Distortion in loudspeakers

[mitchco]

...

If you didn't notice it, it looks like the 2nd harmonic dominates the result...

---

Then I see this:

http://www.avsforum.com/forum/89-sp...-pair-jbl-4722n-speakers-67.html#post50711633

Which throws the UMIK-1 under suspicion for adding 2nd harmonic distortion itself. oh well...

Thanks Ray, yah it was my mic adding 2HD as well. Pulled the mic away from the waveguide a bit and 2HD dropped in half. Move back some more, dropped to 0.2%, but frequency response remains the same. 3HD probably valid measure...

 

[DonH56]

Just to be anal: 1/(2*pi*8*82e-6) = 242.6'ish Hz, not 500 Hz, maybe you were sending more LF energy than desired? Or is it a 4-ohm driver?

 

[mitchco]

Don, meant to say 500 Hz is the XO frequency with cap selection becoming active one octave below the XO point as you correctly calculated and recommended by JBL, attached. That was for an older 3-way active horn system and I just reused the cap, even though the digital XO point is 630 Hz as recommended by JBL for the 2- way JBL 4722's I have with the upgraded JBL 2453H-SL CD's. Cheers.

 

[DonH56]

No worries, thanks. Those are very good distortion numbers!

 

[Brad]

Here is the calibrated measured response at 0.6m of a Celestion CDX1-1747 on a PSE144 horn and the noise floor.
The SPL corresponds to 85dB at the listening position
Distortion to follow

 

[Brad]

Here is the distortion plot

 

[Brad]

In the presentation about the comparison of horns and CDs, the measurement with the LeCleach' horn and TAD2001 used a 30 Ohm resistor to flatten the response and reduce the 3rd order distortion significantly.
On DIY audio the only reference I found to this was that it was found by a TAD engineer that the driver worked better with high output impedance amps and the resistor simulates it. It was explained in an AES paper, but I couldn't find the reference.
Does anyone understand why this would work?
I would be interested if something similar could reduce the rising 3rd order distortion on the celestion driver

 

[mitchco]

Brad, have you measured the other Celestion CD to see if the result is the same? I can't comment on what LeCleach did and would love to see the AES paper. Sounds like a Zobel network, but I am not sure how that would reduce distortion... The only way I have seen it done before is using an active circuit like: https://www.klippel.de/our-products/controlled-sound.html check out the pdf's at the bottom of the page...

 

[Brad]

Thanks for the link.
I have measured 4 of these celestion drivers in total. They are all pretty similar, some worse. With 10dB lower SPL the HD orders higher than 3 fall into the noise.
In absolute terms, the distortion is not much higher than what you have measured.
I also plan to try and determine the Geddes distortion metric using a spreadsheet from diyaudio.
Some of the rising distortion past 4kHz, seems to be contributed by the horn. I am going to try a foam insert at some point.
The microphone is an earthworks M30 - specified to 140dB SPL.
I've looked for an alternative driver (1" throat with 3" mounting holes) but most seem to have similar or worse performance, but its hard to find data for the 'low' levels of ~110dB SPL

 

[mitchco]

Brad, thanks for that. Well, the M30 mic, as far as I know, is as good as it gets from a measurement mic point of view, very low distortion. And if I remember right, you have that nice Rane preamp. I did find another CDX1-1747 distortion measurement from EarlK at: http://techtalk.parts-express.com/f...on-driver-from-peerless?p=1284129#post1284129 If you scroll down a bit, you can see the Celestion on a QSC 152i waveguide (I have those waveguides as well). The distortion plot looks what is to be expected. So not sure why yours is different... Did you see Ray's post 133? http://audiosciencereview.com/forum...ortion-in-loudspeakers.1297/page-7#post-35656 Ray set me straight with the REW settings. Maybe this? If not, maybe reach out to EarlK on diyAudio or the Lansing Heritage site where I see him post from time to time. Maybe he has some thoughts as he has measured the same driver.

 

[Brad]

Thanks for the links.
I am using a Motu 1248 interface. I have measured the loopback (with a cable) response and the distortion is negligible.
I should also add that the driver has a 2.2uF cap in series for DC protection and to flatten the response. This is the main cause of the apparent rising distortion. It's really reducing level and hence distortion at the low freq end.
I have tried 2 different amps driving the CD.
I have tried all the different options for distortion plot display in REW. They don't make a huge difference in my case.
I have tried using a EM8000 mic - it gives the same distortion plot, just a higher noise floor.
I have tried the mic at difference distances, the distortion percentages stay the same
It does seem that the distortion rises quickly with level. According to that PE thread, 2HD changes 2dB per dB of SPL, 3HD 3dB per dB SPL etc
Those peerless tymphany compression drivers are an interesting option though

 

[Robbo99999]

Found this thread recently and read a good portion of it, but not every page, but I did some distortion measurements for my JBL 308p Mkii speakers using REW & UMIK and I wanted to get some conclusion on how valid the measurements are (in relation to anechoic measurements too).....also as to whether I have the various measurement options setup properly.

Basically, I'm measuring both speakers playing together at my listening position which is 2 metres distance from the speakers. Amir measures one speaker at 1 metre distance for his SPL / distortion measurements, so my understanding is that my 2 speakers at 2 metres is basically the same as measuring 1 speaker at 1 metre, in terms of mathematically the dB levels are directly comparable in terms of the speakers being under the same stress.....but correct me if I'm wrong about this.

I decided to measure my JBL 308p Mkii speakers for distortion because it was the most negative variable found when Amir reviewed this speaker, here's his review (https://www.audiosciencereview.com/...ds/jbl-308p-mkii-studio-monitor-review.17338/). So I wanted to see what kind of distortion levels I was getting at my normal listening levels as well as at my maximum listening levels....so I did measurements at 77dB and 86dB (in terms of that is what REW labelled for the fundamental level). Here are my measurements, do I have it set up right and what kind of conclusions can we draw from these? (THD is the highlighted line in each measurement) (Measurements done after applying elements of an Anechoic Listening Window EQ, as well as removing one room mode peak at 155Hz that was exhibited from the left speaker (so that EQ filter only applied to left speaker), room not symmetrical)

Normal Listening Level (77dB):

Absolute Max Listening Level (86dB):

and I also tried to display the measurement above as Percentage on the Y-axis (which is more comparable to how Amir graphs his measurements) - do I have the checkbox options setup in the most relevant manner (top right of the pic below)?:

How comparable is this to Amir's anechoic measurements?

 

[thewas]

https://www.audiosciencereview.com/...ting-thd-in-the-perspective.9085/#post-232274
and more related posts in this thread, also from the REW maker.

 

[Robbo99999]

Thanks @thewas , I think I found that page during my research which prompted me to choose the checkboxes that I show in the last screenshot. I didn't remeasure at a closer distance (to remove more room effect on the fundamental) because it was a hassle and I figured we could still draw some conclusions from my 2 metre measurement at listening position. In my eyes, with what I know so far, the distortion measurements I'm showing show that distortion is not an issue in my particular speakers at the SPL that I'm listening, but because I'm not an expert on this I wanted to get confirmation from more experienced people that can interpret these results with greater confidence/accuracy. (This is my first proper foray into measuring distortion, even though I've used REW quite a lot over the last year for other purposes).

In terms of comparing my results with for example Amir's Anechoic measurments of distortion, I'm thinking the main variable/difference is gonna be an element of room gain accross the frequency response, which means my SPL levels are inflated above that at which Amir measured? Therefore, potentially my distortion measurements could look better than his due to the speaker being able to operate at a lower output level for any given measured SPL - so speaker put under less stress for my measurements. However, if these are my Listening Levels, then I suppose the distortion measurements I show are directly applicable to my listening experience, is this correct?

 

[thewas]

Yes, the problem is that room influence increases the level, also higher measuring distance => lower SPL and more relative room influence like objects resonating reduces the signal to noise ratio and a USB mic like the UMIC is there limited too. Also even the slowest sine sweep offers there lower resolution than a stepped sine, but that is all written also in that thread.

 

[Robbo99999]

Yes, the problem is that room influence increases the level, also higher measuring distance => lower SPL and more relative room influence like objects resonating reduces the signal to noise ratio and a USB mic like the UMIC is there limited too. Also even the slowest sine sweep offers there lower resolution than a stepped sine, but that is all written also in that thread.

I would imagine the reduced signal to noise ratio variable that you mentioned would be influencing the results to show increased distortion, whereas the room influence that increases the measured SPL of the fundamental would do the opposite in terms of to show decreased distortion - although on this latter point the increased SPL of the fundamental is valid because that is the listening volume that I listen at (but not directly valid to comparing against Amir's anechoic distortion measurements)? Either way, I think I'm safe in concluding that the distortion levels I've measured at my listening levels are not a cause for concern, ie they're showing pretty low/good distortion? Perhaps even inaudible distortion at my listening levels?

I remember reading about "stepped sine" in this thread, and I concluded I didn't want to do it for fear of damaging my speakers.

 

[thewas]

I would imagine the reduced signal to noise ratio variable that you mentioned would be influencing the results to show increased distortion, whereas the room influence that increases the measured SPL of the fundamental would do the opposite in terms of to show decreased distortion - although on this latter point the increased SPL of the fundamental is valid because that is the listening volume that I listen at (but not directly valid to comparing against Amir's anechoic distortion measurements)? Either way, I think I'm safe in concluding that the distortion levels I've measured at my listening levels are not a cause for concern, ie they're showing pretty low/good distortion? Perhaps even inaudible distortion at my listening levels?

I also used to think that the distortion shown would be always increased in those case but similar on my own measurements that was not always the case, don't ask me why (possibly some masking of the mic and room?), so I think the only measurements someone can really trust are very slow sweeps or stepped sine in nearfield and with a calibrated and verified measuring chain which most of us hobbyists don't hav, or at least measuring very similar distortion curves to a professional/trusted source.
On the other hand, all these don't really matter for us hobbyists as what is audible or not does not purely depend on some fixed limits but there are several factors influencing them, as music material, room masking, experience etc.
So in few words, if it doesn't bother you, its fine, although on the other hand our references depend on our previous experiences, so a sound we are used can sound fine to us until we have heard it in a better way.

 

[Robbo99999]

I also used to think that the distortion shown would be always increased in those case but similar on my own measurements that was not always the case, don't ask me why (possibly some masking of the mic and room?), so I think the only measurements someone can really trust are very slow sweeps or stepped sine in nearfield and with a calibrated and verified measuring chain which most of us hobbyists don't hav, or at least measuring very similar distortion curves to a professional/trusted source.
On the other hand, all these don't really matter for us hobbyists as what is audible or not does not purely depend on some fixed limits but there are several factors influencing them, as music material, room masking, experience etc.
So in few words, if it doesn't bother you, its fine, although on the other hand our references depend on our previous experiences, so a sound we are used can sound fine to us until we have heard it in a better way.

Gees, you're hard to tie down, I think part of it is you wouldn't want to admit that the distortion I measured is below audible limits (or good) given our heated convo over in the Neumann thread re the 308p.....a low distortion 308p might mess with your Neumann man (melon man)! (last part of that sentence might be be a bit hard to work out my attempted funny reference!) But no, I don't hold grudges against people, I can have an argument or heated discussion with them on one day and start off with a clean slate with them the following....unless they fundamentally cross my lines but that can't happen on an internet forum. I might redo my measurements with the slowest sweep to be totally sure, but I've found the results haven't really changed much from 256K up to 1M, or is 1M enough?

 

[thewas]

Gees, you're hard to tie down, I think part of it is you wouldn't want to admit that the distortion I measured is below audible limits given our heated convo over in the Neumann thread re the 308p!

Sorry "to burst your wish" again , but even if that was true, HD are only one of the loudspeaker distortions.
More problematic and audible is IMD and there exist even harder to measure non-linearities like decay problems, port noises, enclosure vibrations which can make the difference between a good and an excellent loudspeaker.

Here for example the IMD/multitone distortion of two very good loudspeakers with inaudibly low harmonic distortions above the bass:

Source: https://en-de.neumann.com/product_files/7950/download

I own myself a pair of 2-way KH120 which at the normal listening levels have above the bass according to the graphs inaudibly low HD, but again as also other posters have written their sound quality doesn't reach the one of the 3-way KH310.

Although you don't want to believe till now , there is a reason that all 4 posters who have listened to both the JBL and Neumann in the other thread said the loudspeakers sound quality isn't comparable.

But no, I don't hold grudges against people, I can have an argument or heated discussion with them on one day and start off with a clean slate with them the following....unless they fundamentally cross my lines but that can't happen on an internet forum.

Excellent, so do I, we are here all to learn and share our experiences and knowledge.

I might redo my measurements with the slowest sweep to be totally sure, but I've found the results haven't really changed much from 256K up to 1M, or is 1M enough?

1M sounds good but also please do it in the direct nearfield like 20cm and only with one loudspeaker, so only the uncertainty of your measuring chains remains.

 

[Robbo99999]

Sorry "to burst your wish" again , but even if that was true, HD are only one of the loudspeaker distortions.
More problematic and audible is IMD and there exist even harder to measure non-linearities like decay problems, port noises, enclosure vibrations which can make the difference between a good and an excellent loudspeaker.

Here for example the IMD/multitone distortion of two very good loudspeakers with inaudibly low harmonic distortions above the bass:
View attachment 95260
Source: https://en-de.neumann.com/product_files/7950/download

I own myself a pair of 2-way KH120 which at the normal listening levels have above the bass according to the graphs inaudibly low HD, but again as also other posters have written their sound quality doesn't reach the one of the 3-way KH310.

Although you don't want to believe till now , there is a reason that all 4 posters who have listened to both the JBL and Neumann in the other thread said the loudspeakers sound quality isn't comparable.

Excellent, so do I, we are here all to learn and share our experiences and knowledge.

1M sounds good but also please do it in the direct nearfield like 20cm and only with one loudspeaker, so only the uncertainty of your measuring chains remains.

Ha, ok, you're not bursting anything, but I can see you can find ways that Neumann can be better....I can't disprove that. Not sure if I want to bother remeasuring at 1M, because I don't have a mic stand, I've been using the little tripod that comes with UMIK and putting it on the back of my sofa, slightly stacked up on a book to simulate my ear height and using cushions to stack up to allow for slightly further in front variations on my main listening position.....so I can't really measure at 1 metre unless I hold the mic with my hand, and I don't know if that's relevant? But what I will assume from my measurements that I've already shown, I'll assume that they are indeed showing good low distortion within the realms of what UMIK & REW can measure even if you don't want to admit it.