• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Consequences of speaker EQing on amp power needed

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
Take also the Neumann KH 80 DSP in account. It's very small but the reviews here and here are very good, at low levels matching or exceeding the Neumann KH 310. Using EQ for loudness correction should be no problem at 50 db level.

yes,I forgot to mention it. However, I would like more 'physical' bass than my current Proac 1sc, and the 80DSP has even smaller drivers.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
IMHO, for 50 dB listening, it is not worth investing in anything if your speakers are already decent.
"Better speakers" could have a larger bandwidth or less distortion, but at that volume you won't hear much low frequencies (because of human ear various thresholds following the frequencies) nor distortion under 0.3 %.

yes, but modern speakers drivers, like the Beriliums ones, bring more detail. Other driver types, like AMT, have as well their own audible features within the same dynamic range.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
And if you are trying to keep the level low in an office a subwoofer is a bad idea... Sound isolation in offices is usually pretty poor, especially at low frequencies.

yes, that's the thing is stopping me.... the safe bet would be a subwoofer with DSP and cut the lows at the neighbours complain's threshold ;-)
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,982
Location
Riverview FL
If you listen at low levels it would be natural to not hear much bass...

50dB is right at the threshold of hearing for lower bass.

You'll hear the harmonics, you might not "hear" the fundamentals.
Your brain might synthesize the bass it can't hear when it can hear the harmonics.

I don't turn up the bass when playing softly.



Audiogram showing the average human threshold for pure tones:


1557352742432.png


Bass Guitar Harmonics example - with a 100Hz fundamental

Divide frequencies by 2 for 50Hz fundamental, by 3 for 33hz fundamental

1557352971641.png
 
Last edited:

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
yes, that's the thing is stopping me.... the safe bet would be a subwoofer with DSP and cut the lows at the neighbours complain's threshold ;-)

You won't even be able to hear most sub frequencies at 55dB.
iu


That lower trace is the threshold of hearing. Anything below about 40Hz/50dB is going to be inaudible, even in silence - which I'm guessing your office probably isn't ;)
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
amazing, you nailed it!. Indeed I can run direct the speakers from the DAC.

However, somehow I thought that more power is good for higher dynamic range.
You need much more power for higher listening levels. For a doubling of the experienced sound level you need approximately 10 times the power. For listening with an average sound level of 83 dB (cinema standard) with corresponding peak levels of 103-108 dB you must drive your speakers with around 50 -100 W.
Infact, I don't understand why headphones amps are so bulky, when so little power is needed.
Not all are bulky. And some have a DAC and other stuff on board.

One problem for sure is that the impedance range of head phones is very big, going from 600 Ohm down to 8 Ohm. The high impedance headphones need high voltage and low current, the 8 Ohms high current and low voltage.

To feed them all the amplifier must be able to deliver both high voltage and high current which means much more power than you ever need. Feeding an 8 Ohm headphone then creates a lot of heat in the final amplifier stage because the output voltage is low and but the output current is high. The remaining power (output current multiplied by the difference between supply voltage and output voltage) must be dissipated and for this you need space for heat sinks and proper dimensioned output devices.

The bulkiest ones though are usually overpriced highend with imperfect performance.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
yes,I forgot to mention it. However, I would like more 'physical' bass than my current Proac 1sc, and the 80DSP has even smaller drivers.
See @andreasmaaan's posting here. To get more physical bass you need to EQ the speaker, a bigger speaker alone does not solve the problem.

And believe me, such a small active speaker can sound amazing. I haven't heard the KH 80 DPS yet, but I own a similar sized pair of Genelec 8020 and nobody believes his eyes when he hears them the first time. And the KH 80 goes deeper still.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
See @andreasmaaan's posting here. To get more physical bass you need to EQ the speaker, a bigger speaker alone does not solve the problem.

And believe me, such a small active speaker can sound amazing. I haven't heard the KH 80 DPS yet, but I own a similar sized pair of Genelec 8020 and nobody believes his eyes when he hears them the first time. And the KH 80 goes deeper still.

Funny, I have some 8020 as well. At low volume levels the passive old Proac 1sC with the bulky class A integrated definitively sounds 'fuller'. Although my Gennies are from 2007 or so, probaby the oldest version.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
guys,

to express better what I mean for the upgrade, I'm ashamed to post here the REW measures for the left and right Proac 1sc+valve amp @20W+SMSL SU-8. Leaving apart the room mode at 89hz and the placement issue of the left speaker at 167h, the half plane transition aroung 1kh is awful. Well, all looks ugly, but sounds ok ;-)

Proac1sC.PNG
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
Funny, I have some 8020 as well. At low volume levels the passive old Proac 1sC with the bulky class A integrated definitively sounds 'fuller'. Although my Gennies are from 2007 or so, probaby the oldest version.
Mine are also the first version, bought in 2008.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
guys,

to express better what I mean for the upgrade, I'm ashamed to post here the REW measures for the left and right Proac 1sc+valve amp @20W+SMSL SU-8. Leaving apart the room mode at 89hz and the placement issue of the left speaker at 167h, the half plane transition aroung 1kh is awful. Well, all looks ugly, but sounds ok ;-)
Nothing to be ashamed of. I would try to get rid of the 167 Hz bump of the left speaker. This could be a reflection from a wall or the table it sits on.

The 90 Hz bump may actually help since it works like a loudness correction, making the sound fuller. Would be interesting to see the graph of your gennies. Ignoring this bump there is useful response down to 40 Hz but I doubt that you are able to here 40 Hz at 50 db sound level without loudness correction.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
Nothing to be ashamed of. I would try to get rid of the 167 Hz bump of the left speaker. This could be a reflection from a wall or the table it sits on.

The 90 Hz bump may actually help since it works like a loudness correction, making the sound fuller. Would be interesting to see the graph of your gennies. Ignoring this bump there is useful response down to 40 Hz but I doubt that you are able to here 40 Hz at 50 db sound level without loudness correction.

yes, 167 is reflection (near a corner, only some rockwool can help), and when I remove 90hz with a filter from REW+rePhase I have to manually EQ in foobar to recover the 'missing' bas.

Unfortunately, I have the Gennies at home. I plan in the future put them together the Proac and measure, but moving speakers of locations is a pita.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
yes, 167 is reflection (near a corner, only some rockwool can help),
Lots of probably :eek:
and when I remove 90hz with a filter from REW+rePhase I have to manually EQ in foobar to recover the 'missing' bas.
This is what I expected. Anyway if you want to use loudness correction than you have to remove the bump, otherwise the correction makes the bump even stronger.
 

Cortes

Active Member
Joined
Apr 13, 2019
Messages
291
Likes
354
Lots of probably :eek:

This is what I expected. Anyway if you want to use loudness correction than you have to remove the bump, otherwise the correction makes the bump even stronger.

yes, that's what I do. I explain my progresses in a REW thread. However, I'm not sure whether I like more all corrected+low boost with repect to the uncorrected sound.
 

Juicehifi

Member
Audio Company
Joined
Mar 20, 2019
Messages
17
Likes
26
Hi,

Let's say I have a braodband speaker with a sensitivity of 90dB/W/m.
If I apply some EQ, for instance a room curve or any other bass boost (like +10dB at 40Hz descendint to 0dB at 20kHz):
- does it "change" the speaker sensitivity?
- how much SPL will I have at 1 meter with 1 Watt?
- do I need a more powerful amp?

Cheers.

The music signal is complex, and the effect of EQ is spread out in time and blended with the music signal in a way that makes the impact less intuitive. And a lot of folks give an intuitive answer to these questions.

I ran some simulations a few years ago, with fullrange FIR filter that cut all tops and boosted depressions with max 6 dB. The simulation was done with a random signal (noise). The result:

Overall sensitivity (in the digital domain) dropped by around 6 dB in the case I was using. The sensitivity will drop by the max boosting you apply. You can put it back by digital amplification, but then you will experience digital clipping from time to time - at least on compressed music.
Output SPL per watt input was the same for pure frequency correction, but dropped 1 dB for frequency plus time domain correction. In other words: The system will be practically as efficient as before correction.

To answer your questions:

1) No, but the gain in (the digital part of) your system will be reduced accordingly.
2) As much as before if you have sufficient gain headroom. If you don't have 10dB to spare on your volume control, you will not be able to play as loud as you wish.
3) No, it will basically be the same load, except for the very rare situation where you play a pieceof music that has very high spl content on the exact frequency you have boosted.

The only thing you need to consider when you EQ boost a frequency region is how much the subject speaker drivers can handle. There will be more driver excursion on the amplified frequency - 6dB boost doubles the excursion. If this happens in the low end of the speaker it will reduce how loud you can play before the speakers start to sound distorted. A midrange driver can sometimes be a bottleneck too, but usually this will only be an issue for the low frequency driver. In other words: Be mindful about how much you extend the low frequency range if you intend to play loud.
 
OP
daftcombo

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
I have a question about distortion.

If I play music loud enough so that a speaker starts to distort at one frequency (let's say 60 Hz), do the other frequencies played at the same time distort as well?
Isn't that more likely to happen at the impedance peaks?
 

Juicehifi

Member
Audio Company
Joined
Mar 20, 2019
Messages
17
Likes
26
Your second question is a more general one. The impedance peak is fairly easy for the amp, but the slopes on both sides are the more demanding ones. How much this matters depends on the amp. E.g. class D amps can handle reactive loads quite easily.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,982
Location
Riverview FL
If I play music loud enough so that a speaker starts to distort at one frequency (let's say 60 Hz), do the other frequencies played at the same time distort as well?

You decide.

Here's a a little test of my distortion boxes, the JBL LSR 308.

Audio is the in-room (if I got it right)

Same source - Audacity
Different in-room SPL

Lower SPL, low distortion.



Higher SPL, high distortion

 
Last edited:
OP
daftcombo

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
You decide.

Here's a a little test of my distortion boxes, the JBL LSR 308.

Audio is the in-room (if I got it right)

Same source - Audacity
Different in-room SPL

Lower SPL, low distortion.



Higher SPL, high distortion

Thank you but I don't understand what you did. Where is the signal? What are the red curve? black curve?
 
Top Bottom