Generic: How BIG for mid duty driver

[koja]

This thread originally started as my inquiry into speaker measurements but quickly evolved into another interesting discussion, thus I overhauled this first post and moved my questions elsewhere:

so, for those "who like it BIG" the NEW TOPIC is:

>Does the size and mass of the cone driver matter when it comes to mid duty?

>Does a large and heavy cone sound "lazy" and lack "attack" subjectively (for example a 12 incher if used up to say 1kHz)?

How about the difference between the same large size drivers but one meant for bass and the other exclusively declared by manufacturer as "mid duty" (like those from PD for vocal stage monitors for example)?

...and more questions are wellcome if along these lines

Bask in difference of opinions and enjoy .

 

[gedlee]

I would not use the DAC output, I'd add an amp. The DAC may want a more stable higher impedance load. The amp will buffer this.

Choose the highest level that is well within the "safe" range of the drivers. This will yield the best SNR.

Measuring near resonance on any woofer is tricky. Just be careful or the data won't be very good.

Only ever use a system that does a transfer function from voltage to SPL. "Blind" measurements of just SPL can have too many unknown errors.

Check out HolmImpulse.

I live in Novi, just across the river from you, you could come and see how I do things sometime. I am only 1/2 hour from the border (on a good day!)

 

[koja]

Earl, thanks for the pointers. I hope others chime in as well since I am not looking forward to damaging any drivers (been there, done that, paid for it )

off the topic: I was actually over at your place last year. for that matter, right now two of your B&C 12inchers are getting boxed up for sub duty . I also auditioned your system and you got me hooked on managing spk directivity; this for the open stage feeling (i.e. no room) and for not being able to localize the speakers. I think others probably experienced the same when commenting about SL's speakers and the music "not being attached" to them. after this OB project (which is my own "design"), I will try and merge those subs with LX521 top. I want to see if my expectations are right. p.s. the only reservation I had to your design was that you used a large (really bass driver) for lower mid duty to keep the dispersion angle down, whereas the real challenge is to do it with drivers of the right size for the job. on the other hand, the "beef" I have with SL/JK designs is that I have low wattage (pass) amps; thus my own attempt to design a wider OB (before surrender? ).

 

[gedlee]

p.s. the only reservation I had to your design was that you used a large (really bass driver) for lower mid duty to keep the dispersion angle down, whereas the real challenge is to do it with drivers of the right size for the job.

I can measure the polar response problems at crossover when one uses a "smaller" mid-woofer. I do not measure any problems of this sort, or any other for that matter, when I use the larger drivers. You'll excuse me if I stick to quantifiable objective evidence for my designs rather than unsupported dogma about using "too big a midrange driver".

We had this exact same conversation over in another thread and it turned out that there wasn't a single shred of evidence to support the claim. Only unsupported opinions.

My evidence is clearly shown for all my designs on my web site. As I clearly show, there is no problem using a "overly large" driver for the midrange, as long as it is the right driver. The right ones are rare, but they do exist.

 

[a.wayne]

What is an overly large midrange, could you quantify Earl ....?



..

 

[gedlee]

I would suppose that the dogma would claim that a 15" should go no higher than 300-500 Hz, a 12" no higher than 500-700 Hz and a 10" no higher that 700-900. But that's just a guess since I don't buy into the whole concept of cone weight as being a factor for anything audible. Its size matters since that determines it directivity and that is audible. How well the cone is made, its termination, etc. are the important factors in its high frequency response. Mass doesn't even play a role.

 

[Bob Richards]

If you had a 12 inch woofer that crossed over to a 1 inch dome tweeter at say 3kHZ, you could adjust it so the on axis frequency response was relatively flat, but off axis it's highly likely that the response would roll off from about 600HZ on up to the 3kHZ crossover point, where it would abruptly jump back up to being about level with the 200HZ area of the woofer, because the dome tweeter is very likely to have a much wider dispersion at 3kHZ than the 12 inch woofer.

In the case where the upper frequencies are being delivered by a horn with controlled directivity, the abruptness of the dispersion change at the crossover frequency could be made to be minimal, at the expense of high frequency off-axis dispersion. The "power response" or "room response" would have a significant rolloff at high frequencies, because the off-axis emissions would bounce around the room and eventually add to the direct signal at your ear.

 

[gedlee]

Hi Bob

I think that you have that wrong.

With a controlled directivity waveguide (CD) one can perfectly match the polar pattern of a large woofer at the crossover to that of the "tweeter". Since the "tweeter in this case has CD then its response does not "fall off" off axis, it is flat. The forward "beam", if you will, is narrow, but that is exactly what is desired to avoid early reflections.

This is clearly demonstrated on my website for the NS-15 (for example) where at 20 degrees off axis - the design intent because of toe-in - the Directivity Index (DI) is perfectly flat meaning that the direct field and the reverberant field will have exactly the same frequency response along this axis.

A subtle fall-off of the power response towards 10 kHz is intended because this is more like sound in nature where the falloff due to air absorption is ever present. a flat power response will always sound bright.

 

[Bob Richards]

Hi Bob

I think that you have that wrong.

With a controlled directivity waveguide (CD) one can perfectly match the polar pattern of a large woofer at the crossover to that of the "tweeter". Since the "tweeter in this case has CD then its response does not "fall off" off axis, it is flat. The forward "beam", if you will, is narrow, but that is exactly what is desired to avoid early reflections.

This is clearly demonstrated on my website for the NS-15 (for example) where at 20 degrees off axis - the design intent because of toe-in - the Directivity Index (DI) is perfectly flat meaning that the direct field and the reverberant field will have exactly the same frequency response along this axis.

A subtle fall-off of the power response towards 10 kHz is intended because this is more like sound in nature where the falloff due to air absorption is ever present. a flat power response will always sound bright.

I agree with what you're saying here. I'm not sure I'd want the directivity to be that narrow (matching a 15 inch woofer at roughly 1kHZ ?). The sweet spot would be better (less room reflections) but outside of the sweet spot I'd expect it to be more compromised.

Choosing directivity is a game of tradeoffs. If you want it to sound balanced and relatively full range everywhere in the room, the DBX Soundfield speaker designed by Mark Davis back in the early 1980's is the opposite extreme and sounded very good when I heard it, anywhere in the room.

It seems that the worse case is when a room has one major reflection creating comb filter effects. If the room has enough reflections, they all pretty much average out. Most typical living rooms will still have problems in the 80HZ - 300HZ area where there may only be one or two effective reflections.

If a room is substantially acoustically problematic, then controlled directivity could be a valuable asset.

 

[gedlee]

Hi Bob

Again I have to disagree.

With an total 90 degree pattern (+- 45 degrees) and a toe-in of 45 degrees, the entire room is covered by the directivity of the speaker and all seats get the same flat and smooth response because of CD. What is avoided are the walls - they are outside of the coverage angles and as such to not generate early reflections - only late lateral reflections that are the best for the feeling of spaciousness.

My rooms are always quite reflective because I want them to sound spacious and this does that. But early reflections cause degraded imaging. By using a highly directive CD speaker I can achieve both good imaging AND good spaciousness. To my knowledge there is no other simultaneous solutions to these two major problems. Usually a designer goes for one or other - I go for both.

So you would call my rooms "acoustically problematic" because they don't fit the traditional dogma of room acoustics - highly reflective walls, etc.. But yet, all being custom built, these rooms acoustics are precisely what I intended. There are no compromises made for aesthetics, etc.. It is an entirely different paradigm.

Below 300 Hz. is an entirely different problem that directivity has no effect on, so I am not talking about that situation.

Directivity is not so much a tradeoff as it is a compromise. High directivity at lower frequencies requires very large speakers. How large one is willing to use is usually the compromise.

And actually my techniques yield an incredibly wide sweet spot, not a small one. Almost the entire room has good imaging and timbre.

 

[Retsel]

Breakup modes

Hey Earl, I know that Lynn Olson tries to ensure that he uses his drivers below the breakup modes. It seems that the breakup modes of drivers depend on their size and cone material, and probably other factors which I am sure that you know better than I. Is that something that you measure and try to avoid?

I came by your place many years ago with some other local audiophiles (I live in Ann Arbor). You were a gracious host - loved your Summa speakers. It is good to know that you are back in the area.

Retsel

 

[gedlee]

I would never use any driver once it starts to "break-up". But what is breakup? How do we tell this? Its actually quite obvious when one looks at a polar map because the polar response all of a sudden becomes quite chaotic, much like the cone. I don't see this in my woofers until a few kHz, and even then it is fairly well under control. I have seen it below 1 kHz in some smaller drivers however. It is all a matter of design.

The first breakup is always the same and is well know. It is when the rim of the cone and the suspension resonate against one another. This cause a rapid peak and dip to occur. But a well damped surround can alleviate or minimize this problem.

I don't see any apparent breakup problems in the polar maps of my systems so I have to conclude that I am operating below their breakups. Breakup is not something hidden, it is completely out in the open and you can see it very clearly in measurements.

 

[Bob Richards]

Hi Bob

Again I have to disagree.

With an total 90 degree pattern (+- 45 degrees) and a toe-in of 45 degrees, the entire room is covered by the directivity of the speaker and all seats get the same flat and smooth response because of CD. What is avoided are the walls - they are outside of the coverage angles and as such to not generate early reflections - only late lateral reflections that are the best for the feeling of spaciousness.

My rooms are always quite reflective because I want them to sound spacious and this does that. But early reflections cause degraded imaging. By using a highly directive CD speaker I can achieve both good imaging AND good spaciousness. To my knowledge there is no other simultaneous solutions to these two major problems. Usually a designer goes for one or other - I go for both.

So you would call my rooms "acoustically problematic" because they don't fit the traditional dogma of room acoustics - highly reflective walls, etc.. But yet, all being custom built, these rooms acoustics are precisely what I intended. There are no compromises made for aesthetics, etc.. It is an entirely different paradigm.

Below 300 Hz. is an entirely different problem that directivity has no effect on, so I am not talking about that situation.

Directivity is not so much a tradeoff as it is a compromise. High directivity at lower frequencies requires very large speakers. How large one is willing to use is usually the compromise.

And actually my techniques yield an incredibly wide sweet spot, not a small one. Almost the entire room has good imaging and timbre.

I'm wondering if your speakers are available for listening anywhere here in Portland Oregon?

 

[gedlee]

I don't know. They are pretty rare. Kind of a unique product.

 

[Pallas]

I'm wondering if your speakers are available for listening anywhere here in Portland Oregon?

Isn't there a prolific psychopath from this forum in or near Portland?

 

[mayhem13]

I would never use any driver once it starts to "break-up". But what is breakup? How do we tell this? Its actually quite obvious when one looks at a polar map because the polar response all of a sudden becomes quite chaotic, much like the cone. I don't see this in my woofers until a few kHz, and even then it is fairly well under control. I have seen it below 1 kHz in some smaller drivers however. It is all a matter of design.

The first breakup is always the same and is well know. It is when the rim of the cone and the suspension resonate against one another. This cause a rapid peak and dip to occur. But a well damped surround can alleviate or minimize this problem.

I don't see any apparent breakup problems in the polar maps of my systems so I have to conclude that I am operating below their breakups. Breakup is not something hidden, it is completely out in the open and you can see it very clearly in measurements.

.....and it's the resonance you speak of that's easily identified in a magnified impedance sweep. Most of the time, it manifests directly in line with frequency, but lower in frequency is the harmonic distortion peak. So the woofer that breaks up at 1.5khz is usually pumping out some pretty significant odd order HD half an octave lower........or right within the pass band. This is quite typical of 12-15" drivers used to 1khz or higher. One could make the contention that it simply isn't audible......which would of course be a case of convenient subjectivity. Measurements and objectivity speak otherwise.

This certainly isn't to criticize the design choices you've made as clearly there's a huge fan base of your systems and design philosophy. Given a 15" wide baffle, I'd choose a 6.5" mid in a waveguide lens instead with a pair of 12" bass drivers below for added depth and the elimination of floor bounce brought on by stand mounted systems.

 

[gedlee]

Your welcome to do your own designs. I hope they work out. Things usually don't as well as we think.

Since no one has ever shown THD to be audible in a loudspeaker and most investigators, like Toole and Olive (and myself) completely discount this effect as audible, I will simply ignore your comments in this regard as not based in anything objective.

As to the impedance curve, one cannot always correlate SPL aberrations with impedance aberrations, sometimes yes, sometimes no. I have seen this both ways in the drivers that I use. But I challenge you to show me a woofer breakup in my polar maps. Or are you claiming that I fudge those - in which case there is no point in continuing this discussion. One either accepts the data or not. Kind of like global warming. Lots of people don't accept the data, but that is very sad.

 

[gedlee]

Isn't there a prolific psychopath from this forum in or near Portland?

John lives in San Diego now, but he actually lived in Seattle.

 

[koja]

well, engineering (of any sort) is all about trade offs. if there was no appreciable acoustic effect of the inertia force/to stiffness ratio, regardless of the damping treatment used to control a resonance, I am sure the industry would not go to the trouble of offering an array of products which span the full gamma of cone masses etc., even in the same large size. anything larger than say an 8 inch driver in the mid range (let it be lower mid to 1kHz only), is already off the beaten path, unless the shear output is the goal like in PA.

I find it interesting that, for example, JBL LSR6332 monitor (CD +-30deg), has the 12inch hand it over to the 5incher at 250Hz already, (the tweet at 2.2kHz). Earl, on the other hand, on top advocating additional sub support, takes in his speakers the B&C bass driver 12TBX100 up to 1kHz until the compression driver can take over. I would love to be able to listen to both side by side.

 

[Zvu]

well, engineering (of any sort) is all about trade offs. if there was no appreciable acoustic effect of the inertia force/to stiffness ratio, regardless of the damping treatment used to control a resonance, I am sure the industry would not go to the trouble of offering an array of products which span the full gamma of cone masses etc., even in the same large size. anything larger than say an 8 inch driver in the mid range (let it be lower mid to 1kHz only), is already off the beaten path, unless the shear output is the goal like in PA.

I find it interesting that, for example, JBL LSR6332 monitor (CD +-30deg), has the 12inch hand it over to the 5incher at 250Hz already, (the tweet at 2.2kHz). Earl, on the other hand, on top advocating additional sub support, takes in his speakers the B&C bass driver 12TBX100 up to 1kHz until the compression driver can take over. I would love to be able to listen to both side by side.

And that is exactly the case where you would be able to hear the difference that is obvious. It almost the same post as I wrote it in 1-4khz thread. The only thing i would recommend is to listen to 2 way and 3 way loudspeakers with the same woofer and tweeter but with midrange added in between. Letting PA subwoofer to play up to 1khz will IMO compromise the sound a lot and all I said is that people should try it and see for themselves. I do not have the measurements but neither does Erl that backs up his theory that it doesn't matter.

I tried 2way/3way test it and 3way with large mass just does not sound natural to me.

It is not all about polar pattern and SPL. There is more to it . I'm do not posses phd in acoustics and i can not explain what (i'm a biology student) but i can hear quite well. If mass is of no concern than every loudspeaker manufacturer could stop makin' their own 3 od 4 way speakers because it would be tremendous cost savings doing two ways instead of three ways.

And again i do think that some 15" woofers can do the 1KHz with less compromise but their Mms should not be higher than 70-75gr (Deltalite II 2515, AE TD15M etc.)