Hallo!

What do you think about equing studio monitors to extend the linearity
in the low end? We are talking about JBL LSR32. Their frequency
response starts to decrease below 80Hz.

For testing I inserted a 31Band digital Behringer EQ from my
live-emergency(!)-rack and I boosted the 50Hz ca. 5dB 30Hz 10dB.

Since I have no regular measuring equipment, I used a C451 and Wavelab
Analyser and pink noise to "measure" the linearity.
Sounds good to me.

1. What do you think about EQing the low end?
2. What EQ would you suggest? (if yes)
3. How to meassure the frequency response with normal studio equipment?

Greetings Thomas

Thomas Thiele wrote:

What do you think about equing studio monitors to extend the linearity
in the low end? We are talking about JBL LSR32. Their frequency
response starts to decrease below 80Hz.

It does? What are you putting through it? I suspect that the problem
isn't with the speakers as much as it is your room. If you have the
opposite problem, a build-up of bass due to the speaker being too close
to a wall or corner, you can successfully compensate for that by
cutting the low end with an equalizer. But if your problem is that you
want to hear more bass, adding it with EQ might work in one position,
but it will be different in another listening position. Also, you'll be
pumping more power into the speaker at low frequencies and you power
amp or speaker may not be able to take it at the volume you want to
listen. People who want a lot of low end usually want a LOT of low end.

Sounds good to me.

Did you walk around the room listening to music when you had that boost
set?

you suggest? (if yes)

3. How to meassure the frequency response with normal studio equipment?

You can't, unless you consider listening to be measurement and ears to
be normal studio equipment. Cheap and simple spectrum analyzers are
uselessly inadequate for making room measurements at low frequencies.

Thomas Thiele wrote:

What do you think about equing studio monitors to extend the linearity
in the low end? We are talking about JBL LSR32. Their frequency
response starts to decrease below 80Hz.

For testing I inserted a 31Band digital Behringer EQ from my
live-emergency(!)-rack and I boosted the 50Hz ca. 5dB 30Hz 10dB.

Since I have no regular measuring equipment, I used a C451 and Wavelab
Analyser and pink noise to "measure" the linearity.
Sounds good to me.

1. What do you think about EQing the low end?

I think it really doesn't fix the problem. If the issue is the monitor
and not the room, you can use EQ to extend the low frequency corner, BUT
if you do this, you get very little headroom at low frequencies. The
speaker goes into clipping (full driver excursion) very easily.

2. What EQ would you suggest? (if yes)

If you absolutely HAVE to do this, I'd suggest doing it with lumped
sum filters added to the crossover of the speaker.

3. How to meassure the frequency response with normal studio equipment?

The easiest thing to do is to get a $20 SPL meter. Radio Shack sells a
usable cheap one in the US. Now, you can play tones through the speaker
from a signal generator or test disc and get an SPL level in the room.

If you do this, first of all you need to know that the low frequency
response of the meter drops off, so you need to get the C-weighted curve
off the net somewhere, and subtract the curve you draw of the low end
from the response curve of the meter to get the actual response.

Secondly you need to know that, because of room effects, everything will
change when you move a couple feet.

My suggestion is that if you change the room so that the room response
is rising at about the same point where the speaker response is dropping
off, you'll get better results for less money.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Thomas,

What do you think about equing studio monitors

You got the right answer (don't do it), and I'll add a little.

For testing I inserted a 31Band digital Behringer EQ

The peaks and nulls that occur at low frequencies in small rooms are far
narrower than 1/3 octave. So even if EQ was useful, which it's not, you'd
need to use a parametric EQ and adjust it very carefully.

Since I have no regular measuring equipment, I used a C451 and Wavelab
Analyser and pink noise to "measure" the linearity.

The C-451 can be used, but only if you have an omni capsule. I have a 451
with a CK22 capsule and it works very well for room analysis. But you also
need a source of tones in very fine increments, like 1 Hz. Or better, use
software like ETF he

www.acoustisoft.com

--Ethan

Hallo Mike and all others!

Look at http://www.jblpro.com/pub/manuals/lsrman.pdf. You'll see on
page 7 of the manual that the response curve drops off at around 80Hz.
At 50Hz -5dB and at 30Hz -12dB.
My intention was not to fix room problems. Perhabs there are some
little, but not that much. I never measured it, but I think I know how
they would sound.
Btw. a little pictu
http://www.krachwerk.de/ASGStudioreport/studiolinks.jpg
(the stands are some of those temporary solutions that remains
longer...)

I'm just wondering if you can/should extend the linearity in the low
end to say 50Hz by using EQ.
Some ascii-graphic:
___________________
/ /
/ /
/ /

^with EQ
^without EQ

Thomas Thiele wrote:

Look at http://www.jblpro.com/pub/manuals/lsrman.pdf. You'll see on
page 7 of the manual that the response curve drops off at around 80Hz.
At 50Hz -5dB and at 30Hz -12dB.

That's in an anechoic chamber. When you put it in the room, it will
be different because there will be a resonant space between the wall
and the speakers. If you pull the speakers farther into the room,
the resonance gets wider and lower frequency, if you put them in closer
toward the wall it gets narrower and higher. You move the speakers
around until the response is flat, using the room to emphasize the low
end where the speakers drop off.

My intention was not to fix room problems. Perhabs there are some
little, but not that much. I never measured it, but I think I know how
they would sound.

If you had speakers that measured flat in an anechoic chamber, they
would sound very boomy in the room.

If you have never measured the room, do so before thinking about
anything else.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Thomas Thiele wrote:

Look at http://www.jblpro.com/pub/manuals/lsrman.pdf. You'll see on
page 7 of the manual that the response curve drops off at around 80Hz.
At 50Hz -5dB and at 30Hz -12dB.

That's not too bad for a small monitor, actually.

I'm just wondering if you can/should extend the linearity in the low
end to say 50Hz by using EQ.

As Scott, and I think I pointed out, in theory you could do that, but
in practice it might not work out so well. You'd be putting more power
into the speaker at low frequencies which not only takes away power
from the more audible portion of the spectrum (assuming you don't
upgrade to a higher power amplifier, if you have the unpowered version)
but you could damage the speaker by putting too much current into the
voice coil in order to move enough air at low frequencies to keep you
happy.

But like anything else in this business, it doesn't hurt to try, as
long as you understand what the problems might be and be on the lookout
(or listen-out) for them. When you hear farts instead of nice round
bass, back it off.

Have you considered a subwoofer? That's probably the best solution, and
80 Hz is a good frequency for a sub to start working.

"Thomas Thiele" wrote in message
ps.com
Hallo!

What do you think about equing studio monitors to extend
the linearity in the low end? We are talking about JBL
LSR32. Their frequency response starts to decrease below
80Hz.

For testing I inserted a 31Band digital Behringer EQ from
my live-emergency(!)-rack and I boosted the 50Hz ca. 5dB
30Hz 10dB.

Since I have no regular measuring equipment, I used a
C451 and Wavelab Analyser and pink noise to "measure" the
linearity.
Sounds good to me.

1. What do you think about EQing the low end?

Eqing monitors is like tasting wine with a sugar cube in
your mouth.

"Arny Krueger" wrote in message
...
"Thomas Thiele" wrote in message
ps.com
Hallo!

What do you think about equing studio monitors to extend
the linearity in the low end? We are talking about JBL
LSR32. Their frequency response starts to decrease below
80Hz.

For testing I inserted a 31Band digital Behringer EQ from
my live-emergency(!)-rack and I boosted the 50Hz ca. 5dB
30Hz 10dB.

Since I have no regular measuring equipment, I used a
C451 and Wavelab Analyser and pink noise to "measure" the
linearity.
Sounds good to me.

1. What do you think about EQing the low end?

Eqing monitors is like tasting wine with a sugar cube in your mouth.

THAT'S what I've been doing wrong! Damn.

Steve King

Thomas Thiele wrote:

I'm just wondering if you can/should extend the linearity in the low
end to say 50Hz by using EQ.

Yes, but you may well go beyond the ability of the power
amplifier to deliver signal at the boosted LF.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Thomas Thiele wrote:

I'm just wondering if you can/should extend the linearity in the low
end to say 50Hz by using EQ.

No you shouldn't because outside the speaker's desgined bandwidth
performance may (probably) not be linear. You want bottom you need a
sub.

--
ha

Arny Krueger wrote:

Eqing monitors is like tasting wine with a sugar cube in
your mouth.

If the on-axis anechoic response is not flat, why, other
than driving and coil dissipation considerations, shouldn't
they be equalized? Oh, you mean with something like a
graphic equalizer I'll bet. In that case I agree. With the
more sophisticated inverse convolution filtering approach it
can only improve matters.

Then, at least its flat and time accurate before the room
does its dirty and for near field monitoring that dirty
isn't as much as for a leisure listening setup.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

"Bob Cain" wrote in message


Arny Krueger wrote:

Eqing monitors is like tasting wine with a sugar cube in
your mouth.

If the on-axis anechoic response is not flat, why, other
than driving and coil dissipation considerations,
shouldn't they be equalized?

The whole problem with equalizing is about trying to figure
out what sort of response we want to have. AFAIK, there is
no generally agreed-upon answer.

The FR of a speaker isn't a curve, its a large family of
curves, each representing the speaker's response in a
different direction. The ear hears an weighted integration
of these curves.

I doubt that many speakers are designed to have flat
response in an anechoic chamber, at least not without
references to a number of other kinds of frequency response
tests.

Oh, you mean with something like a graphic equalizer I'll
bet.

No, I'm thinking about the best equalizer that we could
think of.

In that case I agree.
With the more sophisticated inverse convolution filtering
approach it can only improve matters.

That would be more like the best equalizer that we could
think of. However, we have no electrical means for
equalizing the response of a speaker in different directions
at the same time.

Then, at least its flat and time accurate before the room
does its dirty and for near field monitoring that dirty
isn't as much as for a leisure listening setup.

Mt attitude is that a speaker is a box that does something
that we need it to do. In monitoring it needs to have broad
and smooth response so that you can hear all the different
audible parts of the music. Every speaker that has broad and
smooth response has response that differs from other
speakers in the details. Once you learn what those details
are, the most important thing is that the speaker's response
be stable.

I'll admit it - I've tried to integrate a subwoofer with my
monitors. In the end, they have been mostly used without the
subwoofer. There's two practical problems with setting up a
subwoofer - one is getting balanced with the mains and the
other is keeping it that way. That became a hassle and I
just wanted to make music.

Bob Cain wrote:
Arny Krueger wrote:

Eqing monitors is like tasting wine with a sugar cube in
your mouth.

If the on-axis anechoic response is not flat, why, other
than driving and coil dissipation considerations, shouldn't
they be equalized? Oh, you mean with something like a
graphic equalizer I'll bet. In that case I agree. With the
more sophisticated inverse convolution filtering approach it
can only improve matters.

If on-axis anechoic response is not flat on the low end, placement
will often compensate for this.

If on-axis anechoic response is not flat in the midrange, the monitors
should be discarded and replaced.

Occasionally equalization might help things a a little, but usually if
this is the case the manufacturers have already designed it into the
crossover.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Arny Krueger wrote:
"Bob Cain" wrote in message



Arny Krueger wrote:


Eqing monitors is like tasting wine with a sugar cube in
your mouth.


If the on-axis anechoic response is not flat, why, other
than driving and coil dissipation considerations,
shouldn't they be equalized?


The whole problem with equalizing is about trying to figure
out what sort of response we want to have. AFAIK, there is
no generally agreed-upon answer.

Why would anything other than flat be desirable for a mixing
monitor?

The FR of a speaker isn't a curve, its a large family of
curves, each representing the speaker's response in a
different direction. The ear hears an weighted integration
of these curves.

True, but for a nearfield monitoring situation the direct
sound from on-axis should much predominate. Enough so, I
would think, that using filters derived from inverting a
pair of in-situ measurements taken at the apex of the usual
nearfield triangle should work very well.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

On Wed, 31 Aug 2005 14:49:51 -0700, Bob Cain
wrote:

Why would anything other than flat be desirable for a mixing
monitor?

I think the term "flat" might be the real sticking point.
IOW, flat for first arrival sound or flat for summed
room response? Or some compromise? Flat for one listener
or flat for an averaged space for x listeners. like that.

A couple years ago, I built a prototype crossover and
bass equalizer for some speakers with essentially *no*
mechanical limitations. Four cylinders with twin 18"
cost-no-object woofers each, intended for the passband
to be below resonance.

The EQ (based on Linkwitz's early Speaker Builder article,
*essential reading* to anyone interested in the topic,
was taste-tested with absolutely no blinding, and yet...

I and several kindred spirits had fairly strong
preferences for a final system Q of 0.5 and F-sub-c
of about 8Hz.

FWIW, maybe not much,

Chris Hornbeck