Speakers distortion

[sergeauckland]

Thank you, I find your data very interesting!

And how exactly do you measure distortion? How do you compensate for distortion of the mic you're measuring with?

Btw, F206 linearity measurement that @Blumlein 88 posted looks very good to me and it doesn't seem to suffer from the rine in HF effect you're mentioning.

I measured distortion by recording some tones at different frequencies through my measuring microphone then analysing them for distortion as normal. The problem with doing that is that room gain isn't compensated for. If, for example, the room has 6dB gain at 3kHz compared with 1kHz, then clearly, third harmonic distortion measurement will be twice as high as it should be. Similarly, if the room gain is 6dB less, then the third harmonic will look better than it really is. I did it just for interest, to compare my original 801s with the active version and my previous Meridians. All were located in the same place in the room, so I expect that any room gain would be the same for all of them.

I also did some measurements using ARTA and REW, but don't have those results to hand any more.

As to microphone distortion, my understanding is that microphones have very low distortion, far lower than loudspeakers, so it's not anything to worry about.

Finally, not all modern loudspeakers suffer from the rise at HF, but many do. Modern B&W seem to be like that, but the worse are probably some of the new 'boutique' manufacturers. Zu and Boenicke comes to mind as two of the worse.

S.

S

 

[Blumlein 88]

I've seen that one, but pity they didn't show distortion below 200Hz. I guess it would be more interesting to seee how much it is in the region where ear is most sensitive.

This is how "silence" currently looks in my room:



My guess is that 50Hz and its harmonics are there from fan in the notebook and refrigirator from the kitchen, or ..?

Lower freq noise is more elevated than I thought it would be. Is that why you said that below 200Hz distortion is noise dominated?

Yes doing your own measurements you will see a better result late at night than in the daytime.

 

[Krunok]

I measured distortion by recording some tones at different frequencies through my measuring microphone then analysing them for distortion as normal. The problem with doing that is that room gain isn't compensated for. If, for example, the room has 6dB gain at 3kHz compared with 1kHz, then clearly, third harmonic distortion measurement will be twice as high as it should be. Similarly, if the room gain is 6dB less, then the third harmonic will look better than it really is. I did it just for interest, to compare my original 801s with the active version and my previous Meridians. All were located in the same place in the room, so I expect that any room gain would be the same for all of them.

I also did some measurements using ARTA and REW, but don't have those results to hand any more.

As to microphone distortion, my understanding is that microphones have very low distortion, far lower than loudspeakers, so it's not anything to worry about.

Finally, not all modern loudspeakers suffer from the rise at HF, but many do. Modern B&W seem to be like that, but the worse are probably some of the new 'boutique' manufacturers. Zu and Boenicke comes to mind as two of the worse.

S.

S

I did some distortion measurements with REW and got results very similar to what @RayDunzl got, but I'm not sure how accurate they are because of room gain.

I would also like to be sure about mic distortion.

 

[Soniclife]

Huh, I'm struggling to find words here..
Why isn't the midrange driver similar in design to high freq driver, like the ones Dynaudio used long time ago? These days all of manufacturers for midrange driver seem to use similar design to bass driver, but smaller. Why is that?

Why cone rather than dome for midrange is the question you are asking. It seems we don't know, https://www.audiosciencereview.com/forum/index.php?threads/test-atc-sm75-150-mid-dome.5628/

 

[Blumlein 88]

Ok yes, I thought you were talking about those Revel measurements you posted, which are from the NRC’s anechoic chamber. Very unlikely to be much noise affecting the measurements there!

I don't know. Anechoic Chambers that are small aren't anechoic at low frequencies. I'm not sure about the one the NRC has. So in this case maybe that low end measure is distortion. Or some of both.

 

[Krunok]

Why cone rather than dome for midrange is the question you are asking. It seems we don't know, https://www.audiosciencereview.com/forum/index.php?threads/test-atc-sm75-150-mid-dome.5628/

Yes, dome - that was the word I couldn't remember!

Oh, you discussed that already. Pity there was no conclusion as i find that dilemma quite intriguing..

 

[andreasmaaan]

I don't know. Anechoic Chambers that are small aren't anechoic at low frequencies. I'm not sure about the one the NRC has. So in this case maybe that low end measure is distortion. Or some of both.

True, typical spec is anechoic down to 100Hz IIRC. But construction and materials will typically result in very good isolation at lower frequencies too. Not sure about the NRC chamber specifically ofc. But I would expect noise levels in there to be well below the 70-ish dB measured in that graph.

 

[andreasmaaan]

Yes, dome - that was the word I couldn't remember!

Oh, you discussed that already. Pity there was no conclusion as i find that dilemma quite intriguing..

I see now

Basically there pros and cons of done mids vs cone mids. I think most manufacturers decide that the cons outweigh the pros for most purposes. It depends on the specific design though.

 

[sergeauckland]

I did some distortion measurements with REW and got results very similar to what @RayDunzl got, but I'm not sure how accurate they are because of room gain.

I would also like to be sure about mic distortion.

My Microphone Data Book gives distortion figures for many microphones, and they are all in the order of 1% for SPLs in excess of 120dB, some as high as 150dB. (The Josephson C550F measuring microphone quotes 1% at 170db SPL!!!) Assuming that distortion reduces with SPL in a linear manner, then at the 80-90dB or so level we measure loudspeaker distortion, distortion should be 30-50dB lower. Whether that assumption is warranted, I don't know, but nobody has ever mentioned microphone distortion as a significant factor in recorded audio quality, so I've chosen to ignore it in the absence of any better information.

S.

 

[Krunok]

My Microphone Data Book gives distortion figures for many microphones, and they are all in the order of 1% for SPLs in excess of 120dB, some as high as 150dB. (The Josephson C550F measuring microphone quotes 1% at 170db SPL!!!) Assuming that distortion reduces with SPL in a linear manner, then at the 80-90dB or so level we measure loudspeaker distortion, distortion should be 30-50dB lower. Whether that assumption is warranted, I don't know, but nobody has ever mentioned microphone distortion as a significant factor in recorded audio quality, so I've chosen to ignore it in the absence of any better information.

S.

This information surely looks relevant, so I find it safe to assume it's correct.

In that case it seems that in the whole recording/playback chain it is the speakers who remain dominantly responsible for the non-so-stelar SQ despite all the advance in the technology in last 20 years. Is that the conclusion?

 

[Frank Dernie]

I have a friend who designs loudspeakers for clients, who generally prefer to let people think they do it themselves.
In a nutshell the two biggest improvements in drive unit distortion have come from making diaphragms which remain pistonic in their pass band and more linear magnetic circuits.
Making the diaphragm pistonic is a material/design problem. Paper and plastic cones and domes won't be pistonic at higher frequencies so the distortion level relies almost entirely on the damping either by doping or by the inherent damping characteristic of the chosen plastic.
The problem with pistonic cones is that when they do break up the distortion is narrow band but huge, so the crossover MUST avoid letting any signal through out of band.
For bass drivers a lot of the magnet, pole piece and coil designs used to be done in a fairly crude way. Now with computer modelling the non linearities can be much reduced giving very significant reductions in bass distortion.
Or this is what he tells me
A pistonic tweeter is very expensive, and may or may not be that important, depending on one's view of the audibility of hf distortion.
A pistonic mid range driver is easily engineered today, in fact I believe they have designed a car speaker which is made for €1 which could be used as a mid range driver in a 3 way system with appropriate crossover which could have <0.1% distortion in its pass-band (I have no idea about the power handling though).
Bass drivers with an excellent linear motor are very, very expensive, and according to him, almost nobody uses one because of cost.
Fabric domes, like the ATCs I have in my Proac EBS speakers depend on the damping glodge hand applied during manufacture to control resonance, I have always wondered how they are now, spec wise, after 40 years...

 

[RayDunzl]

Just to throw in a distortion measurement:

Same room, same signal path through the preamp stage.

I measured my little JBL LSR 308 at the SPL they start singing a different song

and the Martin Logan panels still sounding pure at the same room level...

 

[andreasmaaan]

n that case it seems that in the whole recording/playback chain it is the speakers who remain dominantly responsible for the non-so-stelar SQ despite all the advance in the technology in ast 20 years. Is that the conclusion?

Absolutely. In terms of nonlinear distortion, every other part of the reproduction chain is capable of transparency. Not so for the speakers (for most listeners and most signals at most levels, I think should be added)

 

[RayDunzl]

The only measurements I've seen that smoked the ML on distortion was @dallasjustice M2 setup.

Red JBL M2, Green MartinLogan (at elevated SPL to get closer to the measurement level he took)

https://www.audiosciencereview.com/...ital-crossovers-w-subs.2369/page-3#post-66732

 

[Krunok]

Just to throw in a distortion measurement:

Same room, same signal path through the preamp stage.

I measured my little JBL LSR 308 at the SPL they start singing a different song

and the Martin Logan panels still sounding pure at the same room level...

That is quite a difference between those two!

Can you please describe in detail how did you measure that? I would like to see how my speakers would do.

What is your take on room modes with this kind of measurement?

 

[RayDunzl]

The only other thing I have to add about distortion measurement is that when using a swept tone, there may be an exaggeration of the low-frequency distortion values.

 

[RayDunzl]

Can you please describe in detail how did you measure that?

Basically, using the REW software to generate a tone and view it in its RTA window.

Measure at the listening position with a UMIK-1

 

[Krunok]

The only measurements I've seen that smoked the ML on distortion was @dallasjustice M2 setup.

Red JBL M2, Green MartinLogan (at elevated SPL to get closer to the measurement level he took)

View attachment 19538

https://www.audiosciencereview.com/...ital-crossovers-w-subs.2369/page-3#post-66732

I wouldn't really use the term "smoked" as they seem very similar to me.

Btw, is that a different method of measuring distortion than the previous one you showed?

 

[RayDunzl]

What is your take on room modes with this kind of measurement?

Distortion is at a higher frequency than the fundamental.

I have a 50Hz "node" (I guess).

The fundamental is partially cancelled (much lower in apparent level) at the listening position, the harmonic is not, so it calculates as a higher distortion level (ratio of harmonic SPL to fundamental) than is truly the case.

 

[Krunok]

Basically, using the REW software to generate a tone and view it in its RTA window.

Measure at the listening position with a UMIK-1

I can't play tones from my notebook via Volumio. Can I somehow make REW to generate a WAV file with that tone so I can put it on NAS and play it?
That is how I do when I measure frequency response.