Who has overtaken lambdacoustics - some user tell: "The best bass I have ever heard"

Status
Not open for further replies.
Jump to Latest
#81
Can't resist...

Well, I'm no expert and probably wrong here but this is what I think (guess):

In home application, and woofer(s) up to 15", I think neither Le(X) nor thermal compression would be real problems. The actual excusion and power are just too small. (I assume most of us discussing here are mainly focusing on home application.)

Maybe, in the normal listening level, the deformation of(among) VC/cone/suspension... etc. might have an effect much larger than the above.
 
#82
the only way to be sure...

Hi CLS,

A great line from Aliens 2 is " I say Nuek em from space, its the only way to be sure...!"
True to my Overkill Audio philosophy I do tend to go over the top...

Your quite right, using a the PD 158 in my domestic speakers is Overkill and I doubt I will ever approach thermal compression or variable BL problems when tickling the drivers with a few watts, and that is still very loud!
But its fun when friends come round and I play the sound track to Gladiator or Avatar or something with huge dynamics, I can run full power on all amps and you should see the looks on peoples faces....!!
They love it, even the i-pod designer hi-fi ( the " hi-fi should be seen and not heard" brigade....!) folks admit its another world compared to their £5K worth of polished alloy and glass.

How are your various horn projects comming along?

Cheers


Derek.
 
#83
CLS said:
Can't resist...

Well, I'm no expert and probably wrong here but this is what I think (guess):

In home application, and woofer(s) up to 15", I think neither Le(X) nor thermal compression would be real problems. The actual excusion and power are just too small. (I assume most of us discussing here are mainly focusing on home application.)

Maybe, in the normal listening level, the deformation of(among) VC/cone/suspension... etc. might have an effect much larger than the above.
Click to expand...

Normal woofers have problems all the time in some home environments. I have a 1500 sq ft room and I have a custom HT room 15x25ft room. Seas, scanspeak, etc do not cut it for me. Pro audio or similar are the only choices to produce dynamic peaks without a signal issue.

There is also the amp requirements. Audiophile drivers need 250 to 500 Watts to get to the peaks involved with music or movies.

In the end I believe headroom is key to the best overall performance of any speaker design. Without headroom the speakers and/or the amps are going to be puking on themselves during the most demanding moments.

The most demanding moments define any system.

I wont even get into the mismatch of having 4 18" woofers for bass then thinking a 6" woofer in a main speaker will blend well 😱
 
#85
Doug20 and Overkill,

Re-read Johns post(s) and Dan Wiggins paper CAREFULLY.

Dan's paper is talking about JERK. The rate of change of acceleration.
he calls it "transient response".

I have read it carefully and I can't find where it says that adding mass
has no effect on acceleration.

Johns post states "added mass has NO effect in the time domain as seen
in the impulse response." Again, no mention of mass having no effect on
acceleration.

Here's what John DOES say: "The change in acceleration of the driver is
directly proportional to the current applied"

Again, change of acceleration. The derivative of acceleration. Wiggins
even went so far as to try to explain how the constants drop out when
taking the derivative of the acceleration equation. (I believe he called
them proportional).

This is college level physics - not grade school. If you are looking for
a way to understand what they are saying, go to Wikipedia and read up
on it. If you havent done the excersize of taking the derivative of
position to get velocity (dX/dt) or the derivative of Velocity to get
acceleration (dV/dt), then taking the derivative of acceleration to get
jerk (dA/dt) will not make sense.

herm

P.S. It's hard to write this without sounding like I'm busting your balls. I'm
NOT! i'm trying to help you and everyone else who reads this thread take
advantage of the knowledge being freely distributed here!

P.S.S. Feel free to PM me & we'll discuss. I don't have all the answers, but
maybe we can help each other.
 
#86
herm, Im not sure why you are posting that to me. Im in 100% agreement with John. Remember John posted " 20 grams of added mass added to a cone will not affect the time domain, only the magnitude of the impulse response. "

I was just pointing that out again, my point was in reference to time domain. btw, you can not talk about time domain without acceleration since time domain is one of the variables in acceleration.
 
Last edited:
#87
Duh - I see you posted: "But the paper showed that adding mass alone did
nothing. This means that the ratio is not really a 100% meaningful since the
value can be all over the map with different weights." You obviously get it.

Somehow when I was writing my response I got the idea you had disagreed!
Sorry.

Anyway, did my write-up make sense? My goal is to point out that John
and Dan are talking about Jerk, not Acceleration. That's why the physics
we learned in school arent helping us to understand this phenomena. And
why there are always huge discussions that go nowhere when this
concept is brought up.

I hope I do more good than harm with this topic...

-herm
 
#89
Yes, Im the one that asked about the dip at 700Hz originally (Im penngray on there).

btw, yes, your post make sense. Im definitely not an expert so I will pose opinion to induce discussion. Im learning each minute as the discussion moves along 😀
 
#90
Overkill Audio said:
Can you give a once and for all clear cut yes or no to the basic question:
Does adding / subtracting mass from the cone affect acceleration / deceleration? Yes or no I promise to drop the subject....!
Click to expand...
Well of course it does and John never claimed otherwise. Acceleration is proportional to sound pressure so a light cone will be more sensitive/efficient. BUT.... only at high frequencies above the in-box impedance minimum where mass controls. Below the impedance minimum, the suspension stiffness, including the box's air spring, controls.

There's no free lunch. What you see with a lot of very high Bl and low mass drivers is the Qts is so low that the low end drops way off and you end up losing much of the HF sensitivity gain by the time you EQ it flat in the crossover.

As Herm points out, John was talking about the rate of change of acceleration. That's just another way of saying high frequency response. A driver that can play higher will be able to change acceleration faster. Mass has very little effect on that, just how loud it can play with a given power.
 
Last edited:
#91
catapult said:
Mass has very little effect on that, just how loud it can play with a given power.
Click to expand...

Sure it has
If not, it would be fast enough to use as tweeter

Ofcourse mass slows it down
Thats the point of mass
Or it wouldnt be able to play low
Mass weighs out the spring

Try look at a 2hz signal, and see how slow that will be
Well, ofcourse I have never experienced it, but I bet it will look like slow motion

From this I would conclude that below a certain frequency anything turns into one note signal
 
#92
tinitus said:
Sure it has
If not, it would be fast enough to use as tweeter

Ofcourse mass slows it down
Thats the point of mass
Or it wouldnt play low
Mass weighs out the spring

Try look at a 2hz signal, and see how slow that will be
Well, ofcourse I have never experienced it, but I bet it will look like slow motion
Click to expand...
I have to disagree. Run an SPL sweep on a woofer. Then put some weight on the cone (Mortite, whatever) and run the sweep again. It still plays just as high but at a lower level (above the impedance minimum). Acceleration is proportional to sound pressure. Rate of change of acceleration is proportional to HF response -- how quickly can it change from accelerating forward to accelerating backward.

Your 2 Hz signal would be unaffected by mass. It's just moving too slow for mass to matter. It's simply the motor positioning the cone against the suspension stiffness. At the impedance minimum, mass and stiffness control equally. Above there, mass controls more. Below there, stiffness controls more.
 
Last edited:
#94
For low ressonance/Fs/frequency, in real life size is the main factor

In speaker design this need for huge size is dealt with by using supported weight instead
And smaller means more movement
The more you cheat, by making the model smaller and heavier, the worse and more flawed the model becomes

Then theres the electrical aspects 😱
And the request for SPL, which is another not natural parameter that flaws the model even further

Yeah, seems it does make sense to use 18" woofers, and stay at moderate listening levels 😀
 
#96
tinitus said:
...

Yeah, seems it does make sense to use 18" woofers, and stay at moderate listening levels 😀
Click to expand...

I've been using several 18" woofers in my main system and they're working fine.

Playing loud to REALLY LOUD (in my living room), they are great. However this seldom happens because I don't want to bother the neighbors. When playing at moderate levels (around or slightly louder than conversations), they are just fine (no so great) 🙄 But this is the most frequent usage.

When playing very quiet (late at night), TBH, they're not as good. The sense of effortless (of such big drivers) is still there, but it obviously lacks the micro-dynamics and details. Those transient attacks are rounded, textures are smeared. 🙁

Maybe it's the hearing, just like we can not see every detail of things in a dark enviornment. I'm not sure and I can not measure. Just observation, subjectively.

Or, it's really the case as mentioned above - the cone is moving so little and the deformations among VC/cone/suspension are large enough (proportionally) to mess up the original signals.

Anyone else has the same feelings?

So I've been thinking, 15" might be a better choice...
 
#97
tinitus said:
Hey, wait a minute
Are you suggesting that thinner poleplates are better than thicker poleplates, in terms of flux modulation
It sounds like thats what you say
Click to expand...

No that is not the case most of the time. You have two sides to it though. A thicker top plate gives you more spread out flux field and typically a broader BL curve. Less Bl(x) variation is good. However, if you can't saturate that top plate, the flux will move more. You have more Le(x) and Le(i) then unless you use a proper shorting ring to fix that issue. You ahve to weigh the two options. Ideally thicker gap plate that is fully saturated close to the coil is best.

Overkill Audio said:
John_E_Janowitz said:
But, ( here we go...) the whole point of the old Aidire paper was something very different i.e. it claimed adding mass to the cone did not affect the acceleration / deceleration. Now that is 100% wrong.

Can you give a once and for all clear cut yes or no to the basic question:
Does adding / subtracting mass from the cone affect acceleration / deceleration? Yes or no I promise to drop the subject....!
Click to expand...

Adding mass to a cone does NOT make changes to the time domain as I mentioned before. It does not change acceleration. The cone will start and stop at the same times. What it does do is decrease the magnitude of the peak and as a result the velocity. So lets take a 20hz sine wave and say the driver moves from rest to 10mm outward and back to rest. A full cycle would take 1/20 of a second so 1/2 cycle is 1/40th of a second.

Now take that driver and double the mass, or whatever you wish to do. It now will not go forward as far, lets say only 5mm now. But it still starts moving and comes back to rest at the exact same time. Again 1/40th of a second from rest to positive peak and back. The acceleration or rate of change is the same.

Now, again you have to consider current. In the second heavier driver, increase current so that the driver moves from rest to 10mm and back to rest in the same amount of time again as the first driver. Speed, acceleration, velocity are all equal to the first case now.

By the standard you are stating, the higher the volume level, the "faster" the driver would be because the cone travels a farther distance over the same amount of time. Speed would be the rate of change of position, so technically speed does change. Velocity will change as it has to go farther over a time period. Acceleration is the rate of change of velocity, and acceleration does not change.

However, adding inductance does affect the time domain and the time at which the impulse starts.

Another elephant in the room :THERMAL COMPRESSION due to voice coil heating. This has a far greater effect on resistance than the variations in the magnetic force exerted on the coil. Please read Michael Gerstgrasser brilliant paper attached on the subject.
In addition I would point out that most of the dramatic graphs and figures used to illustrate the variable Bl force refer to extreme cone movements i.e. close to or exceeding Xmax. This is (a) misleading as 90% of the time drivers are not bouncing off their end stops and ( b) is totally avoidable by simply choosing the appropriate drivers i.e. use large fast drivers with low Mms, high Bl and high power handling and run them well within their limits so you never use more than 30% to 40% of their Xmax. This cures the variable BL problem in the vast majority of quality pro drivers, also it ensures the more serious problem of thermal compression is reduced to very low level i.e. less than 0.4 dB in the PD 158 case when I am running it a full power ( 300 watts between140Hz and 600Hz ).
Click to expand...

Your statement above contradicts the other part. As you have stated, you want the driver to move fast or have high speed, meaning traveling farther over time. Again this means it has to travel farther over the same time period requires the driver moving to farther excursions. If you want to keep excursion down, then the speed is decreased. 😉

Power compression raises resistance across the board. This has less effect on the upper end of the response curve than does the varied inductance. The varied inductance changes the shape of the curve, not just the magnitude. It is altering phase continually throughout the stroke, many times per second. This creates IMD and other distortions. A broad rise in resistance of the coil does not. It simply lowers the magnitude of the whole curve.

Also, keep in mind that your requirements are not always what others are. If you wish to have low listening levels, you may be able to keep drivers within 30-40% of Xmax quite easily. However if you want to do reference levels at 10-16hz for home theater, this is just not something reasonable. Increasing the amount of drivers by 3x or more when you may already require 8 15" woofers or so isn't possible. Even in an open baffle speaker, many people require +/-12mm or more from 15" drivers to get the headroom they want at 30hz. Going with 6 15's per side to keep the drivers at a lower Xmax is often not practical from a money standpoint or for the room it requires. Creating drivers that have lower distortion and can cleanly play to the excursion levels required is much more practical.

Again, a low mms, high Bl driver means nothing alone. I can take the same dual VC woofer and wire the VC in series or parallel to change the Bl. Series will double the Bl/mms relationship. By your standards that should make it twice as fast. Does it become faster with the VC in series than with the coils in parallel? Of course not. In a case like this you actually double bl, but as you are increasing the impedance the excursion is lower. Going to the examples above, you have now a "slower" driver as you have less distance over time. In reality Bl^2/Re needs to be looked at to even compare motor strength. You could look this motor strength compared to the mms, but without current it still tells you nothing.

John
Click to expand...
 
#98
Now we are getting to it...!

Hi John,

Thanks for taking the time to reply in such detail and with real clarity that I can finally understand without the help of a maths professor... I never claimed to be a maths expert !
I am really short of time right now and in order to do your reply justice I need to wait for a few days to reply.
As you have only given a " under certain conditions " yes / no answer to my simple question I don’t think I am actually breaking my promise to drop the subject!

Inductor- Please cease and desist with your personal attacks on Doug or any other of our civilised forum members.
Lack of understanding does not excuse extremist attacks...
Moderators- please keep an eye on this one.
 
#99
History

More great stuff!

Dennis (Catapult) it's gonna take me a while to reconcile the concepts you presented. I've never thought of this in terms of mass-controlled response and suspension/box controlled reponse.

And John - I'm right there with Overkill in that it will take a day or two to study what you've said.

Thanks both of you for taking time to explain things to us (me)!

For the new DIYaudio members - Here's another look at this same topic from 10,000 years ago:

http://www.diyaudio.com/forums/multi-way/27810-voice-coil-inductance-vs-transient-response.html
 
#100
Last thoughts on low Le... life is too short!

Hi John,

Sorry about the late reply.
All my time is taken elsewhere right now and as I have no commercial interest in debating ( again...!) the importance of High / low Le values in bass or bass-mid drivers I find the subject very boring. It’s a classic circular debate dependant on how the drivers in question will be used ( passive, active, sealed or open baffle etc ).
But having read your reply two things are clear, (a) I still disagree with you ( sorry!) and (b) it’s of ( at best ) minor importance if using passive crossovers & trying to run bass / mid drivers above 1Khz, and of no importance at all when running high quality Pro drivers in an active system.
Rather than trawl up mathematical arguments I refer you to better qualified ( than I..!) members of this site who have disagreed with your maths in the various threads that the old Aidire / Dan paper rears its ugly head / beautiful face ( you decide! ) in.



Here are a few of my thoughts in bullet points ( time is short!) on your last post, covering the whole paper is not worth the effort for me.


(1) Johns quote “ Adding mass to a cone does NOT make changes to the time domain as I mentioned before. It does not change acceleration. The cone will start and stop at the same times.” I totally disagree! Of course adding mass to a “ mass / spring “ system will affect ( decrease ) acceleration and deceleration. I refer you to my attachment on how “ piston “ drivers actually respond to real music signals.

(2) Johns quote “ So let’s take a 20hz sine wave and say the driver moves from rest to 10mm outward and back to rest. A full cycle would take 1/20 of a second so 1/2 cycle is 1/40th of a second. “ NO it would not....! Any theory predicated on this assumption is fatally flawed. Again see my attachment.


(3) Johns quote “ But it still starts moving... at exactly the same time “ I AGREE ...! Although with REAL MUSIC the cone does not start from the start or “ rest position “ as it is still oscillating out of control from the previous impulse ( drum strike, vocal, piano note etc )

(4) Johns quote “ The acceleration or rate of change is the same. “ No it is not. A mass of X will accelerate at a given rate when a force, Y, is applied. If you double the mass of X it will accelerate at a reduced rate. To use your favoured automotive analogy... If your 200 BHP Volvo estate ( approx 1,750 Kg ) accelerates from rest to 60 Miles per hour in 10 seconds with just the driver in it what will happen if you add another 1,750 Kg of lead weight ( electric car anyone....!) in the boot? It will accelerate at a reduced rate!!! It will also DECELERATE at reduced rate. Heavy coned drivers are Volvo estates full of bricks oscillating on giant elastic ropes....!

(5) Looking at your graphs is interesting. First, these graphs ( and of course the whole paper ) are based on a small 6.6 inch bass / mid driver ( more midrange as its over 10dB down at 60 Hz...), not a serious 12, 15 or 18 inch driver as is the subject of this whole thread... Regardless of that, frequency response first: Does this not say it all, there is almost no difference at all when you ( 78% ) increase the Le ( green trace ) compared to the raw driver ( red trace )? You regularly see far more discrepancy with driver to driver tolerance... only in the last octave of its range at around 2.5KHz to 3.5Khz is there ANY difference at all, and here it is still within a typical plus or minus 2 dB tolerance. PLUS look at the 100% increase in mass driver ( the blue trace ) it, according to your graph, out performs ( greater efficiency) both the raw driver and the high Le driver above 1,800Hz and has exactly the same high frequency extension as your reference driver...! I really am doubtful of these experiments, sorry but they don’t add up.

(6) Now looking at the impulse tests... Again you really have to split hairs to interpret the VERY small differences in the graphs. Here again I question the results of this experiment. You are doubling the mass in a mass / spring suspension system and the settling times are identical.... this just ain’t Penguin....!

John, Could you not hear any difference in the sound quality when you doubled the mass on your driver....? I know it won’t give you a USP or a patent to sell more drivers but hey isn’t sound quality the most important element at the end of the day?


Herm, if you are correct, you can happily use your big heavy coned, low Bl, rubber surround subs as midrange drivers as long as it has a low Bl....


Cheers

Derek.
 

Attachments

Status
Not open for further replies.
Top Bottom