Speakerplans.com Homepage
Forum Home Forum Home > General > General Forum
  New Posts New Posts RSS Feed - LE, not MMS determines woofer "speed"!
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

LE, not MMS determines woofer "speed"!
 Post Reply Post Reply Page  12>
Author
Message Reverse Sort Order
  Topic Search Topic Search  Topic Options Topic Options
IanD View Drop Down
Registered User
Registered User


Joined: 17 January 2009
Location: London
Status: Offline
Points: 400 		Post Options Post Options   Thanks (0) Thanks(0)   Quote IanD Quote  Post ReplyReply Direct Link To This Post Topic: LE, not MMS determines woofer "speed"!
    Posted: 18 September 2016 at 11:56pm
Most of the "it's like a motor, higher force is better" similarities are just plain wrong; as far as bandwidth and speed and transient response are concerned, the T-S parameters tell you how the driver behaves and what the frequency response in the cabinet will be. When people say a driver/cabinet is "fast" what they often really mean is that it's bass-light, ones which are flat down to lower frequencies are often said to be "slow" when actually they're more accurate. A driver with a big heavy cone/coil needs a strong motor force (M-force is the extreme example of this), applying this to a lighter driver leads to low Q and bass rolloff unless the cabinet is designed to work with this -- which usually means a big FLH.

Coil inductance just acts as a bandwidth limiting factor (lowpass filter) if it's linear, and for most PA bass drivers the cutoff from this is well above the crossover frequency -- not always true for drivers aimed at home theatres which can have huge heavy high-inductance coils. But the bigger problem is when the inductance varies a lot with cone displacement, which causes distortion because the filter cutoff changes as the cone moves. So keeping the inductance low and more constant by adding demodulation rings does make for a cleaner sound, which is why the more expensive higher-quality bass drivers often do this.
Back to Top
70,s hero View Drop Down
Young Croc
Young Croc
Avatar

Joined: 14 December 2014
Location: bristol
Status: Offline
Points: 637 		Post Options Post Options   Thanks (0) Thanks(0)   Quote 70,s hero Quote  Post ReplyReply Direct Link To This Post Posted: 17 September 2016 at 7:16pm
I guess you have to start somewhere, measuiring a single cycle is the easiest measurment to make, (in lower frequencies) a Kick or a snare could have a duration of two seconds with a decay in the signal over time. The idea that this assuption (the papaer) can have any credibillity appears to be fundamentally flawed.
Top banana
Back to Top
RoadRunnersDust View Drop Down
Registered User
Registered User
Avatar

Joined: 03 December 2013
Location: United Kingdom
Status: Offline
Points: 432 		Post Options Post Options   Thanks (0) Thanks(0)   Quote RoadRunnersDust Quote  Post ReplyReply Direct Link To This Post Posted: 17 September 2016 at 4:25pm
Having re-read the article I am now somewhat confused about a number of points the author makes:

Firstly, Im not convinced that discrediting values that form a constant directly related to the experiment is such a clever idea unless attempting to force the captured data into an outcome you had previously decided on.

Secondly, its a goddamn loud speaker. I would wager that by the laws of statistics 100% of the loudspeakers on the planet are for playing music.
With that in mind, why does it seem like the author is only interested in movement from a driver at rest across a single impulse (not sure I hear much of that going on in music) whilst addressing and completely disregarding the ability for the driver to reproduce a transient whilst already in motion (like kick drum and snare for example).


Or am I missing something here?
www.guildfordcablecompany.co.uk
Back to Top
odc04r View Drop Down
Old Croc
Old Croc
Avatar

Joined: 12 July 2006
Location: Sarfampton
Status: Offline
Points: 5483 		Post Options Post Options   Thanks (0) Thanks(0)   Quote odc04r Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 12:45pm
There's just too many variables, especially when you get into a large signal model and the small signal TH specs no longer apply. BL is not linear, cones start to flex and leak air, voicecoils warm up and shift parameters etc. All we can offer is general ideas without some very expensive measurement kit and a lot of time.

The only thing that can be defined will, is how you measure transient response. Monitoring decay of an impulse test is a good start but you have to be  careful not to measure room or cabinet effects. Ability of a driver to reproduce a square wave is another potential one. FFT the results and you can see the effects on amplitude and phase shift. Generate the square and you know exactly what harmonics and amplitudes should be present when you measure it.
Back to Top
70,s hero View Drop Down
Young Croc
Young Croc
Avatar

Joined: 14 December 2014
Location: bristol
Status: Offline
Points: 637 		Post Options Post Options   Thanks (0) Thanks(0)   Quote 70,s hero Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 12:28pm
Going back to the OP, (I have not read the paper) it makes assumptions that transient response and speed are not linked which seems odd too.
The whole thrust of the argument is that an 18 is as linear as a 15 given the same power etc. As we know that the linearity is dependant on many things, one of which is cone size then the argument to me fails regardless of any other test.  In part though I think that the difference in perceived sound between an 18 and 15 is due to the physical cone size and angle of construction and how that relates to the size of the size of the sound pressure waves it is being forced to produce.?
Top banana
Back to Top
midas View Drop Down
Old Croc
Old Croc
Avatar

Joined: 27 October 2011
Location: Cumbria
Status: Offline
Points: 2109 		Post Options Post Options   Thanks (0) Thanks(0)   Quote midas Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 12:12pm
Originally posted by 70,s hero 70,s hero wrote:

As I understand it, the inductance ie the given amount of magnetic field , is dependant on various factors, if you take a simple large gauge wire and use that for a coil then the inductance will be much lower than if you use a smaller gauge with multiple windings. The windings in terms of length and gauge have to be selected upon the basis of their ability to handle the current and create a magnetic field. Inductance is directly related to force within the coils flux acting against the fixed magnet,the more windings, the more inductance, common sense would say that a lighter cone  would be more responsive given the same coil assembly otherwise any other view would appear to defy the laws of physics.

What I found with electric motors was, thinner windings with more of them yes gave ultimate higher speed. Less windings with a thicker wire gave faster acceleration/ deceleration. So response time was better. 

Obviously mag strength, coil size, cone composition, surround and voltage impact here.
In bass no one can hear you scream!
Back to Top
Steve20131 View Drop Down
Registered User
Registered User
Avatar

Joined: 28 July 2013
Location: Northants
Status: Offline
Points: 411 		Post Options Post Options   Thanks (0) Thanks(0)   Quote Steve20131 Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 11:27am
I don't know much but for 12" tapped horns down to 40hz mms has to be in excess of 80g, preferably 100g and a bl factor in excess of 18 (big magnet).  But for the power handling the voice coil must be larger and thicker glass fibre/paper for strenth so also resulting in a bigger mass.  Sims do not work with low mass units despite 25bl or 10mm xmax
Back to Top
RoadRunnersDust View Drop Down
Registered User
Registered User
Avatar

Joined: 03 December 2013
Location: United Kingdom
Status: Offline
Points: 432 		Post Options Post Options   Thanks (0) Thanks(0)   Quote RoadRunnersDust Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 11:07am
Once you start talking about controlling the cone whilst in motion you also need to specify the amplifier.
www.guildfordcablecompany.co.uk
Back to Top
kevinmcdonough View Drop Down
Old Croc
Old Croc
Avatar

Joined: 27 June 2005
Location: Glasgow
Status: Offline
Points: 3731 		Post Options Post Options   Thanks (0) Thanks(0)   Quote kevinmcdonough Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 10:40am
Originally posted by bob4 bob4 wrote:

Originally posted by kevinmcdonough kevinmcdonough wrote:

And given stronger inductance in the coil, it will also react faster to the changes.


isn't that exactly the opposite of what he is stating?




Sorry yes, what I meant lol. Was trying to type quickly
Back to Top
bob4 View Drop Down
Old Croc
Old Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1765 		Post Options Post Options   Thanks (0) Thanks(0)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 10:30am
Originally posted by kevinmcdonough kevinmcdonough wrote:

And given stronger inductance in the coil, it will also react faster to the changes.


isn't that exactly the opposite of what he is stating?

Originally posted by kevinmcdonough kevinmcdonough wrote:


Or if they have the same breaks, but one car is a steel body and one carbon and aluminium and so much lighter. Again, the lighter one will stop quicker, even if the reaction time was slightly longer at the start.


That's exactly what I was referring to with my comment about overshoot.....
Back to Top
kevinmcdonough View Drop Down
Old Croc
Old Croc
Avatar

Joined: 27 June 2005
Location: Glasgow
Status: Offline
Points: 3731 		Post Options Post Options   Thanks (0) Thanks(0)   Quote kevinmcdonough Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 10:25am
hey

yeah I don't think it takes very much more than basic high school physics to understand that both parts will effect the result and the "speed" of the driver. 

A stronger magnet will push and pull harder, and so be able to move quicker given the same cone. 

A lighter cone will be easier to move so will move faster given the same magnet. 

And given stronger inductance in the coil, it will also react faster to the changes.

So a combination of all will decide ultimately how quick it sounds.

(Though obviously it becomes far more complicated than this as there are many factors to consider when deciding on your driver design. No point in having a super light cone that can move quickly if it tears up with the pressure of the speaker design. An no point in having a huge heavy magnet if it then affects the other t/s paramiters of the driver and takes it away from it's design goals. )



Now, my problem with the article is that he suddenly discounts a load of parameters that can't just be ignored for no reason. 

I haven't had time to really read the article properly and drill down into his graphs etc at the end, but from my quick skim h<span style="line-height: 1.4;">e seems to be saying that BL and MMS will have no effect on "speed", and doesn't really qualify it so we're left to assume he means between any driver. </span>
<span style="line-height: 1.4;">
</span>
<span style="line-height: 1.4;">He starts off well describing how having stronger motor force will give stronger acceleration of the driver etc. </span><span style="line-height: 1.4;">However then goes on to say because we're dealing with a "time" that we can just discount all of these things. The time is only part of the "speed".</span>

Lets compare it to his car analogy, and assume we've raced our quarter mile and want to slow down again. Now the drivers reaction time can be thought of as his "time" thing that the article speaks of. When the driver receives the signal to slow down, there will be a little delay in his reaction time and then he starts to push the breaks. and so yes one driver will likely react quicker than the other. But this isn't the only measure of how the car will slow down, that's just a measure of when it STARTS to slow down.

Imagine the driver that was slightly slower to react had a set of breaks on his car that were 3 times as good. Yes he may take a slight few milliseconds to react, but will still come to a stop much quicker overall because of his much better breaks. 

Or if they have the same breaks, but one car is a steel body and one carbon and aluminium and so much lighter. Again, the lighter one will stop quicker, even if the reaction time was slightly longer at the start.

So with out speaker, we can't just discount BL and MMS because it's not just how quickly it starts to change, its how quickly it makes the change overall based on the signal. 

If the driver is in the process of an inward movement and gets a signal to change to outward. Yes the reaction time it takes to realise this will have an effect, of course. And this is what the author of the article seems to argue. But then how quickly it decelerates and then accelerates in the other direction to complete this movement. It may start sightly quicker, but if it's a sloopy loose motor and takes ages to then reverse and go in the other direction, it'll sound slow. And it may start a fraction of a second later, but if it has a super strong motor and then completes the change quickly, it'll still sound fast. 

k








Edited by kevinmcdonough - 16 September 2016 at 10:41am
Back to Top
bob4 View Drop Down
Old Croc
Old Croc
Avatar

Joined: 29 February 2004
Location: Finland/Germany
Status: Offline
Points: 1765 		Post Options Post Options   Thanks (0) Thanks(0)   Quote bob4 Quote  Post ReplyReply Direct Link To This Post Posted: 16 September 2016 at 10:20am
Originally posted by odc04r odc04r wrote:


The second is that when you have a series RL circuit, current no longer leads voltage by 90 degrees as it does in a pure inductor circuit. The phase shift will be spread out from 0-90 degrees as the inductor moves from being a short (0 phase) at DC to infinite resistance relative to R (90 degree phase) at high frequencies.

I strongly suspect that it is this variable phase shift (depending on the bandwidth of the driver and additional effects on phase of the cabinet design) that most causes poor time alignment and 'time smearing' of an input signal that was originally time aligned.


Thumbs Up

so does this mean that they should have rather built a second driver with identical specs except for a longer voicecoil with higher Le to make the research valid?
Back to Top
 Post Reply Post Reply Page  12>

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.05
Copyright ©2001-2022 Web Wiz Ltd.

This page was generated in 0.078 seconds.