DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 86 Posts

·
Registered
Joined
·
74 Posts
Discussion Starter #1
Hello anyone who may happen to read this, first of all I would like to say I have two questions, whose correlation, or lack thereof, to each other has yet to be determined, and may possibly be determined by the answer to one of them.

Firstly, I've been debating internally over the importance of certain T/S parameters, and their impact on the quality of my music, as percieved by me, the listener.(subwoofers, btw)
1. Inductance (Levc) - I've noticed that I tend to prefer subs with a lower inductance, my current sub (Jbl w15gti mkii) has a very low inductance of .75mH, and I do very much love the sound of this sub. My question is, has anyone else noticed that subs with lower inductance sound better? Is it just me? Is this coincidence?
2. Sensitivity (Splo) - Why does it seem that people neglect these sensitivity ratings? I've, personally, always considered it to be on of the more important parameters. Will a sub like mine, with a 92db sensitivity not play just as loud as a sub with an 86db sensitivity on only a quarter of the power? (considering that both enclosures are ported and tuned to the same frequency)
3. Xmax - what is the real importance of this specification? I understand that, in general, a sub with higher Xmax will play the lowest notes (20-35hz) more easily, but if a sub can play within 3 decibels of a linear response from 15 to 500 hertz, but only has a 15mm Xmax, why would it matter when compared to another sub with higher Xmax? Is this specification only for show?

And finally could anyone recommend another sub for me to try out? I love my gti, and chose it after listening to dozens of subs available at local reatialers, but i would like to learn more about the less well known brands.

Thank you anyone and everyone who may reply, I'll be looking forward to some intelligent answers or thoughts.
 

·
Banned
Joined
·
4,394 Posts
the nature of specifications is such, that relying on them to make a purchasing decision can be hit or miss.

first, you have to have standardization. Many companies use specs like they use wrapping paper.

second, you have to understand how variances in the specs make subwoofers sound differently from each other. it's a balancing act and if you push the specs in one direction, for say, small box size, you might find unwanted side effects crop up. Most people find the natural balance of speed/agility/output, a useful trade-off to get the low lows in a small box.

third, or thirdly, you have to believe or proof the specs given by manufacturers with independent reviews. Many a manufacturer has passed off great specs on average subs. the Klippel distortion analyzer is a divining rod, separating wheat from chaff, etc.

then the forth, or fourthly, is deciding implementation, goal-centric behavior, goal-seeking behavior, goal-oriented behaviors.

is the roof flex more of a happy, is the hair trick a special kind of club where once you're into it, you don't care who knows it? Is it psychological, do you have a need?

we could plead the fifth, and fifthly, but that isn't necessary...

:)
 

·
Registered
Joined
·
4,982 Posts
Sensitivity ratings are usually pretty meaningless with subs. That rating is often specced in the midrange or simply derived from T/S parameters. What enclosure you put the sub in affects efficiency quite much though. Xmax is also important, if it's rated properly from a klippel analysis or similar... it dictates displacement together with cone area in its "linear range". Dunno about Levc though, I've heard this before but it might be connected to other things as well.

Tapaaatalk!!
 

·
Registered
Joined
·
74 Posts
Discussion Starter #4
So the reason I'm asking all of these things is because I was looking into getting a louder sub, but I want to at least maintain but hopefully improve the sound quality of the system.. I have a dd 9512g sitting around, and I've made a box for it, but would it perform well with onlyt5-600 watts? If I got an re xxx would that do better?
 

·
Banned
Joined
·
4,394 Posts
So the reason I'm asking all of these things is because I was looking into getting a louder sub, but I want to at least maintain but hopefully improve the sound quality of the system.. I have a dd 9512g sitting around, and I've made a box for it, but would it perform well with onlyt5-600 watts? If I got an re xxx would that do better?
a man has 3 apples, one a Granny Smith, one a Gala and another Red Delicious, on his table.

that man wants to know which apple tastes better.

I say to that man, eat the apples.
 

·
Registered
Joined
·
74 Posts
Discussion Starter #6
Haha yeah I know, spl, sql or sq? The war will persist in my mind until I have tried every driver and enclosure available to me! Really looking for sql though I just can't find anywhere near me to audition the dd or the re, and I don't wanna pay to have the dd installed then uninstalled etc etc..
 

·
Registered
Joined
·
1,681 Posts
So the reason I'm asking all of these things is because I was looking into getting a louder sub, but I want to at least maintain but hopefully improve the sound quality of the system.. I have a dd 9512g sitting around, and I've made a box for it, but would it perform well with onlyt5-600 watts? If I got an re xxx would that do better?
If you've only got 500-600 watts what are you gonna do with the extra 1500 watts and likely 40mm extra excursion the XXX can handle?

It seems you're going about this all wrong. If you're limited in power you want a subwoofer that maximizes efficiency.

But you're still battling Hoffman's Iron Law. It distills the basics of subwoofer design into three basic goals. You can have efficiency, a small box or low end extension. But you can pick only two.

You've got yourself a big bad subwoofer but luckily it's one that's designed for efficiency. The XXX simply isn't. Unless you have lots of power the low efficiency of the XXX will get you less sound for your watts.

So what I would definitely do is build a box that's well suited for the DD sub and compare it to your WGTi. You may like the sound of the WGTi or the DD. The thing with the DD is with such a powerful motor and low QTS you'll definitely get a lot of output going ported. But the caveat is you'll still need more power. With the limited wattage you've got you'll get better output with the more efficient WGTi and porting the WGTi you'll get a subwoofer that plays lower and is punchier than the DD which really was made more for competition and burping than ultimate sound quality.

How much louder are you really going for though? Because the WGTi is capable of quite a bit of output as long as you build the box correctly and can give it enough power. It's definitely no slouch, it just can't hang with the dedicated burp subs.
 

·
Registered
Joined
·
1,348 Posts
On the LEVC front their is a system effect that is not often discussed that is proportional to the Le of a driver, Back EMF. A high inductance driver is going to produce more back emf than a lower inductance driver. Depending on the output section of the amplifier it may or may not have much or even any effect, but with most class D amplifiers and the series resistance of their output filter it is much more likely or even certain to make a significant difference in the performance of the system.

Let me start with I only know enough to be dangerous on class D amplifier operation and design. I believe that due to the design of a typical class D amplifier output section and filter they have a very difficult, if not impossible time, making full power into a reactive load.

As for the Xmax thing there is a balance of sensitivity to excursion, such that as Xmax gets higher it yields a diminishing return and about the 25mm one way mark is the limit. You can have more excursion but all you do is achieve the same spl at a higher power level for the most part ( at all but the lowest frequencies) and the Xmax level at which you reach diminished returns is smaller as the frequency gets higher. I am speaking of muscial output and not nasty rumble as you can get a lot output from having a high mass long excursion driver driven to over excursion just making rumble and noise.


Eric
 

·
Banned
Joined
·
4,394 Posts
On the LEVC front their is a system effect that is not often discussed that is proportional to the Le of a driver, Back EMF. A high inductance driver is going to produce more back emf than a lower inductance driver. Depending on the output section of the amplifier it may or may not have much or even any effect, but with most class D amplifiers and the series resistance of their output filter it is much more likely or even certain to make a significant difference in the performance of the system.

Let me start with I only know enough to be dangerous on class D amplifier operation and design. I believe that due to the design of a typical class D amplifier output section and filter they have a very difficult, if not impossible time, making full power into a reactive load.

As for the Xmax thing there is a balance of sensitivity to excursion, such that as Xmax gets higher it yields a diminishing return and about the 25mm one way mark is the limit. You can have more excursion but all you do is achieve the same spl at a higher power level for the most part ( at all but the lowest frequencies) and the Xmax level at which you reach diminished returns is smaller as the frequency gets higher. I am speaking of muscial output and not nasty rumble as you can get a lot output from having a high mass long excursion driver driven to over excursion just making rumble and noise.


Eric
can we infer from this, that if you own a class D amplifier for your sub section, that choosing a low inductance driver is perhaps more important in your decision-making process than poring over manufacturer supplied specifications on frequency response and sensitivity?

is it that notable that pairing an older, non-faraday complement motor with one of the new "nano" class D amplifiers is a no-no, if sound quality objectives are your design goal?

and would it also be congruent to say using high damping factor class AB amps on low inductance drivers, should result in an even greater measure of control and perhaps audible performance increase, so that no other amp/driver combination can best it?
 

·
Registered
Joined
·
1,348 Posts
can we infer from this, that if you own a class D amplifier for your sub section, that choosing a low inductance driver is perhaps more important in your decision-making process than poring over manufacturer supplied specifications on frequency response and sensitivity?

is it that notable that pairing an older, non-faraday complement motor with one of the new "nano" class D amplifiers is a no-no, if sound quality objectives are your design goal?

and would it also be congruent to say using high damping factor class AB amps on low inductance drivers, should result in an even greater measure of control and perhaps audible performance increase, so that no other amp/driver combination can best it?
To draw absolute conclusions I would want to do a controlled comparison test and possibly due some measurements to confirm the comparison results. In my experience I have experienced some dramatic differences between a high quality typical class D sub amp and a class A/B and discussions with Engineers is how I have come to my beliefs. In my talks we have bounced around the idea of building and active reactive load that would handle the power necessary to measure this.

Eric
 

·
Banned
Joined
·
4,394 Posts
To draw absolute conclusions I would want to do a controlled comparison test and possibly due some measurements to confirm the comparison results. In my experience I have experienced some dramatic differences between a high quality typical class D sub amp and a class A/B and discussions with Engineers is how I have come to my beliefs. In my talks we have bounced around the idea of building and active reactive load that would handle the power necessary to measure this.

Eric
but subjectively, in your experience, (which I might add, carries a considerable amount of weight and substance, to me at least) you would remain in the camp that states class AB amplifiers are able to give a higher sound quality in long-term listening examination in comparison to most class D amplifiers available today?

and I'm not saying you have to back that up using a standardization test such as you allude to with a reactive load, but within the many amplifier/speaker combinations you have personally evaluated over decades, you'd consider class D amplifiers as a class, as not being able to deliver the same sonics when the load is producing a lot of back EMF, and an AB class amp can be shown to control a sub with those characteristics, better than class D?

I find this point of contention very much to the heart of amp sound controversies, and one area where a clear line of demarcation may exist. It would be an impetus for someone like captainobvious who would want to subject this notion to a test...

unless of course, we are already able to refer to a test with published results, on this very issue?

thanks, for taking the time to respond, Mr. Stevens. It's always a pleasure to converse with an industry legend.
 

·
Registered
Joined
·
1,348 Posts
but subjectively, in your experience, (which I might add, carries a considerable amount of weight and substance, to me at least) you would remain in the camp that states class AB amplifiers are able to give a higher sound quality in long-term listening examination in comparison to most class D amplifiers available today?

and I'm not saying you have to back that up using a standardization test such as you allude to with a reactive load, but within the many amplifier/speaker combinations you have personally evaluated over decades, you'd consider class D amplifiers as a class, as not being able to deliver the same sonics when the load is producing a lot of back EMF, and an AB class amp can be shown to control a sub with those characteristics, better than class D?

I find this point of contention very much to the heart of amp sound controversies, and one area where a clear line of demarcation may exist. It would be an impetus for someone like captainobvious who would want to subject this notion to a test...

unless of course, we are already able to refer to a test with published results, on this very issue?

thanks, for taking the time to respond, Mr. Stevens. It's always a pleasure to converse with an industry legend.
Yes the class AB amp can be shown to control the subwoofer better than a "typical"class D subwoofer amp. a well noted engineer I have spoken at length about this observed repeatedly that at the same power levels on the same subwoofer system the subwoofer when power by the class D was moving in an uncontrolled manner that was clearly visible to any observer. The same subwoofer when powered by a class a/b amplifier at the same power level was controlled and was moving far less. In his investigation he determined that the back emf was dynamically modulating the damping factor of the amplifier or stated in other terms reducing the amplifiers amility to deliver current into the load.

Not sure how the switching output amps with feed back and control loops with smaller output inductors do in the grand scheme of things. But my intuition tells me that it would need to switch at a very high frequency and use a simple first order filter to perform at the same level as a class A/B.

My opinion is that at power levels below 1500 wrms class A/B is the better solution since at 1/3rd or 1/2 power the current consumption is still reasonable enough for a stock charging system. Only disadvantage is the larger size needed for the heatsink so a variable rail voltage design like H or G would allow you to have your cake and eat it too.

Something I have been told also is that most if not all companies that give damping factor for their class D amps list the damping factor prior to the output filter. The series resistance of the output inductors will dramtically alter the damping factor specifications both static and dynamic. Only meaningful damping factor is at the output terminals of the amplifier IMHO when stated into a given load at a given frequency.

Eric
 

·
Banned
Joined
·
4,394 Posts
thanks Eric.

so, if another congruent relationship exists, as those speakers with large voice coils and strong magnets produce ever higher levels of BL, would it also be fair to say that the trend towards higher Le can be observed in mid bass frequency transducers, and that the combination of Dynaudio/Morel and class D amplification is possibly subject to this same criticism?

And the use of XBL technology, or neo-radial technology, with their inherently shorter coils and lower inductive mess, along with most underhung topology product, will pair up better as a general observation, with the class D amp design?

it's as if there is a parallel that can be observed where the move away from EMF generating motor structures, works with the new technology amplifiers.


Like running EON, JBL differential drive pro audio off of class D plate amps built right into the speakers...

I kind of have to wonder if the move to switched power supplies is as much a profit generator as it is a conservator of raw materials, since power huge transformers or their subwoofer motor counterparts, are being shifted from the public domain in favor of more efficient design, and the sound quality is promoted to not be affected, when it can be demonstrated that it has?

maybe that's just the general trend, in technology as Moore's law dictates the miniaturization of materials so more people can get away with less natural resource depletion, kind of a big picture thing.

or it could simply be profit driven, anyways it's good to see the return of the amplifier as an audible modifier of the signal, good clean, usable signal...

:)
 

·
Registered
Joined
·
1,681 Posts
If the output inductor was such a bane to sound quality because of its series resistance/reactance, then you could also infer that a length of speaker wire on an A/B amplifier of equal length if coiled up will similarly act as an inductive filter increasing inductance and lowering cone control as well.

If you assume that the speaker wire isn't affecting the A/B amps damping factor then you can also assume the output inductor on the amplifier shouldn't affect the damping factor much if any on a properly designed amplifier.

And modern class D amplifiers operate at much higher frequencies now, hence why you can get full range class D's so that the inductors necessary become much less inductive, and also present minimal if any DC resistance, which reduces their overall effect they can have. This is especially true with subwoofer amps as the only way back emf can be affected by the output inductor is if the back emf is at a frequency that the output inductor filters out, will definitely reduce the output sections ability to control the cone.

But in terms of subwoofer amplifiers it hasn't been my experience that I can hear any difference in cone control. You can simulate loss of cone control really easily, put a series resistor inline with the output, the more resistance you add the less the cone can be controlled, you'll realize unless you really go over half the series impedance of the subwoofer you really can't hear a difference. So unless the damping factor is so bad as to be less than 2 you're really not going to hear it in a subwoofer.

Sure it'll be "visible" probably on an o-scope or servo sensor, but audible? I've yet to really hear it.
 

·
Premium Member
Joined
·
4,235 Posts
Uh oh...

The output filter on a class D amplifier is indeed a passive low pass filter. If the coil is too small, then it may have a high DCR, which is a series resistance. In order for it to have much of an effect on the circuit damping, the DCR would have to come very close to the same resistance as the voice coil's DCR. If you're using a 1/4 ohm speaker and a class D amplifier, this might actually happen. If you're using a 1 ohm speaker, it might also happen, but probably not. If you're using a 4-ohm speaker, then it's likely not going to be an issue.

damping in the circuit is a matter of the speaker dissipating the energy stored in the coil's inductance, but it's the coil's DCR that regulates the flow of current, whether that current comes from the back EMF or from the amp, so long as the DCR is the LARGEST resistance in the circuit. That's precisely why we choose woofers that have a 2-ohm, 1-ohm or 4-ohm coil. To cause the amp to make more or less current.

It's a sloppy generalization to say that Class-D amps are better or worse in terms of screwing up transient response due to really high series resistance. It's less sloppy to say that cheap class D with with either no provision for minimizing the effects of the output impedance or with high DCR filters may cause poor performance IF the output impedance is similar to the woofer's DCR.

When you do this visual test, be sure the amp isn't clipping.
 

·
Premium Member
Joined
·
4,235 Posts
If the output inductor was such a bane to sound quality because of its series resistance/reactance, then you could also infer that a length of speaker wire on an A/B amplifier of equal length if coiled up will similarly act as an inductive filter increasing inductance and lowering cone control as well.

If you assume that the speaker wire isn't affecting the A/B amps damping factor then you can also assume the output inductor on the amplifier shouldn't affect the damping factor much if any on a properly designed amplifier.

And modern class D amplifiers operate at much higher frequencies now, hence why you can get full range class D's so that the inductors necessary become much less inductive, and also present minimal if any DC resistance, which reduces their overall effect they can have. This is especially true with subwoofer amps as the only way back emf can be affected by the output inductor is if the back emf is at a frequency that the output inductor filters out, will definitely reduce the output sections ability to control the cone.

But in terms of subwoofer amplifiers it hasn't been my experience that I can hear any difference in cone control. You can simulate loss of cone control really easily, put a series resistor inline with the output, the more resistance you add the less the cone can be controlled, you'll realize unless you really go over half the series impedance of the subwoofer you really can't hear a difference. So unless the damping factor is so bad as to be less than 2 you're really not going to hear it in a subwoofer.

Sure it'll be "visible" probably on an o-scope or servo sensor, but audible? I've yet to really hear it.
If you really want to know the effect the series resistance has on the transient behavior of the speaker at resonance, just add it to the calculation for Qts of the driver. Q is a measure of damping.
 

·
Registered
Joined
·
1,348 Posts
I am pretty certain that this is not as simple as a model of output impedance vs load impedance and circuit damping. The switching output amplifier works in a very different way to the class A/B amplifier and I believe that the back emf has a significant effect on the output.
 

·
Registered
Joined
·
1,681 Posts
I am pretty certain that this is not as simple as a model of output impedance vs load impedance and circuit damping. The switching output amplifier works in a very different way to the class A/B amplifier and I believe that the back emf has a significant effect on the output.
You believe but you do realize that if the back EMF on a switching section of a class D amp really did effect its output, then A/B amps would be just as affected.

A/B amps have a switching power supply. Therefore the output of the switching power supply after going through the varying resistance of the output transistors would be EVEN more susceptible to back EMF because there is no output filter, the power supply is essentially directly connected to the speaker via the output transistors and any back emf would work its way to the power supply. It doesn't have an effect because the negative feedback in the output section would prevent back emf from really affecting the power supply and output section.

So the fact that class D amps don't have a separate output section, essentially the power supply is the output section and has the same built in negative feedback, there really isn't any more susceptibility to back EMF than the A/B design.
 

·
Banned
Joined
·
4,394 Posts
You believe but you do realize that if the back EMF on a switching section of a class D amp really did effect its output, then A/B amps would be just as affected.

A/B amps have a switching power supply. Therefore the output of the switching power supply after going through the varying resistance of the output transistors would be EVEN more susceptible to back EMF because there is no output filter, the power supply is essentially directly connected to the speaker via the output transistors and any back emf would work its way to the power supply. It doesn't have an effect because the negative feedback in the output section would prevent back emf from really affecting the power supply and output section.

So the fact that class D amps don't have a separate output section, essentially the power supply is the output section and has the same built in negative feedback, there really isn't any more susceptibility to back EMF than the A/B design.
but isn't the ability of the amp circuit, to modify the signal according to changes in impedance a general design, and all AB class amplifiers aren't "susceptible" to back EMF when connected directly to the output transistors, because the current that comes from the transistor section is stronger than any back EMF, basically because what you put in, minus the work (energy loss as acoustic output) can never exceed what the speaker can generate?

the class D design has a series inductor, and that series inductor absorbs and modifies the current passing through it just like as if it were another speaker attached to the one you're powering, and reactance from that magnetic field created by the inductor is going to take away the amp's ability to control the speaker, we see this as the loss in actual damping factor, down to near single digit levels in an otherwise tightly regulated and controlled amp section.

if you can have class D design with a series inductor that is so high in frequency it has no impact on the circuit or whatever other measurable parameter, amps, watts, volts, ohms... obviously I have no grasp of what's going on, except to agree that class D amplifiers in some permutations have had audible, degenerative impact on the signal coming out of speakers, since day one, and if that design criteria has been nullified by new technology and back EMF is no longer able to modulate the class D amp's circuit as a result, that would be good to know as well.
 

·
Registered
Joined
·
1,681 Posts
I definitely do remember old class D's that definitely had low damping factors. And you knew it. They didn't sound god at all and it was obvious.

But nowadays with decent design if that loss of cone control were present we wouldn't be having a discussion. Everyone would just say give it a listen you'll be able to tell. Since we can't do that anymore, can't hear an objective difference, and I distinctly remember the old class D's had cone control problems. We can infer that the inductors are of such very low inductance as to not be in any resonant mode that can affect audio.

The full range class D amplifiers nowadays operate at a switching frequency in the megahertz range, I remember early class D car amps in the mid 90's were much lower, basically modified car audio power supplies in the 40 khz range, and these are the amps that gave class D its reputation it's still trying to shake off even if the technology today is so much more advanced. They needed huge high value inductors to filter out the noise. And their DC resistance most definitely affected cone control especially if you ran two ohms or less.

The output inductor needed on those ultra high frequencies Mhz frequencies is simply so low in value that it wouldn't affect the audio. This is why I say if it was a large enough inductance to affect audio, simply loop the speaker wires on an A/B amp and see if you can tell the difference in cone control loss just by hearing it, because that's the equivalent . You can't because the added inductance is so little and the relative frequency of the back emf is so low that there wouldn't be a change in audio. And that's the output inductance levels we're talking about here.
 
1 - 20 of 86 Posts
Top