Value of subwoofer amplifier power ratings?

Mr Wolf

Well-known Member
It became very clear to me recently that not all subs are created equal and that specs can be very misleading. To my surprise, using REW and a calibrated UMIK-1 microphone I recently measured a 500W 12" sealed sub as having significantly more in-room SPL output than a 750W 15" ported sub at all frequencies.

Obviously, in addition to amplifier power, a sub's actual SPL output is also down to the efficiency of its driver/cabinet design which is why I'm primarily interested in measured SPL output (CEA-2010 style) but a lot of forum members go around quoting sub power ratings like they really do mean a lot when comparing subs when in practice they probably do not.

Let's consider AVR power ratings for a minute. To make them comparable, all AVR power output wattages have to be expressed in terms of a given frequency (or range), RMS/peak, impedance load and distortion level e.g. 140W RMS (20-20,000Hz, 8 Ohms) @ 0.05% THD. Tweaking these assumptions can result in well over a 100% variance in an AVR's power output rating.

By comparison, while they may state whether it's RMS or Peak, subwoofer wattage specs leave out any mention of frequency, impedance or distortion. This therefore surely makes it impossible to compare sub wattage specs so that any talk along the lines of "sub X has 300W more power than sub Y" is totally meaningless. In practice, a "300W sub" from one brand might actually have a more powerful amplifier than a "500W sub" from another brand.

Does anyone know how any sub manufacturer actually measures this figure? I would expect that some are far more conservative than others.

Do you agree?
 

MemX

Well-known Member
It's all about the displacement with subs - more air moved = more loudness :D

Power is just one of the three key ingredients to Hoffmans's Iron Law.

Low Power
Small Box Size
High Output at the low end

Pick two only ;)
 

Mr Wolf

Well-known Member
It's all about the displacement with subs - more air moved = more loudness :D

Power is just one of the three key ingredients to Hoffmans's Iron Law.

Low Power
Small Box Size
High Output at the low end

Pick two only ;)
I understand that, it's very much like the Speed, Armour, Firepower design triangle of Main Battle Tanks - you pick two.

While interesting, it's a little bit off point. What I'm challenging is how that amplifier power ingredient in the subwoofer recipe is being measured, not how a sub is made to play loud and/or low.
 

MemX

Well-known Member
I understand that, it's very much like the Speed, Armour, Firepower design triangle of Main Battle Tanks - you pick two.

While interesting, it's a little bit off point. What I'm challenging is how that amplifier power ingredient in the subwoofer recipe is being measured, not how a sub is made to play loud and/or low.
Ha, the tanks thing had never occurred to me but I totally see the comparison! :D

I take the point I may be a little off point ;) but I think we are getting at the same point - stated power is a fairly irrelevant metric / piece of information on its own; it needs to be in the context of other things :) (and in subwoofers it is arguably about displacement if one is looking to go lowwwww :cool: ) so I'm not sure it really matters how it is measured - it probably shouldn't even matter because the output, not the input, is what counts!!
 

Dobbyisfree

Well-known Member
I have thought this as well Wolfy! I don't believe there is any kind of consistency in the sub quoted power measurements.
I don't even think there is consistency amongst the same supplier. E.g. I have taken out the plate amp of a (called in its day) 60w REL Strata and had it next to the plate amp from a (again quoted) 100w REL Quake.
They are the same "class", i.e. neither is class D etc.
The Strata's plate is nearly double the weight, most components look bigger... do we really think that it has less power than the Quake?
As you point out, it depends on so many factors.
 

Conrad

Moderator
As others have said, it's a parcel. You have to look at the DSP as well.

Take for example the new Kef sub. Stats suggest 11Hz AND 105dB. That's an amazing spec. But that's not 11Hz AT 105dB. And in fact, when you measure it, 11Hz stops rising at about 85dB. It's the same with the B&W subs.

I don't know how that translates into CEA stats as you'll never get the harmonics no matter how loud you turn it as the volume for that frequency doesn't rise.

Having a 2000w amplifier driving a 10" driver is fine, and you'll definitely get it moving, but at what cost? (well, super high distortion). The caveat here is to over-engineer the excursion like JL and others do. It's why a JL 12" sub, or the Velodyne DD12+ can hit low teens at high SPL with low distortion.

You can give up some cone area if you can move that cone far enough.

And just to add to the triangles, in software development (and other businesses) it's Speed, Cost, Quality.
 

Mr Wolf

Well-known Member
Ha, the tanks thing had never occurred to me but I totally see the comparison! :D

I take the point I may be a little off point ;) but I think we are getting at the same point - stated power is a fairly irrelevant metric / piece of information on its own; it needs to be in the context of other things :) (and in subwoofers it is arguably about displacement if one is looking to go lowwwww :cool: ) so I'm not sure it really matters how it is measured - it probably shouldn't even matter because the output, not the input, is what counts!!
Well yes, I guess the bigger picture point here is the strength of the relationship between subwoofer specs and performance, amplifier power being just one of those specs.

Unfortunately few manufacturers publish any SPL output figures and we have to rely on third party testing (e.g. Audioholics) to get anything reliable.

The fact remains though that specs sell big-time in this industry. And sub power specs are bandied around on this forum (even by some experienced hobbyists) like they're something that's comparable when I would suggest they are not. When we don't know a sub's output, we try to deduce it from the inputs (i.e. specs) but for subs (even if we understood the science) the specs we are given are too simplistic to work with do that reliably.

Interestingly, my ported 750W (1500W peak) 15" Velodyne sub is convincingly outgunned by a sealed SVS SB-2000 500W (1100W peak) despite having a more powerful amp (???), a larger driver, a larger cabinet and a port. The SVS has visibly more excursion though and I presume it's in this and the efficiency of that driver that account for the performance gap here.
 

vm1451

Active Member
Could you explain "excursion" please? Desperately trying to understand subs and their integration!
 

Conrad

Moderator
It’s how far back and forward a driver can move. There are more specific terms like xMax and xMech which describe specific portions of the movement.

XMax is the total distance a driver can move in tolerance (I think). XMech is the physical limit a driver can move to (when it starts to hit other parts of the driver). Excursion is a generic term for the back and forth movement.
 

vm1451

Active Member
So why do subs seem to have excessive amounts of power? By that I mean my AV amplifier has 170w, (I think!), of power across at least five channels and I have never run it anywhere near what I believe is called "reference" level, i.e. volume control at 0db, (please correct me if I'm wrong, I'm not proud!!!). That said , why would I need/want a sub with 1500w of power for example?
 

Conrad

Moderator
Because for every octave you go lower (halving of frequency) you need twice the power to generate the same SPL.

So 20Hz at 100dB takes twice the power of 40Hz at 100dB. 10Hz takes double 20Hz (so 4 times 40Hz). Plus, there's a loudness curve that means you actually need more output at 20Hz than you do at 40Hz to hear the perceived loudness.

The same goes up as well. So 80Hz takes half the power of 40Hz, 160 half of 80, etc. So you can imagine that from 100Hz up to 20,000Hz really doesn't need much power at all, depending on distance and speaker sensitivity.
 

vm1451

Active Member
Will running a sub amp that has say 100w of power at 90% or higher tend to introduce distortion?
 

David Alexander

Active Member
I may get my REL HT1003 down again and measure some REW responses vs the XTZ 12.17 Edge (ported) I seem to rmember the 12.17 had was louder overall whith the same input test tone but that the REL certainly went much lower and louder than the 24HZ (-6db) that the handbook stated. But then you factor in the room gain to that and it's going to be boosting the sealed subs response from around 40Hz (my longest room dimension is just short of 16').

Although room gain wouldn't explain why the smaller/less powerful sealed sub would beat the larger ported sub across the frquency range like Mr. Wolf is measuring.

Do you have set both subs to zero gain as well to equall things out? Or is gain intrinsically part of a subs performance? For example if I have a gain setting on the bigger sub at x to give a power output which it (or I) can handle but to acheive the same output the smaller sub has to have its gain upped to such a level that the output becomes a distorted mess.
 

Mr Wolf

Well-known Member
So why do subs seem to have excessive amounts of power? By that I mean my AV amplifier has 170w, (I think!), of power across at least five channels and I have never run it anywhere near what I believe is called "reference" level, i.e. volume control at 0db, (please correct me if I'm wrong, I'm not proud!!!). That said , why would I need/want a sub with 1500w of power for example?
Further to what @Conrad said about extra power for lower frequencies, movie soundtracks require the LFE subwoofer channel to play up to 10dB higher than any other speaker channel. All other things being equal, 10dB louder requires exactly 10x the power.

Add in the redirected bass under (say) 80Hz from any speakers set to small and you have an even greater power requirement for the sub. This is why subwoofers really have to be quite powerful even if you are not listening at reference levels.
 

Dobbyisfree

Well-known Member
But back to the original question, based on the comments do we think it actually matters what the power figure is? As discussed, there are so many aspects of a subwoofer design that affect its output at a given frequency that surely frequency response curves, each at a target db level would be (for example) a better measure and comparison tool to show its ability?
Some great data is available on the internet but only for a small number of products. Which sadly leaves people not really knowing what they are buying. At least forums like AVF are there so that people can share their own data or subjective opinions on the products!
 

vm1451

Active Member
I do get confused with some of the terminology used. Room gain for example. A room does not have an amplifier, so how can it have gain/volume control? My understanding is that a room can have loss through shape, furnishings etc which may "appear" to have an effect of increasing volume in certain areas. However the apparent increase is actually the volume you should be hearing anyway!
 

Mr Wolf

Well-known Member
I do get confused with some of the terminology used. Room gain for example. A room does not have an amplifier, so how can it have gain/volume control? My understanding is that a room can have loss through shape, furnishings etc which may "appear" to have an effect of increasing volume in certain areas. However the apparent increase is actually the volume you should be hearing anyway!
 

vm1451

Active Member
So I'm a cynic and would say that SVS sell subs so have an axe to grind! Arn't they essentially saying the same as I have though? i.e. there is no "gain", but by placing any speaker in any given place in any room, you can hear a difference and that difference is created by room loss, not "gain".

However if I am wrong, (more than possible!), why is the word "gain" used to indicate volume control on a sub, (why not just call it a volume control?), and then use the same word to indicate something else when used in conjunction with the word "room"?

 

Conrad

Moderator
My understanding is that it's not really "room" gain, it's "boundary" gain - of course those boundaries form the room.

If put the sub against the wall the sound waves come out of the sub and bounce off the wall and meet up with the waves moving away from the wall. As the waves are so long they're basically still in phase, so you get a positive gain. I believe you get ~3dB by putting a sub against a wall. If you put it in the corner (so against two walls) you get double the gain.
 

Conrad

Moderator
However if I am wrong, (more than possible!), why is the word "gain" used to indicate volume control on a sub, (why not just call it a volume control?), and then use the same word to indicate something else when used in conjunction with the word "room"?
It's not a volume control on the sub, it's a gain control.
It's designed so that you can gain match the sub amp and the amp driving your main speakers. For a given signal the sub should play a tone at the same level as the speakers do. That level will depend on the gain of your AVR amp/power amp, sensitivity of your speakers, listening distance, and room influences.

Volume is managed by the levels and trims in your AVR or processor.

 

MemX

Well-known Member
It’s how far back and forward a driver can move. There are more specific terms like xMax and xMech which describe specific portions of the movement.

XMax is the total distance a driver can move in tolerance (I think). XMech is the physical limit a driver can move to (when it starts to hit other parts of the driver). Excursion is a generic term for the back and forth movement.
What he said re: excursion ^^^ :)


'Excursion' multiplied by 'surface area of the cone' = displacement (i.e. how much air it 'displaces' / moves)

e.g. a driver of 100 square inches surface area with 1" excursion in each direction = (2x1)x100 = 200 cubic inches of displacement.

Same excursion but larger cone = larger displacement.

Multiples of a given sub = larger displacement.

This is why bass speakers are usually large in diameter and often come in multiples - they move more air for a given excursion.

More air moved = more loudness :D


For example - 18" Stereo Integrity HT18 V2 drivers 600wRMS

23mm Xmax (specifications)
117841 sq mm
Displacement = 23 x 117841 = 2710343 cubic mm
1 cu cm = 10x10x10mm = 1000 cu mm
2710343 / 1000 = 2710 cu cm
1 litre = 1000 cu cm
2710 cu cm = 2.71 litres of displacement per driver.

4 of those drivers = 10.84 litres of displacement at Xmax :D
 

vm1451

Active Member
It's not a volume control on the sub, it's a gain control.
It's designed so that you can gain match the sub amp and the amp driving your main speakers. For a given signal the sub should play a tone at the same level as the speakers do. That level will depend on the gain of your AVR amp/power amp, sensitivity of your speakers, listening distance, and room influences.

Volume is managed by the levels and trims in your AVR or processor.

So should this be used while the test tone is being sent from the AVR? Or should the trim settings be used within the AVR to set the volume levels and then the gain to fine tune?
 

Conrad

Moderator
Set the trim to the same as a main channel and then adjust the gain so it matches that main would be a good start.

You want a balance between enough output as desired in room but you also want a negative trim level in the AVR to avoid clipping. Sometimes subs can hum or buzz when you turn the gain up too high so you want to balance that as well.

It's just like if you had two power amps with different gains, you'd adjust the trims in the processor to take account of that, the only difference is that you have a control for it on the sub. If you didn't you just adjust the trim. Do that, but move the gain control on the sub if you're getting to levels that leave you no room to move (usually -12 is bad, and something like -3 gives you nowhere to go up). -6 to -11 is about the right spot, and then adjust the gain on the sub accordingly.
 

Conrad

Moderator
What he said re: excursion ^^^ :)


'Excursion' multiplied by 'surface area of the cone' = displacement (i.e. how much air it 'displaces' / moves)

e.g. a driver of 100 square inches surface area with 1" excursion in each direction = (2x1)x100 = 200 cubic inches of displacement.

Same excursion but larger cone = larger displacement.

Multiples of a given sub = larger displacement.

This is why bass speakers are usually large in diameter and often come in multiples - they move more air for a given excursion.

More air moved = more loudness :D


For example - 18" Stereo Integrity HT18 V2 drivers 600wRMS

23mm Xmax (specifications)
117841 sq mm
Displacement = 23 x 117841 = 2710343 cubic mm
1 cu cm = 10x10x10mm = 1000 cu mm
2710343 / 1000 = 2710 cu cm
1 litre = 1000 cu cm
2710 cu cm = 2.71 litres of displacement per driver.

4 of those drivers = 10.84 litres of displacement at Xmax :D
Nicely described.

However, excursion only helps with output, not extension. Only driver size does that, and I don't know why.

If you have a 12" sub in a room and it's capable of outputting down to, say, 20Hz. Adding 10 more 12" drivers will get you loads more output, but won't get you any more extension. With more output you can EQ hard and use the increased low level output to generate extension, but a larger driver is just capable of moving more air at the slow speeds required for low Hz.

To hit 7Hz for example the driver needs to cycle 7 times a second, that's slow enough to be easily visible. If a 12" driver is moving that slowly it's barely moving any air. If an 18" driver does it, it gets windy!

I'm starting to get the limits of my knowledge here so if anyone can clarify or correct it would be really useful.
 

The latest video from AVForums

Podcast: Anthem AVM 70 Review, EQing Your Home Cinema Sound System and More...
Subscribe to our YouTube channel

Latest News

Sky bringing Apple TV+ to Sky Glass and Sky Q customers
  • By Andy Bassett
  • Published
AVForums Podcast: 27th October 2021
  • By Phil Hinton
  • Published
Linn unveils latest Klimax LP12 flagship turntable
  • By Andy Bassett
  • Published
Samsung Tizen plans include cloud gaming for smart TVs
  • By Andy Bassett
  • Published
Freeview Play racks up 10 million users
  • By Andy Bassett
  • Published

Full fat HDMI teeshirts

Support AVForums with Patreon

Top Bottom