Decibel meter.

[sunquest]

Originally bought one from Maplin for around £20. This proved inaccurate against one of their more expensive models, as did the replacement, so got my money back. Downloaded one for my IPad free and appears pretty good although I have nothing to measure it against. Has anybody any thoughts on this?
Thanks Geoff

 

[st599]

Originally bought one from Maplin for around £20. This proved inaccurate against one of their more expensive models, as did the replacement, so got my money back. Downloaded one for my IPad free and appears pretty good although I have nothing to measure it against. Has anybody any thoughts on this?
Thanks Geoff

The decibel is a ratio of A vs B.

Without the reference you're comapring too, it's a meaningless value.

 

[Bodach na mara]

Cheap sound meters are the classic illustration that you "buy cheap and buy twice". The cheapest one that I found that worked reasonable accurately cost £250 and was checked with the £700 standard calibration source and against the "cheap" class 2 meter that I had access to in my lab. I also tested £60 meters to see if they would do for student use and found that they were inaccurate and also drifted in the error of the reading. Even a reliable standard sound source to calibrate your iPad app will not come cheap. Sound is one of the hardest physical measurements to make.

 

[AngusMcDoon]

The decibel is a ratio of A vs B.

Without the reference you're comapring too, it's a meaningless value.

That's the correct technical answer, but not very helpful. As I'm sure you know, "decibel" on its own is used commonly for a measure of sound level. The ratio that the measured value of sound level in decibels is compared to is the threshold of normal human hearing. I expect this is what the OP means.

 

[sunquest]

That's the correct technical answer, but not very helpful. As I'm sure you know, "decibel" on its own is used commonly for a measure of sound level. The ratio that the measured value of sound level in decibels is compared to is the threshold of normal human hearing. I expect this is what the OP means.

Thanks for the explanation, my objective is to measure the noise levels at various engine speeds. The Maplin version showed readings in the high 80's and the IPad in the 70's which is comparable with the noise level tests Conducted by MBY on their boat tests, so I assume it is reasonably representative of the noise levels.

 

[blackbeard]

Thanks for the explanation, my objective is to measure the noise levels at various engine speeds. The Maplin version showed readings in the high 80's and the IPad in the 70's which is comparable with the noise level tests Conducted by MBY on their boat tests, so I assume it is reasonably representative of the noise levels.

The conventional reference level is 20 micropascals which is approximately the threshold of normal human hearing for mid-range frequencies. An "A" weighting is usually (not always, depending on what the measurements are for) used; this reduces the effect on the measurements of low frequencies to which the human ear is less sensitive. Where the "A" weighting is used, this should be stated, ie measurements given as dB(A) (but is often omitted). I expect you have measured an A weighted sound pressure level (ie dB(A).

I wouldn't expect a cheap sound level to be accurate but I might reasonably expect it to be somewhere near correct. A difference between two results exceeding 10 dB for the same noise source indicates something seriously wrong!

A reading in the high 80's (as dB(A)), if anywhere near accurate, is trouble. You are getting into an area where prolonged noise exposure can permanently impair hearing, although this might imply several hours of exposure on a reasonably frequent basis. It would also be unpleasant and tiring, and noise at such levels would make conversation difficult and you might have to resort to shouting. It's an unreliable and inaccurate guide, but if a normal conversation is possible then the noise is unlikely to be in the high 80's.

Are you seeking to reduce the noise, eg by sound insulation? this gets a bit more complex especially if you are trying to determine how the sound is transmitted.

I assume the boat is a mobo? I have always assumed that mobos have very effective sound insulation and nothing is audible apart from a distant purring. Maybe I'm wrong. An auxiliary engine in a sailing yacht shouldn't be unduly noisy, even a single cylinder diesel on a small yacht (and a single cylinder engine can't really be said to be engineered for low noise and vibration) shouldn't be too bad. On the other hand, a good old British Seagull can kick up a fair old racket.

here endeth the lecture

 

[JumbleDuck]

The decibel is a ratio of A vs B.

Without the reference you're comapring too, it's a meaningless value.

Not in acoustics, when the reference is defined - it's the threshold of hearing, basically.

 

[JumbleDuck]

Cheap sound meters are the classic illustration that you "buy cheap and buy twice". The cheapest one that I found that worked reasonable accurately cost £250 and was checked with the £700 standard calibration source and against the "cheap" class 2 meter that I had access to in my lab. I also tested £60 meters to see if they would do for student use and found that they were inaccurate and also drifted in the error of the reading. Even a reliable standard sound source to calibrate your iPad app will not come cheap. Sound is one of the hardest physical measurements to make.

I used a £50-ish one from Maplin for some outreach work. It tested fairly well against the ones our acoustics people use, although professional standard it was not.