[hubert]

Hi Dave and all,

A discussion began in the 3-way topic, I wish to continue it in a dedicated one if you agree.
Dave, what is your opinion about Cms and what related listening caracteristic do you guess/feel about?

Thanks.

[David Ellis]

I... am not really sure what you are looking for with regard to my opinion on the matter.  Many of the T/S parameters are interrelated.  Cms is certainly one of them - rooted primarily in the spider.  As such it's either tight/loose or somewhere in-between. 

I generally consider things in terms of the Q - numbers.  A tighter suspension/lower Q is should sound cleaner BUT will have less bass reach.  Such a suspesion would normally have a LESS compliant spider. 

Is this what you were looking for?

Dave

[hubert]

I generally consider things in terms of the Q - numbers.  A tighter suspension/lower Q is should sound cleaner BUT will have less bass reach.  Such a suspension would normally have a LESS compliant spider. 

Is this what you were looking for?

I'm interested by the T/S parameters values and their related listening characteristics; indeed Q-numbers give probably enough informations...
For Cms, somebody consider a high compliance of the mass/spring effect allows to get listening details at low level, I just wanted to know if you were looking at this particular parameter when choosing a driver.

[David Ellis]

I have not considered Cms as an isolated variable.  This is because it can't be isolated.  It exists in a system of parameters T/S parameters and can't be separated from Qms. 

I am VERY certain that driver builders consider this parameter intimately when selecting a spider for the implemented cone weight and desired Qms outcome. 

I only asked that Qms be kept under 4.  I don't have any hard-science support for my desire.  I only have a belief that the higher impedance peak created by higher Qms woofers creates a difficult phase angle for amplifiers. 

I must offer a caveat herein.  I have read 1 article conveying that high Qms woofers are actually superior to low Qms woofers.  I believe this article was published by a respected Canadian loudspeaker manufacturer.  I don't recall his detailed explanation, but vividly recall his assertion.  Certainly the SCC300 (now gone) was a high Qms woofer and sounded very good.  The SCC300 has many other positive attributes that may overcome the high Qms.

A possible explanation for the subjective preference of a high Qms woofer is that creating the high phase angle creates a DESIRABLE distortion in the amplifier.  The bass sounds slightly full, warm and pleasant.  However, this is only a guess. 

Opinions anyone?

Dave

[hubert]

A possible explanation for the subjective preference of a high Qms woofer is that creating the high phase angle creates a DESIRABLE distortion in the amplifier.

Regarding that the impedance peak can be compensated by a RCL network, you could be sure about your guess by comparing one woofer with, and another without this network.

Something is for me astonishing in this matter of mecanical values: for exemple CMS and even QMS are single (fixed) values...and in my point of view, such mass/spring systems have NOT constant stiffness/elasticity, at least for well-builded spyders...or these numbers are correct for some frequencies/power levels and not at some limits of conditions (i.e. fs)? 
Opinions?

[David Ellis]

Regarding that the impedance peak can be compensated by a RCL network, you could be sure about your guess by comparing one woofer with, and another without this network

Well, not really.  However, is a complex issue on the fringe of my knowledge.  I will convey that I know, and also admit what I don't know. 

I have attempted using electrical compensation for the impedance hump in the midrange of the loudspeaker.  Dennis has also done this.  Another very respected DIY amp guy has also accomplished this.  The prevailing subjective sentiment was marginally a dull sound or no impact.  I believe the best words used to describe this phenomena were, "the impedance compensation stole the life from the speaker".  I really don't know why this happenend, but it did.

So, if an a/b test were performed, it wouldn't be completely valid. 

Further, flattening the impedance of a woofer in the bass region is difficult or impossible in the real-world.  This is because as wattage increases the normal woofer impedance hump in-cabinet will shift frequency.  I am fairly certain that in a sealed system the shift is to the left/lower.  Hence, an impedance compensation network designed for a specific/fixed frequency would only be marginally effective. 

Something is for me astonishing in this matter of mechanical values: for example CMS and even QMS are single (fixed) values...and in my point of view, such mass/spring systems have NOT constant stiffness/elasticity, at least for well-built spiders...or these numbers are correct for some frequencies/power levels and not at some limits of conditions (i.e. fs)?

Thick issues!!  My answers will be short, and will only address the surface of these issues. 

 - Spiders are progressive to a designed extent. 
 - Testing voltages will impact measurements, but if the voltages are consistent between drivers there is a valid means for comparison and implementation based on experience.
 - I am told there are better and worse spiders.
 - I am also told that spiders are extremely difficult to design and build.  What appears very simple is actually very complex.

Overall, your questions are very good.  I might suggest that you visit with a real loudspeaker manufacturer.  I learned more about drivers in 1 hour with a SEAS engineer, and 1 hour with a TC sounds engineer than from anywhere else in the past 7 years.  The guys who have design and build drivers for reputable companies really know their stuff.

I recommend reading this article:

http://www.speakerbuilding.com/content/1039/

I read the last paragraph on the first page several times, but it hasn't soaked-in yet.  Some of this information seems backwards.

I also recommend you read this article.  There is a section titled "Variance of Loudspeaker Parameters" that is quite clear and should be helpful.

http://www.klippel.de/pubs/aes2000/klippel%20paper%20109.htm

Dave

[hubert]

Yes Dave,

I agree: visiting a well-known driver manufacturer and speaking with the engineers certainly would be very know-full; I think about Germans ones who aren't far away from Alsace
Thanks for the links, particularly the last one; what I keep in mind: as allways, mecanical issues are subject of compromises; however I feel that for high-fidelity drivers, I would prefer strong and well builded motors allowing to electrically control the moving parts and no stiff suspensions made for high power handling. Indeed, my guess is that letting the parts moving freely (all things being equal) certainly is one of the keys to high resolution...but only a guess at this moment.