Precieved Loudness and LF Demand

[derrickdj1]

LF
damand interested me since I have seen my sub clip light go from blue
to bright red. I was at 120 db at the listening position when this
occured. The red means I have around 5 more db before the amp shuts
down for safety. I knew that bass uses a lot of power but, never
thought of it in the terms that Don H on AVS discussed. This is not new
for some of you but, for beginners like me it is a good read.

Don
H:

Bass
frequencies don't take any more power than any other frequencies at
the same SPL, but we are much less sensitive to bass frequencies so
for a low note to sound/feel as loud as a midrange or high note
(until our hearing drops off) it takes much more power. For example,
to sound as loud as a 1 kHz tone at 70 dB SPL, a 40 Hz tone must be
at 90 dB. That extra 20 dB is 100 times the power! So if your
speakers only need 1 W to produce that 70 dB, 1 kHz tone, then a 40
Hz tone requires 100 W to sound as loud. Wow!

Your speaker have only
so much displacement and dynamic range, and the same goes for the
amplifiers in your AVR (or whatever). If you are sending 100 W out to
create that 40 Hz tone, that your speakers might not even reproduce
well, then that is essentially wasting headroom (dynamic range) in
the amplifier and speakers. Your AVR could be happily putting out a
few Watts with the sub handling the low stuff, or the AVR could be
stressing out trying to put 100+ W to reproduce the low frequencies
plus all the rest. Most "large"
speakers have very high distortion when trying to produce loud LF
sounds -- like 10% to 50% ! You are using all their excursion and
thus getting very nonlinear (distorted).
It is even worse when the main speakers cannot reproduce the low
frequencies; it simply causes large excursions that are not heard, or
the power is converted into heat in the voice coil. Neither is a good
way to spend your power budget.

Again
using 1 W to produce 70 dB at 1 kHz as a reference, 80 Hz requires an
extra 10 dB, or ten times the power. So, if you cross at 80 Hz (ideal
brick-wall filter just for this simple example), your AVR and main
speakers need to deal with about 11 W (it is not linear, but say
roughly 10 W for the 80 Hz signal to sound as loud as the 1 W signal
at 1 kHz and 10 + 1 = 11). Meanwhile your sub is putting out 100 W to
produce a 40 Hz tone at the same loudness, sparing your main speakers
and AVR from having to deal with 100+ W.
http://www.avsforum.com/t/1112470/official-pioneer-mcacc-thread/4170

This is why subwoofers
have their own power amps, and why more power is always better when
it comes to bass. Some subs have as much power as entire AVRs.

[Coytee]

Interesting read!

[Mongo171]

Thanks for posting that. It explains alot of the darkness of needing more power for low freqs. [li]

[derrickdj1]

I was suprised to see just how much power LF damands on the system and also why settting speakers to small help the amp/avr with heavy loads. Trying to get good output at 20 Hz or lower is a real job for any system. What is going on with speakers set to large was also interesting.

[Mongo171]

The question that people always asks is "Why do my efficient Klipsch speakers need so much power?" I know that bass zaps an amps power rather quickly. I couldn't really answer the question in an intelligent way.

I have never run into a problem with the headroom from my ATI amp. Now I know why sub amps go into Kilowatts to keep up with HT systems.

[derrickdj1]

I notice a significant chage in my Yamaha M 80 meter readings running speakers as Large. Understanding the LF demands on the system make me re-think running speakers as Large or using the PLUS SUB option in the avr. I kown there is more speaker cone breakup running things as Large or with very low xo which means more distortion. How audible is the question.

I consider getting the MQ 600 or Sub 1 amp for the new subs since they are more powerful amps, but chose the Dayton amp because of the EQ feature. I was running a frequency sweep on the subs when I hit 120 db. I was somewhere between 20-30 Hz. I quickly backed off because some pictures fell off the wall and did not want the wife to find something broken again, lol. The red clip light is just an indicator letting the user know that the Dayton amp is getting near the limit. I never listen to things at 120 db: I do want to keep my hearing. Damage does occur even with LF sounds that don't appear as loud as HF sounds.

[Mongo171]

The highest measured reading I have gotten from my system is 106db. And stuff was definitely no sitting still in the house.

I think LF hearing damage is due in part to the moving of the air and most LF freqs are inaudible. Combined they put a lot of pressure on the ear drum.

My hearing is already shot at 400Hz due to military aircraft power freqs.

[djk]

OSHA allows 105dBA for one hour.

dBA has a 6dB slope below about 1Khz, so that is about 130dB at 50hz.

Below 50hz they allow up to 140dB for certain types of impulse noise.

A Chevy I owned had a brand new exhaust installed and it sounded very quiet, it also measured 100dB at 30hz at the driver's ears.

[Mongo171]

A stereo I put into my Mustang in the early 90's was measured at 132db. It was 90 watts total.

[derrickdj1]

The highest measured reading I have gotten from my system is 106db. And stuff was definitely no sitting still in the house.

I did have things running on the hot side, the most I have gotten with the subs at autocalibration setting is 110 db in the 70-100 Hz band and 104 in the 20-30 Hz band with the mic 16 ft from the subs which are 1/8 spaced in the corners and nothing running hot. This was at -5 on the avr with no other speakers running. I don' t remember how many db you loose per ft. of distace. I think it is like 1 db/ 3 ft. At 0 i doubt if the spl will change much since these are sealed subs.

[Mongo171]

I have downloaded REW to my laptop. Just need to get the other components and learn how to use them. I'm sure it will dramatically improve the sound of my system and give me spots where I need to setup my sub. Right now it's pretty much just running on factory settings. Haven't even delved into the PEQ on it.

[derrickdj1]

I don't know much about sub EQ but, I guess it is time to learn. I will try some EQ next week since this week is fairly busy. I have the REW downloads but not the mic and computer card. The REW is the best way to go. All I know is if the graph look like you are climbing a moutain,you need EQ, lol.

[garyrc]

How large is your listening room, and how far from the sub is your listening position, and are you sitting in a zone of high or low bass re: room modes?

120 dB is rather easy to create in a smallish room without approaching clipping of most sub amps, if you are sitting in the right place ... with a large or very large room, a lot more is demanded of the amp, and you may get closer to clipping.

At that SPL (100 to 120 dB) the Fletcher-Munsen curves flatten out quite a bit, so the need for more power to the bass is less dramatic, i.e., at low SPLs the bass needs to be turned up (in SPL terms) to sound as loud as the midrange. That is why "loudness" is considered a psychological/perceptual phenomonom, while SPL is a physical one. At very high levels, the bass droop is less, so it doesn't need to be turned up as much relative to the midrange to sound as "loud" as the midrange. At one time, at least (I'm not up on it), the THX people used a reference level of 108 dB with a band limited pink noise signal (probably 500 to 2Khz, but I don't remember for sure). When they ran The Empire Strikes Back through the system set up for 108 dB, the bass produced 110 dB, which took about ~~~70% more power (111 dB or a 3dB difference would have required twice the power over that needed for 108 dB). This is the other part of the picture; bass passages are often recorded at levels higher than the usual "loud" midrange, so they will require more amp power to get there. This is especially true with modern movies.

P.S. Don't run frequency sweeps at 120 dB!! That could cause speaker damage, frying tweeters (I don't know if your sweep extended into the high frequencies) and stressing all other speakers. Your sub might be called on to produce a steady tone at a high SPL in a movie, but even the "infernal bass machine" that crops up in movies is usually a warble and doesn't last long.

[derrickdj1]

Thanks for the advice Garyrc. I have a large open concept home around 6800 3ft. the drop down family room is roughly 20x18 with 10 ft ceilings. I know not to sweep into the HF. A true 120 db sub sweep can't be handel by a lot of subwoofers. I don't listen to things that loud, I do like my hearing, lol. Just testing the limits of a new toy and getting a better understanding of the LF power demands. I think I get a fair amount of room gain and the subs are 1/8 space. They are not boomy though. Normaly I don't run things hot. The point about the short bass spikes in movies is true and peak power is a little more important than RMS power. When I hit the 120 db level the power conditoner watts reading went from 300+ to over 800+ . I never seen it go that high before. Most of the time the watts are in the 300-400+ range. I have all the amps, avr, cable box, BD and TV hooked up to it.

Now I need to learn how to use the Parametric EQ with Pioneers MCACC. It should be a good combination since MCACC filters takes care of 125,185 and 193 Hz with att and Q. I only have to focus on 100 Hz or lower to bump up or take down a peak. I need to look into REW. The output of these subs is amazing considering the size room and open space. I have the subs facing the wall to help load the driver since they are front firing. Beside, I am not one that likes to look at the drivers and prefer grills on the speakers, lol.

[JBryan]

..and lets not forget how folks perceive the bottom end and many actually prefer an EQ'd boost. I have on several occasions, offered or otherwise was roped into helping a friend set up a system. I either used an SPL meter or software and microphone to try and set the speakers flat. In almost every case, the friend would adjust for more gain in the bottom end - sometimes to ridiculous levels. The odd thing is that I have only 1 older friend who prefers a 'tipped up' bottom end but almost every younger audio bud I've come across has in some way boosted the bass up quite a bit... maybe its something physiological or possibly the way music has been mastered in the last 20 years. I do remember using the 'Loudness' button more often when I was young - now, I don't even have the option available. Maybe its just me but lately, when someone asks for my opinion of their sound, I often find myself suggesting the bottom end might be 'toned down' a bit...or a lot.

[Quiet_Hollow]

but almost every younger audio bud I've come across has in some way boosted the bass up quite a bit... maybe its something physiological or possibly the way music has been mastered in the last 20 years. I do remember using the 'Loudness' button more often when I was young - now, I don't even have the option available. Maybe its just me but lately, when someone asks for my opinion of their sound, I often find myself suggesting the bottom end might be 'toned down' a bit...or a lot.

You can blame the "80 Hz crossover" for that. If I had a nickel for everytime I see a non-THX subwoofer low-passed at 80Hz I'd be a millionaire.

The average Joe, for whatever myriad of reasons, is lead to believe that "crossovers" and filters are one of the same...and hence forth set every subwoofer filter they see to the industry standard 80Hz.

The problem is a use of terms. The result is bad sound.

First, some preamble:

"Crossover" is usually associated with a piece of PA gear that filters sound (ie. an Active Crossover) or a piece of home audio gear consisting of a pile of passive components soldered together used to filter sound (ie THE Crossover).

Either piece, active or passive, is not a "crossover", they are filter networks by definition and usually consist of two or more filter types working in tandem, the high-pass and the low-pass.

This is an important distinction because "crossover" is a noun that is being improperly used to describe an action.

Electronic signals cross over one another. Electronic crossover. Driver performance envelopes cross over one another. This is acoustic crossover. Their handshake, is the act of crossing over.

We don't set the crossover. We set a couple of filter points and the result is crossover.

Now for the crux:

Back to the average Joe. He / She reads in certain literature or on the internet that crossovers are typically set to 80Hz and everything will be cool.

Wrong.

80Hz is the filter set point, not where the crossover happens...especially with respect to non-THX compliant gear which constitutes an overwhelming majority of the equipment out there.

Yes, you ideally want the crossover to happen at approximately 80Hz the majority of the time. This is true. [8]

The problem arises because the frequency response of any typical subwoofer, haphazardly placed in the typical room, requires a filter set point much lower than 80Hz to achieve proper acoustic crossover at 80Hz.

Unless they are aware of basic acoustic principals, and have the gear to RTA their system, most people instead take the online advice verbatim (often against the will of their ears), set 80Hz and live with the result.

Which is often a bloated, boomy, over-emphasized hot mess in the bass region. 80-200Hz is situated waaay too hot while everything below 50 is needlessly muted.

The result is even worse with smaller "subs" that have no business being labeled as such. Those "little boxes" tend to waltz right up into the mid-range when set up incorrectly...with little or no capability below 60 Hz even if they weren't.

To the listener over a short period of time, the result quickly becomes the norm and from that point forward any deviation in the bass quality from there is perceived negatively as "thin" or "weak" or "different".

They are lead to believe that because everything is set correctly, as they've been instructed, that that is what "correct" must sound like.

Oddly enough, people will go so far as to claim the above, even though their perception is clearly telling them, it doesn't sound correct or "real" , in staunch denial that it possibly ever could. [:$]

If it doesn't sound real, that's because it's wrong.

In an overwhelming number of cases, the "80Hz" set point is the culprit.

The advice should read:

Set the subwoofer filter to between 40-50 Hz. Check for correct phase with the mains at 80-90Hz. The subwoofer & mains will then be situated to crossover in the recommended 80Hz region.

From that point the average listener's response would be reasonably flat, and sub-bass frequencies would actually be presented at near correct levels without the bass/mid-bass region AFU at the same time. [8]

[derrickdj1]

I understand what you are saying on setting the sub xo and then the avr xo. Most of us have learned to turn the xo on the sub all the way up, disable the xo on the sub or turn it off. How dose that work setting sub 40-50 Hz and avr 80 Hz and this will equal a xo of 80 Hz combined.

[Quiet_Hollow]

It's strictly case by case...but often goes something like this:

The majority of AVR's mistakenly label their unit's main speaker high-pass filter as "Crossover" when it doesn't crossover anything.

It's the combination of that high pass filter and the low pass located in the sub that constitutes the entire network.

To add to the problem, a handful of late model AVRs actually provide their own built-in, adjustable low-pass on the subwoofer channel while the channel is active.

Which is fine for using a featureless power amp to drive the sub... [Y]

But it just adds to the confusion in any other instance because it's way too easy to cascade filters when using the typical powered sub or component subwoofer amplifier (that already has the correct active filter network packaged onboard).[N]

How dose that work setting sub 40-50 Hz and avr 80 Hz and this will equal a xo of 80 Hz combined.

It's in the math. If the response is completely flat to begin with, the filter points are intuitive and make sense. (like those pretty pictures found in wikipedia articles and technical papers that talk about crossovers)

But that's almost never the case with any subwoofer once in-room.[8] You use RTA to sample the wide band response, apply a filter, then take another look.

If you have access to MCACC, Audyssey, REW, or an SMS-1 you can actually see whats going on as you adjust the filter points. With an SPL meter I've also hand-plotted response old-school Ethan Winer style. The results are good enough to show what's taking place.

[derrickdj1]

Thanks for elaborating on what is going on. Crossovers and whatis actually going is confusing when you consider the different brands of avr's and the bass management system. Recently on AVS I read that Pioneer xo setting affects to LFE channel.

[Quiet_Hollow]

Thanks for elaborating on what is going on.

You're welcome.

Crossovers and whatis actually going is confusing when you consider the different brands of avr's and the bass management system.

No doubt. With the complexity of today's AVRs, there's a lot more chaff to wade through to get to the "nuts & bolts" features that truly equate to good sound.

With respect to what I explained in the post above, here's an example below using my sub:

The raw in-room response with my amp's low-pass set as high as it'll go, 200Hz:

Using the amp's low pass filter (12 db / oct) set to 30Hz, it flattens the response and initiates the correct roll-off for crossover near 80Hz. Flat response + perfect bass all in one twist of the knob.[Y]

Subtract 12 dB for every doubling of frequency, and you should be able to pick out right where the filter point is by overlying the two plots.

From there, all I needed to do was set the AVR to high-pass my mains at around 85-90Hz, then use a test tone at that frequency and my SPL meter to set the sub and mains in-phase at that point.

Note: If I had set my subwoofer amp's low-pass filter point to 80 Hz (like most people typically end up doing), the response would've have sounded not much different than the first sample plot. The response would be a big "hump" shape with plenty of action between 60-70 Hz and not so much any where else, with a very steep roll-off above the filter point. A typical subwoofer is going to have this mound-looking response in-room. Not all, but most. Effectively a powered woofer more than a subwoofer, which might sound okay with some types of music, but would be lacking some serious bacon for anything else.....and all because I simply didn't know any better. [:$]

Going from -15 dB at 20 Hz to essentially flat with useable response dipping into the low teens was night & day [6]