cabinet frequency response

[12bass]

Interesting to read about the phenomenon of power compression. Reminds me of the oft-repeated mantra of matching the amp's watts to the cabinet's watts. This line of reasoning suggests putting more power into smaller cabinets with fewer drivers to make an ideal power "match", a strategy which is an ideal recipe for increased power compression and overexcursion. Glad to see some more in-depth discussion which dispels such notions....

 

[AstroSonic]

Interesting to read about the phenomenon of power compression. Reminds me of the oft-repeated mantra of matching the amp's watts to the cabinet's watts. This line of reasoning suggests putting more power into smaller cabinets with fewer drivers to make an ideal power "match", a strategy which is an ideal recipe for increased power compression and overexcursion. Glad to see some more in-depth discussion which dispels such notions....

You got it. Most drivers (yes, even the 'super' drivers) have well under 5% efficiency. That basically means that for every 100 watts input about 5% are turned into sound while the rest is turned into heat, mainly in the voice coil. The voice coil gets very hot. The resistance goes up with temperature, so that voice coil resistance goes up by a factor of typically 2-3 (some are better than others). That being the case, the spl increase from doubling the power is well under the expected 3db at high power levels. Further the loudspeaker's spl is no longer following the electrical signal (which is why this phenomenon is know as thermal compression - the peaks are flattened). So the intrepid bass player turns up the master volume. Voice coil temperature increases even more, and the volume increase is even less. Kind of a downward spiral. For some well designed drivers thermal stabilization occurs (at least within the rated power handling) and the amount of thermal compression stabilizes. For lesser drivers the 'spiral' results in voice coil burn out. Note that reports of increased spl from turing the master volume up at high power levels are most likely due to amplifier overload distortion, which is perceived as increased spl. A distorted boom box will cut right through a clean high spl sound field even though its spl is much lower.

I would like to see thermal compression curves for the new high power drivers. Thanks to improved adhesives and thermal management designs, I've little doubt that these drivers will survive the stated high power levels, but I have some doubts about their ability to deliver the high spls suggested by those power levels (even with sufficient Xmax).

 

[KJung]

You got it. Most drivers (yes, even the 'super' drivers) have well under 5% efficiency. That basically means that for every 100 watts input about 5% are turned into sound while the rest is turned into heat, mainly in the voice coil. The voice coil gets very hot. The resistance goes up with temperature, so that voice coil resistance goes up by a factor of typically 2-3 (some are better than others). That being the case, the spl increase from doubling the power is well under the expected 3db at high power levels. Further the loudspeaker's spl is no longer following the electrical signal (which is why this phenomenon is know as thermal compression - the peaks are flattened). So the intrepid bass player turns up the master volume. Voice coil temperature increases even more, and the volume increase is even less. Kind of a downward spiral. For some well designed drivers thermal stabilization occurs (at least within the rated power handling) and the amount of thermal compression stabilizes. For lesser drivers the 'spiral' results in voice coil burn out. Note that reports of increased spl from turing the master volume up at high power levels are most likely due to amplifier overload distortion, which is perceived as increased spl. A distorted boom box will cut right through a clean high spl sound field even though its spl is much lower.

I would like to see thermal compression curves for the new high power drivers. Thanks to improved adhesives and thermal management designs, I've little doubt that these drivers will survive the stated high power levels, but I have some doubts about their ability to deliver the high spls suggested by those power levels (even with sufficient Xmax).

Another thing to remember is that there is lots of 'breathing space' between the notes when amplifying a bass guitar. With compressed programmed music, it is a bit different thing (i.e., a DJ cranking tunes for hours at a time with massive sub-bass, etc.). For the typical bass guitar backline application, I would think that most power amps will never reach even close to their rated output, and if they do, it is at very, very brief transients.

 

[AstroSonic]

Originally posted by alexclaber

Our '69er uses a similar format but with one of the 10"s as the mid/treble driver. The issue I've found with cone midranges in musical instrument cabs is that they sound clearly different to cabs without them - yes, they improve polar response but they don't do so without changing the character of the cab. You can't just go sticking mid drivers in cabs, even with really well thought out and carefully designed crossovers and expect everyone to like the end result.

A decent quality midrange, whether cone or horn, will improve articulation and dispersion, which many players find desirable. The sound of a paper cone midrange tends to blend better with a paper cone bass driver than say a titanium horn. To my ear, the sound of a horn with a good compression driver equipped with a phenolic diaphragm and coupled to a cd horn or waveguide can blend almost as well (with the right crossover frequency and topology), but offer better articulation and coverage over a wider frequency range than a cone mid.

Of course, you can't just stick any midrange driver into a cab and have a positive result, any more than you can stick just any 15 inch driver into a cab and have a positive result.

Originally posted by alexclaber

I'm not sure it's so much multi-driver cabs as cabs with a larger amount of radiating area.

In as much as radiating area is 'quantized' (into more or less standard driver sizes) multiple drivers are required to increase radiating area beyond that of a single driver. A 2x15 cab is a multi-driver cab.

Originally posted by alexclaber

...if you want a cab that you can gig in loud bands and still fit in your car you need to do some careful balancing of compromises.

Yes, some compromises are necessary based on the required acoustic power and bandwidth.

Originally posted by alexclaber

I'll just add that I think anyone focusing on F3, F6 and F10 without considering the sensitivity of the system and both forms of power handling is somewhat missing the point. If you have two cabs of equal size, one with 96dB sensitivity and F6 at 45Hz, one with 93dB sensitivity and F3 at 45Hz what's the difference in the low frequency ability?

Hoffman's Iron Law: Size, bass extension and sensitivity: pick any two.

The question is, will a given size cab produce the tone a player wants (bass extension, punch, growl, articulation, presence, etc.) at the required spl, with a sufficiently wide coverage (for the venue), and using reasonably available amplifier power?

Originally posted by alexclaber

Further investigation shows that at very low frequencies the limiting factor is almost entirely down to excursion limitations (both speaker based and port issues).

Yes, it comes down to total volume displacement (drivers and ports).

Radiation resistance may also play a role. A small diameter driver and a large diameter driver producing the same volume displacement [Sd x Excursion] will not produce the same spl. The larger driver has more radiation resistance, so more of the air mass is compressed (or rarified) instead of being pushed aside (acoustic resistance vs acoustic reactance). The load on the cone has both acoustic (radiation) resistance and acoustic reactance. The difference can be relatively small (insignificant?) for similar sized common drivers (like 10 and 12 inch drivers), but could be 'audibly' significant between a single 12 inch and four 10 inch drivers clustered on a baffle. It may be a matter of quantitative insignificance vs audible/perceptual significance.

Radiation resistance was not included by Novak or T&S because it was thought to be quantitatively insignificant and it greatly simplified the analysis. What is quantitatively insignificant may not be audibly insignificant. I am not sure if this contributes audibly to the sound quality (that I perceive) of cabs with relatively high radiating area (like 2x15's, 4x10's, etc.) or if it is some combination of factors. What I hear is a more massive, effortless, immersive sound from cabs with a large radiating area vs smaller radiating area (for example 4x10's vs single 12's) at levels well below displacement limits. The cab with the larger radiating area may have less total displacement volume capability than the cab with smaller radiating area, but it will still sound more effortless, etc. Of course, other performance factors may completely swamp this effect when comparing real world cabs.

And yes, bigger cabinets are, well bigger. Once you reach a size limit (determined by vehicle size, cab weight, etc.), and venting has been optimized, increasing cone excursion is the most effective way to improve low end spl capability.

Those who desire a small, but very capable cab will find a number a good candidates available. Those who desire that massive, effortless sound characteristic of cabs with large total radiating area might just find what they are looking for in a larger, multi-driver cab employing advanced drivers, and constructed with light weight materials and advanced bracing to be a practical solution. The Barefaced multi-driver cabs are good examples. A similarly built 'X by 10' or 'X by 12' cab might also serve well.

 

[AstroSonic]

Another thing to remember is that there is lots of 'breathing space' between the notes when amplifying a bass guitar. With compressed programmed music, it is a bit different thing (i.e., a DJ cranking tunes for hours at a time with massive sub-bass, etc.). For the typical bass guitar backline application, I would think that most power amps will never reach even close to their rated output, and if they do, it is at very, very brief transients.

Good point for amplifiers, provided that the bass player is not intentionally overdriving the output stage (for a particular 'sound'). With respect to drivers, the heat produced by a high power 'transient' may or may not dissipate before the next one comes along. If it's not, then the voice coil temperature gradually increases. The 'breathing space' between the transients can be substantially decreased by compression. Some music genres commonly use a rapid succession of bass notes, with compression and the amp overdriven. Either of these cases spells increased temperature for voice coils.

There has been a tendency in recent years for players to rely on smaller cabs with fewer drivers. 4x10's, 4x12's , 2x15's 6x10's and 8x10's are giving way to 1x12, 2x12 and 2x10 cabs. Yet players want the same, if not more spl. The drivers in these smaller cabs must be able to handle more power (and more heat) while at the same time delivering more spl. I am not sure that we are getting the spl capability that is suggested by the higher thermal power ratings. No doubt, these smaller 'super' cabs are technical wonders, a gift to aging backs and, for many, an easily lived with compromise. Many players are amazed and pleased with their performance, but they may be just a little over-hyped.

 

[KJung]

Good point for amplifiers, provided that the bass player is not intentionally overdriving the output stage (for a particular 'sound'). With respect to drivers, the heat produced by a high power 'transient' may or may not dissipate before the next one comes along. If it's not, then the voice coil temperature gradually increases. The 'breathing space' between the transients can be substantially decreased by compression. Some music genres commonly use a rapid succession of bass notes, with compression and the amp overdriven. Either of these cases spells increased temperature for voice coils.

There has been a tendency in recent years for players to rely on smaller cabs with fewer drivers. 4x10's, 4x12's , 2x15's 6x10's and 8x10's are giving way to 1x12, 2x12 and 2x10 cabs. Yet players want the same, if not more spl. The drivers in these smaller cabs must be able to handle more power (and more heat) while at the same time delivering more spl. I am not sure that we are getting the spl capability that is suggested by the higher thermal power ratings. No doubt, these smaller 'super' cabs are technical wonders, a gift to aging backs and, for many, an easily lived with compromise. Many players are amazed and pleased with their performance, but they may be just a little over-hyped.

+1 Just to be clear, I was using 'compression' as a proxy for programmed music, which from what I understand can put a big strain on a system by having no 'rests' at all (or very few) over large time periods. Of course, you don't have those big peak transients either, but if the DJ is wumping, EVERYTHING becomes a peak transient!

And yes, it is just recently (within the past couple of years) that I've been able to finally retire the 410 format and move to 212's. My current 212's seem to easily handle everything my old 410 could, in a smaller, lighter format with more low end extension, and of course a smidge better upper midrange spread. This would not have been the case a few years ago!

 

[alexclaber]

Our research has found that amplifying bass guitar is pretty much as tough as it gets on speakers! The issue is the vast range of approaches across the spectrum of players, so some want a very dynamic uncompressed clean tone whilst others want to use huge amounts of effects. Some are in very 'professional' bands where everyone leaves each other space whilst some are in very loud bands where 'volume wars' occur. Some are using tube amps, some are using solidstate amps. Some have loads of power available, others don't. Some crank their EQ in demanding ways, others don't.

Yes, DJ's can be demanding but they're in control of their sound, just as FOH engineers are. Although a bassist is technically in control of their sound, they're often an artist expressing themselves through music in an often emotional environment and the sound being produced is a group effort, so if the band is rocking out and getting louder and louder they need to keep up, and an amp that's overdriving and clipping pretty hard can actually sound good when things are reaching a crescendo. That is really really hard on both voicecoils and suspension.

Suffice to say we've blown quite a few drivers whilst on our mission to go another step better.

 

[B-string]

Yes there is no "free lunch" or "magic", only hard work and compromise.

 

[ArtechnikA]

As long as we're porting cabinets and pumping significant amounts of air, has any thought been given to ducting the port air past the voice coils for a 'semi-active cooling' effect? (I understand the issue with damping batting...) Or even in the case of sealed cabinets, employing internal finned heat exchangers thermally bonded to the coils and/or frame and running a (possibly fan-assisted) convection air path through the cabinet?

If 95% of 1000W/cabinet turns into heat, it seems as if even a small improvement in heat dissipation could be A Good Thing...

 

[B-string]

First understand that a tuning port is not an "air vent" and is no longer active above 150 Hz. Forcing air into the air gap between the voice coil and magnet structure also will force dirt/contaminants into it, we REALLY don't want that at all.

A sealed cabinet with a air flow path through it is no longer sealed.

 

[ArtechnikA]

A sealed cabinet with a air flow path through it is no longer sealed.

Sure it is, if the path is not open inside the cabinet...

You conduct heat from the voice coil to a finned structure mounted inside a duct. The duct is open to the outside only.

This is how heaters in air-cooled engines work. Or air-to-air intercoolers in turbocharged engines - the two air streams never mix although they share a thermally conductive structure.

Attached is a photograph of a sealed structure with a port.

 

[B-string]

Sure it is, if the path is not open inside the cabinet...

You conduct heat from the voice coil to a finned structure mounted inside a duct. The duct is open to the outside only.

This is how heaters in air-cooled engines work. Or air-to-air intercoolers in turbocharged engines - the two air streams never mix although they share a thermally conductive structure.

Attached is a photograph of a sealed structure with a port.

A bottle of Downey fabric softner?

 

[ArtechnikA]

Sure, why not ?
It is a sealed structure, with a hole right through it.
It's a visual aid...

Now - you make the structure square-ish, and where the hole is, you attach your thermally-conductive interface. It conducts heat from the inside of the sealed container to the outside. The cooling duct (hole) can be fan-assisted for greater efficiency allowing it to be smaller in cross-section and internal volume (since we're subtracting volume from the cabinet interior. Make it stiff enough (e.g. carbon fiber) and it can probably substitute for internal bracing - perhaps a net gain...

I'll concede the port ducts as cooling mechanism, even though it seems there's significant signal content under 150 Hz in a bass speaker array, and HEPA filters are cheap. But probably not enough energy to work the filter...

 

[B-string]

I suppose it would be an interesting experiment. Speaker motors are vented and would eliminate the pressure differential between inside and outside the enclosure so assist would be needed. You no longer have the natural "pumping" action of the pressure differential. Not sure how this would affect the compliance of the center spider?

 

[dhsierra1]

not ever having put my hand on a speaker magnet that's been worked hard how hot can the driver motor assembly get?

 

[AlexanderB]

Some transducer manufacturers (Volt comes to mind for new-ish gear) have inverted the motor and frame structure and has by that created a big heat sink area OUTSIDE the cabinet. Must work wonders for thermal compression.

Invalid Link Removed

Others, like JBL and Cerwin Vega have had the shorting rings in their woofers for decades. This is also beneficial for "heavy duty" use.

I definitely think things are getting better and better in the loudspeaker world. The vintage gear is available to those that want that and we also get to use state of the art speakers that are amazing engineering and still affordable: woofers with 5" or 6" voice coils, kW in power handling, massive excursion and low weight.

 

[B-string]

not ever having put my hand on a speaker magnet that's been worked hard how hot can the driver motor assembly get?

It would surprize you just how warm they can get, as stated above upto 95% of the wattage that goes into a speaker from an amp just produces waste heat. I've received second degree burns when changing drivers doing live sound, before learning why welding gloves were around (fast lesson learned).

 

[AlexanderB]

In my older bass rig (inefficient 15"s) both the cone and magnet were warm to touch. The dome got approx. to mid 40s (C) and the magnet the same. And then I still had the bass reflex port aiming straight at the magnet (not cone) to improve cooling. I find it very likely that the voice coil would het over 150 degrees (C) during gigs.

 

[ArtechnikA]

...assist would be needed. ... Not sure how this would affect the compliance of the center spider?

For the 'power' I was figuring convection falls into the 'cheap and better than nothing' category, especially if the duct is vertical. We're talking about hundreds of watts of heat although there's an efficiency loss across the thermal interface. If you can cool a 1000W amp with a 'quiet enough' fan surely you can do the same with a cabinet... They only make enough heat for it to matter when they're loud...

I donno how exactly you'd get the heat conducted but so far I've only spent 20 minutes on the problem ;-)

I believe high-performance tweeters are using ferrofluid to get heat from the transducer to the frame for improved cooling. Maybe super-duper copper braid a la grounding strap (or solder-wick on steroids...) would conduct enough heat to the relatively fixed 'radiator' in the heat-exchanger duct without affecting the sound (by conducting it to the structure).

 

[dhsierra1]

It would surprize you just how warm they can get, as stated above upto 95% of the wattage that goes into a speaker from an amp just produces waste heat. I've received second degree burns when changing drivers doing live sound, before learning why welding gloves were around (fast lesson learned).

95% of several hundred watts is just heat generation?! I hadn't considered that before, no wonder the specs change during hard playing.....