-science-blog/2014/04/09...
John
Gassing Station | UK Club Motorsport | Top of Page | What's New | My Stuff