Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Art Welter]

In a recent thread Ivan indicated that front loaded subs have more distortion compared to tapped horn subs. Is the distortion a function of the low frequencies produced by the sub or is distortion inherent to all front loaded speakers?

Robert,

The harmonic distortion generated by subwoofers is inherent in any type of driver enclosure. Drivers can be designed to have less distortion, but at considerable expense.

Given the same drive level, “horns are cleaner than BR (bass reflex)” is a myth.
The horn’s "acoustic transformer" between the stiff suspension of the driver and the "loose air" requires the cone to do more “work” to achieve the higher output level, and that “work” causes more cone stress and deformation than in a BR.
A light cone that works well in a BR may actually permanently deform when subjected to the acoustic loading of a horn when driven to Xmax, and may distort badly before that.

The charts below are distortion results for the BC18SW115-4, (a very heavy duty cone) in a bass reflex cabinet and the Keystone Sub, a tapped horn with similar frequency response to the DSL TH-118, (which uses the same driver) using sine wave tests. Both the BR and the TH have a similar Fb (box tuning, minimum excursion) of 38 Hz.

The BC18SW115-4 BR 38 Fb averaged 120.6 dB, 2.69% distortion.
The BC18SW115-4 TH averaged 126.52 dB, 5.438% distortion, about 6 dB more output, with about twice the distortion, though excursion was lessened through much of the range compared to the BR. Granted, distortion would be less if the output level was dropped to that of the BR, but the bottom line is when powered at “reasonable” levels, the TH does have more distortion than the BR.

Excursion was measured at the same frequencies at a lower drive level to save my ears and avoid power compression while measuring.
The TH, although having considerably less excursion from 60Hz to 90Hz, has considerably more distortion than the BR, demonstrating distortion is not just excursion related.
That said, most of us would be hard pressed to identify less than 10% THD (total harmonic distortion) from frequencies below 100Hz with most musical signals, which are rich in harmonics.

Art

[Frank Koenig]

Robert,

The harmonic distortion generated by subwoofers is inherent in any type of driver enclosure. Drivers can be designed to have less distortion, but at considerable expense.

Given the same drive level, “horns are cleaner than BR (bass reflex)” is a myth.
The horn’s "acoustic transformer" between the stiff suspension of the driver and the "loose air" requires the cone to do more “work” to achieve the higher output level, and that “work” causes more cone stress and deformation than in a BR.
A light cone that works well in a BR may actually permanently deform when subjected to the acoustic loading of a horn when driven to Xmax, and may distort badly before that.

The charts below are distortion results for the BC18SW115-4, (a very heavy duty cone) in a bass reflex cabinet and the Keystone Sub, a tapped horn with similar frequency response to the DSL TH-118, (which uses the same driver) using sine wave tests. Both the BR and the TH have a similar Fb (box tuning, minimum excursion) of 38 Hz.

The BC18SW115-4 BR 38 Fb averaged 120.6 dB, 2.69% distortion.
The BC18SW115-4 TH averaged 126.52 dB, 5.438% distortion, about 6 dB more output, with about twice the distortion, though excursion was lessened through much of the range compared to the BR. Granted, distortion would be less if the output level was dropped to that of the BR, but the bottom line is when powered at “reasonable” levels, the TH does have more distortion than the BR.

Excursion was measured at the same frequencies at a lower drive level to save my ears and avoid power compression while measuring.
The TH, although having considerably less excursion from 60Hz to 90Hz, has considerably more distortion than the BR, demonstrating distortion is not just excursion related.
That said, most of us would be hard pressed to identify less than 10% THD (total harmonic distortion) from frequencies below 100Hz with most musical signals, which are rich in harmonics.

Art

Thank you, Art, for some actual data. I think you're the only person here who does serious distortion measurements and is willing to talk about them.

I'm curious how you are measuring excursion?  It occurred to me after an earlier discussion here that one way might be to drive the speaker with a sine wave long enough to come into the steady state and then open-circuit the speaker input while measuring the voltage. Once open-circuited, the voltage should be proportional to the velocity and, for a sine wave, the excursion should be proportional to the velocity, too.

v = B*L*u

x = Int{ u(t) dt }

Where v =  open-circuit voltage, B = flux density in the gap, L = length of the voice coil winding under the influence of B, u = voice coil velocity, x = excursion, and Int{ * dt} represents the time integral.

My assumption is that since the electrical damping is gone the Q of the mechanical/acoustic system should be high enough to give at least a few cycles to measure. I could do the math based on typical values of Qms but haven't. Re (power compression) won't enter into it as the measurement is open-circuit. I think the main source of error would be variation in B*L with excursion, and this may indeed be fatal once we're flirting with the non-linear range.

I should just try it at some point. It would not be hard to set up using a relay and a storage scope.

--Frank

[Art Welter]

Thank you, Art, for some actual data. I think you're the only person here who does serious distortion measurements and is willing to talk about them.

I'm curious how you are measuring excursion? 

Frank,

Josh Ricci has done far more serious distortion measurements on a much wider variety of cabinets than I have, but seems the Data Bass site http://www.data-bass.com account has been suspended..

I simply measured excursion using a dot near the edge of the cone, persistence of vision and a ruler make it fairly easy to see the end points for excursion over a millimeter or so, though at very high SPL, "eyeball wobble" makes it take longer to determine the actual excursion.

Art

[Mark Wilkinson]

Frank,

Josh Ricci has done far more serious distortion measurements on a much wider variety of cabinets than I have, but seems the Data Bass site http://www.data-bass.com account has been suspended..

I simply measured excursion using a dot near the edge of the cone, persistence of vision and a ruler make it fairly easy to see the end points for excursion over a millimeter or so, though at very high SPL, "eyeball wobble" makes it take longer to determine the actual excursion.

Art

Hi Art, he's migrated to a new site.      https://data-bass.com/#/?_k=3v6efr

And thanks for the distortion tests....always good to see assertions held up to measurements....

[Art Welter]

Hi Art, he's migrated to a new site.      https://data-bass.com/#/?_k=3v6efr

And thanks for the distortion tests....always good to see assertions held up to measurements....

Mark,
Thanks for the thanks and the data-bass update!

Josh's measurements are invaluable for myth-busting. I've attached an example of his 18 cubic foot Othorn TH compared to a 15 cubic foot BR at similar voltages.
The TH has more than 5dB more output for a given voltage, but distortion levels are for the most part a bit higher when driven at similar voltage levels.

Art

[John Roberts {JR}]

Perhaps ironic unsophisticated listeners often perceive the extra very LF distortion as natural bass content.

I have experienced similar disconnects with listeners who preferred the sound of power amps allowed to clip vs. unclipped thanks to fast limiting.

FWIW the clipped amplifiers actually put out more power while clipping, and louder typically sounds better. Simple harmonic distortion on top of LF tones, can sound musical, and listeners accustomed to the extra LF distortion, can have difficulty evaluating clean systems, assuming they have less bass response.

The customer is always right even when wrong... not my problem any more...

JR   

[Ivan Beaver]

Mark,
Thanks for the thanks and the data-bass update!

Josh's measurements are invaluable for myth-busting. I've attached an example of his 18 cubic foot Othorn TH compared to a 15 cubic foot BR at similar voltages.
The TH has more than 5dB more output for a given voltage, but distortion levels are for the most part a bit higher when driven at similar voltage levels.

Art

It would be interesting to see the distortion measurements at the same SPL.

After all, that is what it is about in the end.  Not the drive voltage.

Does speaker A have less or more distortion at X Hz than speaker B while producing some specified SPL?

After all, 5dB is almost 4 times the power.

I don't know the answer, because I haven't done it.  But curious.

[Tim McCulloch]

It would be interesting to see the distortion measurements at the same SPL.

After all, that is what it is about in the end.  Not the drive voltage.

Does speaker A have less or more distortion at X Hz than speaker B while producing some specified SPL?

After all, 5dB is almost 4 times the power.

I don't know the answer, because I haven't done it.  But curious.

I think drive voltage is used because we can mostly agree what 38VAC RMS is, for example, compared to "how was that SPL measurement taken?"  Then again, it could be something else. 

[Ivan Beaver]

I think drive voltage is used because we can mostly agree what 38VAC RMS is, for example, compared to "how was that SPL measurement taken?"  Then again, it could be something else. 

I totally agree, it all "depends" on how you look at things.

But if one speaker cannot produce as much SPL with the same drive level, you will need to increase that drive level (and distortion as well) in order to get it "loud enough".

Or else use more subs, and amps, with a larger truck to get it there, and more trips to and from the truck and so forth.

That is why it is important to look at more than the "simple answer" in order to get the whole story.

And even then, the "true story, can have different meanings to different people.

[Mark Wilkinson]

Mark,
Thanks for the thanks and the data-bass update!

Josh's measurements are invaluable for myth-busting. I've attached an example of his 18 cubic foot Othorn TH compared to a 15 cubic foot BR at similar voltages.
The TH has more than 5dB more output for a given voltage, but distortion levels are for the most part a bit higher when driven at similar voltage levels.

Art

You bet Art !

I was gonna ask you the same thing Ivan has, if you have any equal SPL data...
cause Ivan, that makes the most sense to me too.

Once I have set my SPL and low end extension targets, then I want to know how many boxes of what types ….does it take to get there cleanly.


The only time I've done any distortion comparisons among box types was in this thread    https://forums.prosoundweb.com/index.php/topic,167351.0.html
But it was a double 18 bass-reflex and a Labhorn…..not sure if any distortion conclusions can be drawn from it at all.

The best comparison of the same driver with different box types I've seen is Ricci's article on sealed, vs ported, vs tapped, that uses a 12".
https://data-bass.com/#/articles/5cb774640ca6e70004e10828?_k=7muugk

[ProSoundWeb Community]