[Foglght]

The other day, I was looking at the specs for some of my drivers. I never did end up matching the sensitivity ratings for my drivers, and some range from 85db @ 1w to 93db.

Of course, as one adds power you get a logrithmic (db) increase in spl, but do your ears work the same way?

Does a 75W signal at 50hz and at 2khz sound the same as a 1w signal at 50 and at 2k?

Discuss?

I get the impression from a pure listening perspective that they aren't linear.

 

[aznboi3644]

edit:delet please

 

[Neil]

To summarize: you're wondering if the equal-loudness contour scales with amplitude. Is that correct?

 

[kinchu007]

current is higher for lower frequencies so that has to have something to do with it...wish I was better at EE

 

[Xander]

Human hearing linearity changes with volume.

Generally speaking...

At lower volumes we are less sensitive to lower frequencies.

At high volumes we gain low-end sensitivity in our hearing.

That is why when you play music really quiet it always sounds like it has zero bass. Turn it up to respectable levels and you can hear the bass more equal to the rest of the audio spectrum.

 

[Neil]

This is a paper you'll want to read:

http://www.nedo.go.jp/itd/grant-e/report/00pdf/is-01e.pdf

If you look at page 6 and onward, you can see how they break down our sensitivity vs frequency vs sound pressure.

 

[Foglght]

Ahh, that was what I was looking for. So, based on how loud someone wants to listen to a vehicle, one would have to somehow match the drivers so that at a desired listening range, the "loudness" would be at the desired level.....am I getting that right?

Also, from what I gather there, if you wanted equal loudness, choosing high sensitivity drivers for the low range, and lower sensitivity for upper range would balance out nicely for a "soft" system?

 

[tspence73]

Ahh, that was what I was looking for. So, based on how loud someone wants to listen to a vehicle, one would have to somehow match the drivers so that at a desired listening range, the "loudness" would be at the desired level.....am I getting that right?

Also, from what I gather there, if you wanted equal loudness, choosing high sensitivity drivers for the low range, and lower sensitivity for upper range would balance out nicely for a "soft" system?

That's an interesting thought. Are you thinking of having the bass region drivers hit power compression effects first? However, it would be far more effective if your headunit or sound processor accounted for volume increases with an equal loudness curve adjustment rather than hoping your speakers will fall in line as you increase the volume.

 

[briansz]

Hence the difference between 'A' weighted and 'C' weighted SPL measurements.

 

[Foglght]

Hence the difference between 'A' weighted and 'C' weighted SPL measurements.

Those are a static curve, and doesn't seem to incorporate a wide variety of SPL.

As you can see in the paper that Neil linked, the curves change at "loudness" increases.

 

[Electrodynamic]

Ahh, that was what I was looking for. So, based on how loud someone wants to listen to a vehicle, one would have to somehow match the drivers so that at a desired listening range, the "loudness" would be at the desired level.....am I getting that right?

Also, from what I gather there, if you wanted equal loudness, choosing high sensitivity drivers for the low range, and lower sensitivity for upper range would balance out nicely for a "soft" system?

Not necessarily. If you have an infinite amount of room for your speaker cabinets it may be an option to persue that route, but the different drivers sensitivities alone aren't going to make the curve that you see in the charts.

Before I forget, the paper that Neil linked may have stated it (I didn't read through it) but the curve is called a Fletcher-Munson curve: Fletcher?Munson curves - Wikipedia, the free encyclopedia

IMO, if you wanted that curve to change so that your systems curve would change with increasing or decreasing volume levels you might want to store 2-3 different EQ curves and set them every time you change volume levels. However, getting a system that is flat enough to begin with is hard enough.

 

[Neil]

Technically it's called an Equal-Loudness Contour. The original Fletcher-Munson curves were abandoned for the Robinson-Dadson curves, and then those were dropped when they developed iso226 (which actually found that the Fletcher-Munson curves were closer than the Robinson-Dadson curves).

Anyways, you should not be designing a playback system to have an axial response that mimics your ear's sensitivity.

 

[Xander]

Anyways, you should not be designing a playback system to have an axial response that mimics your ear's sensitivity.

Exactly. You should be designing a system that sounds good at normal listening levels.

And when you turn it up, you will hear more bass. Nothing wrong with that.

 

[Electrodynamic]

Technically it's called an Equal-Loudness Contour. The original Fletcher-Munson curves were abandoned for the Robinson-Dadson curves, and then those were dropped when they developed iso226 (which actually found that the Fletcher-Munson curves were closer than the Robinson-Dadson curves).

Anyways, you should not be designing a playback system to have an axial response that mimics your ear's sensitivity.

*edit*

It was called the Fletcher-Munson curve. Interesting that no matter what you call it that the FM curve was closer than the RD curve even after the 'official' abandonment of the FM term. Call me old school, but that's what I still refer to it as.

 

[Foglght]

Exactly. You should be designing a system that sounds good at normal listening levels.

And when you turn it up, you will hear more bass. Nothing wrong with that.

Define normal listening level.

Anyways, you should not be designing a playback system to have an axial response that mimics your ear's sensitivity.

I can't do a whole lot of designing. I find drivers that fit in the holes, and where they lay is where they stay.

All my speakers are off axis anyway......WAY off axis.

 

[Xander]

Define normal listening level.

The level you normally listen at.

That's where it should sound balanced. In my opinion.