AVS Forum banner
1 - 13 of 13 Posts

·
Registered
Joined
·
227 Posts
Discussion Starter · #1 ·
I understand that the speakers are by far the weakest links in the audio chain, because they distort so much. Is REW capable of measuring this distortion, or are we left with whatever data speaker manufacturers make available - zero, in my case... Are higher efficiency / larger speakers distorting less, as a general rule?


Luca
 

·
Registered
Joined
·
6,162 Posts
Measuring speaker distortion accurately is tough because whatever microphone you use will add its own distortion. That said, you can get an idea of the distortion amount by recording test tones through a microphone using audio editor software, then viewing an FFT display to see the added distortion components. I use Sony Sound Forge, but other audio editor programs also include an FFT feature. You'll play various single tones to assess THD, and pairs of tones to measure IMD. For IMD both tones should be well within the crossover range for each driver. For example, if your speaker has a crossover at 1 KHz, then you'll use 200 Hz and 250 Hz (or whatever) to measure the woofer's IM distortion, and 3 KHz and 3.5 KHz (etc) to measure the tweeter's IM distortion.


--Ethan
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by BarracudaDelGato  /t/1471702/speaker-distortion#post_23288453


I understand that the speakers are by far the weakest links in the audio chain, because they distort so much. Is REW capable of measuring this distortion, or are we left with whatever data speaker manufacturers make available - zero, in my case... Are higher efficiency / larger speakers distorting less, as a general rule?

Of course we are not left at the whims of manufacturers particularly in this day and age of high quality free analysis software and low cost measurement hardware including microphones.


Here is an example of speaker measurements that an above-average amateur should be able to duplicate for himself if he is willing to put forth the effort:

http://www.hometheaterhifi.com/surround-sound-speaker-systems/surround-sound-speaker-systems-reviews/secrets-review-of-speakers-and-subwoofer/page-5-on-the-bench.html




While the above measurement was made using megabuck dedicated hardware, it can be closely duplicated at nominal expense using the kind of measurement tools frequently discussed on AVS.


One important point is that the room that a speaker is used in profoundly affects its performance. Here is one way to address this for a reasonable cost:

 

·
Registered
Joined
·
227 Posts
Discussion Starter · #4 ·

Quote:
Originally Posted by Ethan Winer  /t/1471702/speaker-distortion/0_50#post_23288614


Measuring speaker distortion accurately is tough because whatever microphone you use will add its own distortion. That said, you can get an idea of the distortion amount by recording test tones through a microphone using audio editor software, then viewing an FFT display to see the added distortion components. I use Sony Sound Forge, but other audio editor programs also include an FFT feature. You'll play various single tones to assess THD, and pairs of tones to measure IMD. For IMD both tones should be well within the crossover range for each driver. For example, if your speaker has a crossover at 1 KHz, then you'll use 200 Hz and 250 Hz (or whatever) to measure the woofer's IM distortion, and 3 KHz and 3.5 KHz (etc) to measure the tweeter's IM distortion.


--Ethan

Thanks for replying Ethan. All I have is an EMM-6 mic with Art USB pre, and a Galaxy CM-140 SPL meter; do you think the EMM-6 would be OK to get a rough idea? On the software side I have used REW. I believe Audacity has an FFT filter, I'll have a look.


I found this thread on the DIY forum. In it REW, as well as other software like WinMLS and Steps, are discussed as a tool for measuring THD. A lot to try and digest, this may take me a while.



What distance should the mic placed from the speaker? If trying to minimize the interaction with the room, I guess that setting the mic very close to the speker, or even move the speaker outdoors, would make sense. OTOH, placing the mic at the MLP would be more related to what actually happens when I use the system, though the results may not be meaningful for a different space.
 

·
Registered
Joined
·
227 Posts
Discussion Starter · #5 ·

Quote:
Originally Posted by arnyk  /t/1471702/speaker-distortion/0_50#post_23291392


Of course we are not left at the whims of manufacturers particularly in this day and age of high quality free analysis software and low cost measurement hardware including microphones.


Here is an example of speaker measurements that an above-average amateur should be able to duplicate for himself if he is willing to put forth the effort:

http://www.hometheaterhifi.com/surround-sound-speaker-systems/surround-sound-speaker-systems-reviews/secrets-review-of-speakers-and-subwoofer/page-5-on-the-bench.html


Thanks arnyk. I notice that hometheatherhifi measures THD for subs, but not for regular speakers. Why is that? Is it fair to say that distortion mainly a LF problem with modern speakers? Why don't manufacturers release THD figures, as they do for FR or efficiency?
Quote:
While the above measurement was made using megabuck dedicated hardware, it can be closely duplicated at nominal expense using the kind of measurement tools frequently discussed on AVS.


One important point is that the room that a speaker is used in profoundly affects its performance. Here is one way to address this for a reasonable cost:


Poor man's anechoic chamber eh?
Wouldn't measuring my speakers in my room be closer to what I actually hear when I listen to music? Do measurements in-room hold any water, at least at mid / high frequencies?
 

·
Registered
Joined
·
4,587 Posts

Quote:
Originally Posted by BarracudaDelGato  /t/1471702/speaker-distortion#post_23292082


Thanks arnyk. I notice that hometheatherhifi measures THD for subs, but not for regular speakers. Why is that? Is it fair to say that distortion mainly a LF problem with modern speakers? Why don't manufacturers release THD figures, as they do for FR or efficiency?

Poor man's anechoic chamber eh?
Wouldn't measuring my speakers in my room be closer to what I actually hear when I listen to music? Do measurements in-room hold any water, at least at mid / high frequencies?

Actually the measurement Arny posted is for a center speaker, so not a sub. THe measure THD for other speakers they review, too.



I suspect speaker manufacturers don't specify THD because, first, the numbers will look awful compared to electronics and second, a THD number is likely to be fairly meaningless in the context of speaker response. Haven't studied it, but since mechanical forces are at play, I'd guess that speakers make distortion differently from amps, and whether a particular distortion level is problematic might vary from speaker to speaker.
 

·
Registered
Joined
·
227 Posts
Discussion Starter · #7 ·

Quote:
Originally Posted by JHAz  /t/1471702/speaker-distortion/0_50#post_23292150


Actually the measurement Arny posted is for a center speaker, so not a sub. THe measure THD for other speakers they review, too.

You are right, I just connected HSU with subs, my bad.
I can see that, whenever speakers were tested, distortion was there.

Quote:
I suspect speaker manufacturers don't specify THD because, first, the numbers will look awful compared to electronics and second, a THD number is likely to be fairly meaningless in the context of speaker response. Haven't studied it, but since mechanical forces are at play, I'd guess that speakers make distortion differently from amps, and whether a particular distortion level is problematic might vary from speaker to speaker.

This makes sense, thanks. How very convenient for speaker manufacturers though. Is there even a standard way to measure THD in speakers?


I always associated that sort of 'fatigue' that sometimes one experiences when the system is playing louder with distortion. I am ashamed to admit that much money was wasted replacing amps and pre-amps in my early years in the hobby.


Thanks to AVS I know a little better these days, and my focus has switched to speakers and room. Yet I wish there was a "number", or at least a group of standardized measurements that could tell me if I'm likely to have a problem.
 

·
Banned
Joined
·
14,420 Posts

Quote:
Originally Posted by BarracudaDelGato  /t/1471702/speaker-distortion#post_23292082



Wouldn't measuring my speakers in my room be closer to what I actually hear when I listen to music? Do measurements in-room hold any water, at least at mid / high frequencies?

You are starting to ask questions whose answers are in this book:


Sound Reproduction: Loudspeakers And Rooms By Floyd Toole


You can get the book and read it and ask questions here about whatever you don't understand right away, or you can try to pry many of the same answers out of online forums one at a time and maybe not always get the best answers. ;-)
 

·
Registered
Joined
·
7,989 Posts
If you're trying to understand the speaker, you isolate it from the room. If you're trying to understand the room, you must first understand the performance of the speaker, and then measure it in the room.


One of my favourite pics from when I tested a speaker a couple years ago. I usually don't go that high unless I need information down to 80hz:

http://s874.photobucket.com/user/tuxedocivic/media/Challenger/0e93c9db.jpg.html


We hear both the direct sound of a speaker, and the room sound (depedning on frequency). The book Arny recommended is a good one.


Measuring distortion is no easy task. Be careful how you interpret the results to first, make sure they're right. And second, make sure they matter. If both of those are true, then third, why is there distortion? Bad speaker, poor design, high spl...?
 

·
Registered
Joined
·
227 Posts
Discussion Starter · #10 ·

Quote:
Originally Posted by arnyk  /t/1471702/speaker-distortion/0_50#post_23292976


You are starting to ask questions whose answers are in this book:


Sound Reproduction: Loudspeakers And Rooms By Floyd Toole


You can get the book and read it and ask questions here about whatever you don't understand right away, or you can try to pry many of the same answers out of online forums one at a time and maybe not always get the best answers. ;-)

I'll do that. Thanks arnyk.
 

·
Registered
Joined
·
227 Posts
Discussion Starter · #11 ·

Quote:
Originally Posted by tuxedocivic  /t/1471702/speaker-distortion/0_50#post_23293225


If you're trying to understand the speaker, you isolate it from the room. If you're trying to understand the room, you must first understand the performance of the speaker, and then measure it in the room.

thanks tuxedocivic. Nicely put.
Quote:
One of my favourite pics from when I tested a speaker a couple years ago. I usually don't go that high unless I need information down to 80hz:

http://s874.photobucket.com/user/tuxedocivic/media/Challenger/0e93c9db.jpg.html

Great picture. I am not sure I'll go that far, WAF, neighbours AF, and all that.
I'm guessing that what you're measuring was a DIY project? Is there a thread here on AVS?
Quote:
We hear both the direct sound of a speaker, and the room sound (depedning on frequency). The book Arny recommended is a good one.


Measuring distortion is no easy task. Be careful how you interpret the results to first, make sure they're right. And second, make sure they matter. If both of those are true, then third, why is there distortion? Bad speaker, poor design, high spl...?

Yeah, I'm buying that book, and will go from there.


Luca
 

·
Registered
Joined
·
6,162 Posts

Quote:
Originally Posted by BarracudaDelGato  /t/1471702/speaker-distortion#post_23292018


do you think the EMM-6 would be OK to get a rough idea?
I have no idea how much distortion that mic adds. One way to estimate THD added by a microphone is to measure the IMD which is easier to do. This is from my Audio Expert book:
Quote:
Originally Posted by The Audio Expert 

The best way to measure microphone IMD is with two speakers, with one tone played through each speaker. This avoids contaminating the resulting sum and difference frequencies with the loudspeaker’s own IM distortion, which is typically higher than most microphones at low frequencies. You’ll create separate files for each test frequency, panning them hard left and right for playback. The microphone should be halfway between each speaker and also on-axis to capture the flattest response from both speakers. Assuming your monitoring is set up correctly, with each speaker angled toward your ears, just put the microphone where you listen at the same height as the speakers. Figure 22.3 shows the IM distortion I measured for my audio-technica 4033 and DPA 4090 microphones using this method, by playing 300 Hz and 500 Hz through the JBL 4430 speakers in my home studio. As you can see, the AT 4033 has half the IM distortion of the DPA 4090. Adding the sum and difference signal levels, the AT is about 0.2 percent IMD versus about 0.4 percent for the DPA.

Even though THD and IMD are different, their amounts are often related fairly closely. So if you duplicate this type of test with your microphone, you'll have a good ideas of how the mic's distortion compares to what you measure from the speakers.
Quote:
If trying to minimize the interaction with the room, I guess that setting the mic very close to the speker, or even move the speaker outdoors, would make sense.

Exactly.


--Ethan
 

·
Registered
Joined
·
7,989 Posts

Quote:
Originally Posted by BarracudaDelGato  /t/1471702/speaker-distortion#post_23295996


I'm guessing that what you're measuring was a DIY project? Is there a thread here on AVS?

Ya, but it never totally panned out. I got it sounding pretty good, about 90% of the way there, but never totally satisfied. SEOS finally became a reality so I went in that direction and abondoned that project. I just like the picture cause it's the highest I've ever measured. I usually just measure 6ft up or ground plane.


The EMM-6 is supposed to be pretty clean. The UMM-6 has more noise according to Cross Spectrum. I use an ECM8000 and it seems to be pretty clean, but I don't get to worked up about distortion. Usually the ambient noise adds more "distortion" than the mic does. Watch your furnace, other family members, fridge, various appliances, traffic outside, etc. Keep it dead quite. And you always need to measure against some reference or method. For instance: 30cm from face of baffle at 95db. Or something. You can measure such that the distortion is buried in the noise, but what good does that do if at 100db it's got loads of distortion. It's a tricky thing.
 
1 - 13 of 13 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top