• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

How to measure closed-back headphone SPL

JanesJr1

Senior Member
Forum Donor
Joined
Jun 9, 2021
Messages
346
Likes
255
Location
MA
I would like to measure SPL with a hand-held SPL meter, which has an SPL log and PC interface. My headphones are closed-back. It seems to me that I could get best results by placing the two cups of the headphones together around the microphone wand, so that the pickup would be close to where my ears would be, and in a closed-back configuration.

However, doing that with two transducers into a single mike would seemingly almost double the SPL that a single ear would hear. Is the right way to disconnect one side, then the other, and average the two samples? If I do that, however, the sound from one side is playing into a larger cavity than a single ear would encounter, perhaps under-estimating the SPL; or is that hairsplitting?

I thought that this kind of improvisation must be old hat for someone at ASR.... I am doing this mostly to make sure I preserve my hearing, not to do detailed research. (In other words, I don't plan to spring for Mini-DSP Ears, at least not yet.)
 
Last edited:

staticV3

Major Contributor
Joined
Aug 29, 2019
Messages
1,643
Likes
2,316
How about going off the measured sensitivity of your headphones and the signal amplitude that you're feeding them?

For example, I turned on some music with my HD600, adjusted the volume to my preference, then played a 0dBFS 60Hz sine wave without touching the volume while measuring the RMS amplitude with a DMM.

Using those two values I can calculate my own peak SPL.
With EDM I measured about 66.5mVrms, in combination with the 103.7dB/Vrms HD600 that's 103.7+20*log10(66.5/1000)=80.15dB SPL peak.

Switching to some much more dynamic Toto, I got about 135-170mVrms -> 86.3-88.3dB SPL peak.
 
Last edited:
OP
JanesJr1

JanesJr1

Senior Member
Forum Donor
Joined
Jun 9, 2021
Messages
346
Likes
255
Location
MA
How about going off the measured sensitivity of your headphones and the signal amplitude that you're feeding them?

For example, I turned on some music with my HD600, adjusted the volume to my preference, then played a 0dBFS 60Hz sine wave without touching the volume while measuring the RMS amplitude with a DMM.

Using those two values I can calculate my own peak SPL.
With EDM I measured about 66.5mVrms, in combination with the 103.7dB/Vrms HD600 that's 103.7+20*log10(66.5/1000)=80.15dB SPL peak.

Switching to some much more dynamic Toto, I got about 135-170mVrms -> 86.3-88.3dB SPL peak.
Well, good thing I posted my question to the noob forum. "Played a 0dBFS 60 Hz sine wave", and "measuring RMS amplitude with a DMM" are out of my league. I'm not going to say that I can't learn to do it, but I will say I know what a sine wave is but not how to generate one, and I know sort of what RMS amplitude might be, and I'm clueless on what a DMM is.
 

staticV3

Major Contributor
Joined
Aug 29, 2019
Messages
1,643
Likes
2,316
You can use the tone generator that's built into Room EQ Wizard or REW for short to generate all sorts of tones. Here's how to generate a 60Hz 0dBFS tone:
Screenshot 2022-04-12 224431.jpg

A DMM or Digital MultiMeter can be bought for very little money off Amazon, eBay, or the nearest home improvement store.
I'm using a Benning MM 1-3 for example.
In case you want to look into multimeters a little, here's a good resource: https://lygte-info.dk/info/DMMReviews.html
 

Aerith Gainsborough

Addicted to Fun and Learning
Joined
May 4, 2020
Messages
747
Likes
1,066
What is the purpose of the measurement?

If no super high accuracy is needed I could see the "poke the SPL meter's mic through a compact Disk hole and place it on top of one cup, sealing it gently" approach working. I did so with a UMIK and the measured value was within 1dB of the theoretically calculated one.
 
OP
JanesJr1

JanesJr1

Senior Member
Forum Donor
Joined
Jun 9, 2021
Messages
346
Likes
255
Location
MA
Thanx for the responses. I will follow up with REW, since I've been planning to become familiar with it at some point anyway.

My purpose is pretty limited, to listen to a variety of music I normally listen to, chart avg and max SPL's on my PC for those listening sessions, along with volume settings, to get a solid sense of when I'm putting my hearing at risk either for transient SPL peaks or longer-session average SPL's. (My SPL meter will support this.) I was exposed to shotguns quite a bit as a boy, and have some hearing loss, though thankfully pretty even up and down the FR spectrum up to 10k.

I more or less understand staticV3's approach in at one level, but am not clear on how that approach will help me develop good real-time practices in setting volume levels for safe listening with a variety of music. A DMM would be wasted on me because of my limited experience with the basics of electrical circuits. (I'm working on it. Finance or biology, I'm your man, but I never learned about electrical circuits beyond a single middle-school shop project in the 1960's.) Just getting a hands-on, real-world feel for SPL's that is also anchored on equipment volume settings will get me going on this. I fear volume creep. Or more accurately, I need to fear volume creep.
 
Last edited:

Aerith Gainsborough

Addicted to Fun and Learning
Joined
May 4, 2020
Messages
747
Likes
1,066
This thread might lend some insight. We've discussed a few methods to assess the general SPL we're subjecting our ears to when using headphones:


Here's what I did:
I used my speaker calibration mic (UMIK-1), poked it through a Compact Disc, sealed the headphone that way and put on a bass heavy track You cannot measure high frequencies accurately this way but bass is easy and accurate enough for the purpose, so pick some track where the bass frequencies are the loudest (that's the case in most music anyway) and use actual music as a test signal.

Then I measured the Z-weighted transient peaks with REW. That gave me the maximum possible SPL. As I stated earlier: I did the calculations in a more theoretical way and the results were pretty much the same. Both approaches are described in the afore mentioned thread.

Now the transient SPL in itself isn't all too useful in isolation but it gets you started. Because now you can use any program to determine the average loudness of a given piece (typically output in negative dB from 0dBFS) and get the average SPL you are listening to by calculating Transient dB + (- average dB). That highly varies according to music genre and even from track to track within one album. So be aware.

Last but not least, you need to consider exposition duration. It's not a problem to crank it up for a track or two but do that for hours and you're asking for trouble.

Lots of variables to consider but the effort is worth it.
 
OP
JanesJr1

JanesJr1

Senior Member
Forum Donor
Joined
Jun 9, 2021
Messages
346
Likes
255
Location
MA
This thread might lend some insight. We've discussed a few methods to assess the general SPL we're subjecting our ears to when using headphones:


Here's what I did:
I used my speaker calibration mic (UMIK-1), poked it through a Compact Disc, sealed the headphone that way and put on a bass heavy track You cannot measure high frequencies accurately this way but bass is easy and accurate enough for the purpose, so pick some track where the bass frequencies are the loudest (that's the case in most music anyway) and use actual music as a test signal.

Then I measured the Z-weighted transient peaks with REW. That gave me the maximum possible SPL. As I stated earlier: I did the calculations in a more theoretical way and the results were pretty much the same. Both approaches are described in the afore mentioned thread.

Now the transient SPL in itself isn't all too useful in isolation but it gets you started. Because now you can use any program to determine the average loudness of a given piece (typically output in negative dB from 0dBFS) and get the average SPL you are listening to by calculating Transient dB + (- average dB). That highly varies according to music genre and even from track to track within one album. So be aware.

Last but not least, you need to consider exposition duration. It's not a problem to crank it up for a track or two but do that for hours and you're asking for trouble.

Lots of variables to consider but the effort is worth it.
Really, really helpful! I did look for such a thread but guess I didn't look hard enough... Thank you!
 
Top Bottom