[S Clark]

My only addition to this thread is to say that Danny Richie is a gentleman and a generous man who is eager to share his knowledge with anyone interested in audio.  Some may disagree with his philosophy, dislike his products, or have some other divergence of opinion, but it is absolutely untrue to ever accuse this man of being a "snake oil salesman" or to question this man's ethics. 

   Danny may be embarrassed by my relating this, but since this thread has reached a personal level, I thought it important that the readers here get a cleared picture of his character.  I am a high school physics teacher whose students make a set of speakers for their end of year project. He has helped underwrite this project for several years, commits usually 3 and sometimes 4 days of his time, travel several hundred miles, and endure my questions by phone and email in order to help generated enthusiasm in the students of a small town high school. He has never received payment for this nor asked for any.  His involvement in time and resources  undoubtedly gives my students an opportunity to get excited about their DIY projects and definitely has helped channel many young men and women to pursue engineering.  Every year, underclassmen approach me to tell me that they have signed up for my class because of this project-- all because of the generosity of this man. 

    In short, what I am saying is that if Danny disagrees with John's measurements, it is not because of commercial reasons.  And although Danny may make a mistake, if I were John, I would go back over my measurement procedures and look for error there.  Above all, Danny is so low key that I would be shocked if a phone call from John would not find this issue settled.

[Scotty]

My bad, I got my Dans confused,sorry.
Scotty

[ttan98]

I cannot help it.......

.....Scotty beam me up..........

[dlr]

dlr, A question on an unrelated topic, which driver reproduced music with higher fidelity when listened to,the SS 12m4631 or the  RS125.
Scotty

I can't do an immediate side-by-side since I use SoundEasy in Digital Filter mode for the RS125 and it's used with the same woofer, but the difference is not small. The 12m is used LR2 with an XT19 and the RS125 is used LR4 with a SS 9300 (it can't do 2nd order due to metal cone breakup). How much of the difference is in the tweeters is hard to say.

The most obvious difference that I hear is the more natural, laid back sound of the 12m. The RS125 is more forward in the midrange. The 12m seems to let more of the recording space come through. There's a smoothness lacking in the RS combo. But like I said, that may not all be in the mid itself. There's enough difference in the rest that absolute comparison is a bit problematic. However, the 12m remains solidly on my "best" list. I also prefer 3/4" tweeters such as the XT19 and the OW1, so the RS125 can't be paired well with them due to tweeter requirements in the low end when used with that driver. I cross the 12m/xt19 LR2 @3000.

[Kevin Haskins]

I don't think there is anything wrong with Johns measurements.   You just have to realize they are just ONE set of measurements that incompletely describes transducer behavior. 

A lot of guys are making this into some sort of personal Danny is right or John is wrong sort of thing.   Most of the judgements are being based upon things other than understanding the measurements, transducers and the limits of three measurements to describe a transducer.

T/S parameters change with break-in but not in some unpredictable way.  I think we have picked over this bone before Danny.   I'd concur with the mechanical changes (but perhaps not with some of the details) but not the electrical break-in. 

For the most part it is the spider changing compliance as it mechanically works back and forth under travel.   For 6.5" drivers I've found about 80-90% of the change occurs within the first ten hours of continuous drive (25Hz sine wave @ high signal levels).   The sole effect of break-in (that is measureable) is the change of Cms (compliance/stiffness of the suspension) measured in meters/Newtwon (or mm/N).   

The change in Cms (raises as it breaks-in) effects the following:

Fs:  calculated by mms & cms
Qms: loss of compliance set by Rms
Qes: mms, BL, Re & cms
Qts:  combo of Qms & Qes
Vas: cms & Sd

Of course once you model the driver in the actual enclosure the break-in tends to be complimentary.    These are all small signal level characteristics though and any of them by themselves probably have almost zero significance (within reason) on subjective sound quality.   They help you determine the appropriate box to use for low frequency behavior but they have little significance when it comes to either the linear (FR/Phase) and non-linear components (distortion spectrum).

The best indicator of in-use distortion behavior is to look at the source.   The three major components are suspension non-linearity, BL (motor) non-linearity and inductance non-linearities.   The XBL^2 driver is at a disadvantage in John's measurements simply because he doesn't show a low level vs. high level distortion measurement.    Since BL non-linearities scale with power the BL non-linearities of the XBL^2 driver should have an advantage in high signal level measurements (which are more realistic).   

Just my 0.02

 

[Danny Richie]

T/S parameters change with break-in but not in some unpredictable way.  I think we have picked over this bone before Danny.

Yea, I am in agreement with you there Kevin. Give me a fresh out of the box woofer and I can tell you with fairly good accuracy where the parameters will be after X amount of time burning in.

I'd concur with the mechanical changes (but perhaps not with some of the details) but not the electrical break-in.

 

  One of these days Kevin, I'll have to include you in some subjective comparisons. Are you going to go to RAMF next year?

These are all small signal level characteristics though and any of them by themselves probably have almost zero significance (within reason) on subjective sound quality.

I have found driver break in to have a very notable affect on how the speaker sounds. I have often built out a fresh set of speakers and compared them to a set that has been playing for a long time. Some of the differences are due to burn in of components and not just mechanical, and they always sound different. The fresh pair always sounds "different" in the same and very predictable way too. Then after 50 to 100 hours of play time the two pair some the same.

The best indicator of in-use distortion behavior is to look at the source.   The three major components are suspension non-linearity

This brings up a very good point. Sense suspension non-linearity is a source of distortion and suspension non-linearity changes with burn in time, then one can only conclude that the distortion will change as well. Just a theory at this point but this could have a lot to do with the subjective changes noted by many as a result of driver burn in. This would be worth while to pursue.

The XBL^2 driver is at a disadvantage in John's measurements simply because he doesn't show a low level vs. high level distortion measurement.    Since BL non-linearities scale with power the BL non-linearities of the XBL^2 driver should have an advantage in high signal level measurements (which are more realistic).

Very true.

[Kevin Haskins]

The compliance changes in magnitude but not necessarily the Cms curve shape.   The cms curve via the Klippel measurements shows the non-linear characteristics based upon position (forward or rearward).  It would also show magnitude but I'd imagine the shape of the curve wouldn't change in any significant way.   The spider & the surround are breaking-in in both directions equally (I assume). 

The non-linear behavior is a result of how it is terminated to the cone, spider design (progressive vs. linear) and some other complicated factors (glues/epoxy, materials used etc..)  . 

Suspension non-linearities primarily give you even order distortion products.   They are what I'd consider to be the LEAST offensive distortion by-products. 

I may or may not go to RMAF next year.   If I do I'll come by after hours and we can do some cable swapping and some SBT.   I think I've stated my issues with subjective test in general though.   They are virtually always non-conclusive in any kind of meaningful way that could be used via research or a study.   To do them in a way that is meaningful requires group studies and lots of money.      

Don't get me wrong.   When I'm making a personal choice for what I want to use in my system it is done mostly with subjective listening test.    But for finding "truth" subjective test are extremely flawed.

That's my position and I'm sticking to it!   

[Hank]

XBL^2 designs are also harder to get built right (most build houses don't understand how critical the VC length and placement is to the performance).

Very true.  The specs are a lot tighter and manufacturers are just not used to them.  Given that and the tendency for volume mfgrs to keep costs at a minimum, it's difficult to expect them to pay super close attention to getting the specs perfect.

[laserman]

XBL^2 designs are also harder to get built right (most build houses don't understand how critical the VC length and placement is to the performance).   It wouldn't be a surprise to see a problem in the production run. 

Interesting comment, that Kevin, Danny, and Hank agree on.  For us consumers and end-users, how do we know we will end up getting a correctly spec'ed/functioning unit???

[Kevin Haskins]

XBL^2 designs are also harder to get built right (most build houses don't understand how critical the VC length and placement is to the performance).   It wouldn't be a surprise to see a problem in the production run. 

Interesting comment, that Kevin, Danny, and Hank agree on.  For us consumers and end-users, how do we know we will end up getting a correctly spec'ed/functioning unit???

The main problem is getting them to build it right the first time.   Most of the build houses can manufacture like there is no tomorrow.    The good ones have great repeatability and do good unit to unit quality.    The problem is they don't have many good design engineers.   They don't have a clue WHY they are doing something but once you get them doing it right they do a superb job.

As an end user you should depend upon the company you are dealing with.   If you buy a driver from Danny you are trusting his skills as a designer to know when a product is right.    That is probably why this whole even torqued him the wrong way.   All of this is something that shouldn't really concern the end-user.   Just deal with people you trust to get it right.

[Vapor Audio]

I hate to bring this up again when things are just settling down a bit ... but

I was measuring my completed speaks with ARTA, and though I'd see if I can reproduce that 1900hz suckout both John talks about in his summary, and Mark K shows in his data.  Sure enough, if I put the mic as they do, about 1-2" from the dustcap there's a very deep high Q suckout right at 1900hz.  I tried in on all 4 of the M-130's I have, and it's there on all of them. 

But here's the almost laughable part, if I move the mic to about 4" away from the dustcap, that suckout is gone completely.  Come on guys, I have no problem admitting both John and Mark K know way more about this stuff than me, but don't silly little mistakes in testing like that pretty much lend reason to question anything you publish?  I'm an automotive journalist among other things, and if I'm reviewing a product and something weird pops up, I don't publish my findings without first talking to the manufacturer or someone who knows the product thoroughly.  I guess it's hard to know what's 'weird' when reveiwing a speaker, but I'd say this is one perfect example. 

[Kevin Haskins]

It is a near-field artifact.   The near-field measurements are only meaningful for low frequency data.   At higher frequencies you have cancellation effects as out of phase content from the cone cancels and combines.   People who know what they are doing understand it and know how to interpret the data. 

The problem is that not everone looking at their measurements necessarily understands these things.   The distortion data taken near-field for high frequency content is all unreliable as far as I'm concerned.   It doesn't matter that all the drivers where measured in the same way.   The cones & arrival times for the near-field content is unpredicable and you don't know what is and is not an artifact of the near-field.

[Danny Richie]

Kevin is absolutely correct.

[Daygloworange]

I was talking to Danny about miking the other day. When we record stuff in the studio, man, we can spend literally hours on mic position. We then take measurements, photo's anything that'll help us recall, as mic position of even fractions of an inch are a huge deal. Mic selection is just as important. Every mic comes with a frequency response graph, and a graph showing pickup patterns. A lot of mics have similar looking frequency charts but sound like apples and oranges.

Danny concurred as it pertains to shooting a response of a speaker with a mic.

Great job bringing your experiment Turbo.  This helps shed some light on some things for sure.

You know, John Krutke took down his blog recently, and I'm not feeling so good about that for the guy. He's obviously put a lot of (his spare) time into doing his thing. I hope this somehow turns into a beneficial experience for everyone, John Krutke included.

Cheers

[DanWiggins]

NF Distance = D² / λ - a good rule to remember.  

NF Distance is the distance from the speaker to the near-field/far-field transition point.

D is the diameter of the speaker (effective diameter; for a typical 6.5" woofer this is about 12.5 cm).

λ is the wavelength in question.  Basically, in cm, it's 34300 / f where f is the frequency in question.

When you are closer than this distance, you're in the near-field of the speaker.  And the output of the speaker is "chaotic" in that it does not follow the typical 1/r² fall-off, nor is it even continuous.

For a 6.5" woofer tested at 1900 Hz, the NF Distance would be (12.5 * 12.5 / (34300 / 1900)) 8.6 cm, around 3".  Distances closer than that are in the near-field; distances beyond that are in the far-field.  Frequencies higher than 1900 Hz would have a larger NF Distance; frequencies lower then 1900 Hz would have a smaller NF Distance.

Note that this equation basically means the near-field INCREASES with increasing frequency.  If you're making frequency-amplitude-dependent measurements within this distance you are not getting an accurate measurement of the speaker's performance in a typical installation.

Also note that it implies bad things for higher harmonics when doing distortion measurements.  You can significantly skew your measurement results because the harmonics may be higher AND lower than what they would be in the far-field, where you usually sit.  For the example above, a measurement distance of 10 cm would be about 2200 Hz; if you did a THD measurement at 1200 Hz, you would be comfortably outside the near-field for the fundamental, but ALL the harmonics would be in the near-field, meaning your THD measurement is highly suspect.  And will change with just small movements of the mic.

Note, too, that when you reach a dip in the response because of the nearfield the impact of the noise floor is increased; you can see a THD contour that quite reasonably is the inverse of the frequency response at the harmonics.

Personally, I don't use near-field measurements if I can avoid them.  I use them when troubleshooting a system (trying to track down an airleak, or incomplete glue joint), or when I have to because of the nature of the system (like headphones, or computer speakers).  But otherwise, good 0.5m or further (I prefer 2 meter anechoic) is a good way to go.

For reference, I've been doing a TON of THD/IMD/SC (spectral contamination) and other non-linear effects (such as compression - power and BL) measurements the last year for one of my clients - Microsoft - in the development of some new telecom products (the RoundTable being one such unit).  Distortions - linear and nonlinear - are EXTREMELY detrimental to echo cancellers, and as such must be dropped to as low as possible (and yes, there is an XBL² speaker in that unit - it comfortably dropped THD by 60% compared to the best underhung units tested).

The first part was getting reliable THD/IMD/SC measurements.  Thankfully we had a way of checking the results, because with proper modeling and hard-core theory, we could predict the level of echo cancellation for a given THD/IMD/SC level from the speaker.  If something didn't line up, we could check the setup.  It took a few months to get things dialed in so that it all agreed with physics, theory, models and measurements.  This included construction of an anechoic chamber, tightly-calibrated ultra-low-distortion mics, and quiet buildings.

Oh, and it's a real two-way effect; your microphone also has a near-field associated with it.  In most cases it's not an issue since most mics are small diaphragms, but with the larger 1" units you still have issues with getting too close for high frequency THD measurements.

[Mark K]

There is some type of FR irregularity at around 1900. This is without a doubt certain. And the proof is there on the page on a different graph.

Since this type of windowed FFT has some resolution limitations, one may argue about the exact nature of the FR irregularity, but it is not a nearfield artifact.

You can either seek the truth or not.

[DanWiggins]

XBL^2 designs are also harder to get built right (most build houses don't understand how critical the VC length and placement is to the performance).

In reality, VC length is crucial to a good linear XBL² motor.  And in reality it is EXTREMELY easy to get right; the issue has been getting manufacturers to actually ACCEPT that when you say 4 layers with 162 turns of 0.3mm single-coat wire, you MEAN exactly that.  With most drivers, anywhere from 152 to 180 turns would be fine.

Winding voice coils, though, is quite simple to get to the exact number of turns.  You just have to convince the manufacturer to pay attention to it.  Often you'll call out a 6.2mm long voice coil, and they have 6mm and 6.5mm in stock, so will use one of those.  No, it needs a specific length.  It's not hard - in fact, getting accurate, repeatable VC winding lengths is trivially simple!  It just takes time to convince the factories that it needs to happen.

Once it does, though, watch out!

[Vapor Audio]

There is some type of FR irregularity at around 1900. This is without a doubt certain. And the proof is there on the page on a different graph.

Since this type of windowed FFT has some resolution limitations, one may argue about the exact nature of the FR irregularity, but it is not a nearfield artifact.

You can either seek the truth or not.

I'm open to any possibility, but I measured 4 drivers ... all of them showed the suckout when measured at under 2".  But on all the suckout disappeared when measured slighly farther away, like 4-5".  What else could it be?

[Danny Richie]

There is some type of FR irregularity at around 1900. This is without a doubt certain. And the proof is there on the page on a different graph.

I have seen similar effects with other drivers as well. You are too close and getting a cancellation at that wavelength.

Dan, thanks for your insight (typo corrected  ).

[Vapor Audio]

There is some type of FR irregularity at around 1900. This is without a doubt certain. And the proof is there on the page on a different graph.

I have seen similar effects with other drivers as well. You are too close and getting a cancellation at that wavelength.

Dan, thanks for your incite.

Hey Danny, might wanna hit edit real quick and spell insight correctly before the keyboard warriors on the other forums notice