[Jim]

Sorry if this is a repeat question, but I can't seem to find what I'm
looking for. I need a cheap, easy to use, but effective software to
test some speakers I've built. I need to test the frequency response,
the cross over, distortion, phase. . . that sort of thing.

Can anyone recommend some software?

Better yet. . . if there is a member in the San Francisco Bay area who
is willing to help me test the speakers I could use the help.

Thanks, JH

[James Lehman]

"Jim" wrote in message
oups.com...
Sorry if this is a repeat question, but I can't seem to find what I'm
looking for. I need a cheap, easy to use, but effective software to
test some speakers I've built. I need to test the frequency response,
the cross over, distortion, phase. . . that sort of thing.

Can anyone recommend some software?

Better yet. . . if there is a member in the San Francisco Bay area who
is willing to help me test the speakers I could use the help.

Thanks, JH



The most important testing that you can do at home is done before you build
the speakers. You can find the T/S parameters of the woofers and make the
right box for them.

I wrote software that uses a sound card to measure frequency vs. voltage.
With this you can also read frequency vs. impedance. You can measure all of
the T/S parameters and also the individual and summed outputs of the
crossovers. There are a number of other doohickeys out there that can do the
same thing.

http://www.akrobiz.com/speakers/metal-graphs.html

None of this will tell you how the speakers will actually sound. That is a
matter of choosing the right combination of drivers and putting them
together into a good design.

James. )

[Jim]

Thanks for the information.

James Lehman wrote:
"Jim" wrote in message
oups.com...
Sorry if this is a repeat question, but I can't seem to find what I'm
looking for. I need a cheap, easy to use, but effective software to
test some speakers I've built. I need to test the frequency response,
the cross over, distortion, phase. . . that sort of thing.

Can anyone recommend some software?

Better yet. . . if there is a member in the San Francisco Bay area who
is willing to help me test the speakers I could use the help.

Thanks, JH



The most important testing that you can do at home is done before you build
the speakers. You can find the T/S parameters of the woofers and make the
right box for them.

I wrote software that uses a sound card to measure frequency vs. voltage.
With this you can also read frequency vs. impedance. You can measure all of
the T/S parameters and also the individual and summed outputs of the
crossovers. There are a number of other doohickeys out there that can do the
same thing.

http://www.akrobiz.com/speakers/metal-graphs.html

None of this will tell you how the speakers will actually sound. That is a
matter of choosing the right combination of drivers and putting them
together into a good design.

James. )

[TimPerry]

James Lehman wrote:
"Jim" wrote in message
oups.com...
Sorry if this is a repeat question, but I can't seem to find what
I'm looking for. I need a cheap, easy to use, but effective
software to test some speakers I've built. I need to test the
frequency response, the cross over, distortion, phase. . . that
sort of thing.

Can anyone recommend some software?

Better yet. . . if there is a member in the San Francisco Bay area
who is willing to help me test the speakers I could use the help.

Thanks, JH


RMAA seems like it will fit your needs
http://audio.rightmark.org/index_new.shtml

assuming you have adequate sound cards, measurement mic, etc.

[Jim]

The more I read about this the more I wonder if I can get any useful
data running these test at home with a pc, even with a descent mic,
etc.

This is my problem: I built a pair of MTM 2 way speakers using a kit
from Zalytron. I bought a kit because I am a novice speaker builder at
best. I've never been able to test the speakers after I put them
together and I've always wondered if I am getting the optimum
performance out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively, they sound
very good to me but there is "something" missing.

I realize that testing them in my living room with a pc is not the best
condition but can I get some usable data? Of course, once I get the
data I need to know what to do with it. Maybe the speakers will be
just fine and its my ears that need tweeking (after all they are more
than 50 years old), but I wouldn't be surprised to find that the
crossovers need a bit of work.

So, if there is some kind sole out there who wouldn't mind giving me
the benefit of his or her vast knowledge, I would greatly appreciate
it.

Thanks, Jim
TimPerry wrote:
James Lehman wrote:
"Jim" wrote in message
oups.com...
Sorry if this is a repeat question, but I can't seem to find what
I'm looking for. I need a cheap, easy to use, but effective
software to test some speakers I've built. I need to test the
frequency response, the cross over, distortion, phase. . . that
sort of thing.

Can anyone recommend some software?

Better yet. . . if there is a member in the San Francisco Bay area
who is willing to help me test the speakers I could use the help.

Thanks, JH


RMAA seems like it will fit your needs
http://audio.rightmark.org/index_new.shtml

assuming you have adequate sound cards, measurement mic, etc.

[[email protected]]

Speaker Tests are usually done in an anechoic chamber. This is a room
on huge springs in another room and all the walls, incl floor and
ceiling are made of absorbing materials so that no reflections bounce
of them. They are completely silent (you'll hear your heart beat and
nervous system sing if you're in them and listen carefully) and try to
sound like an 'open field', that is to say, not sound at all. Perfect
anechoic chambers do not exist but are the closest you can come to
forming an 'objective' opinion about your speakers. Running tests at
home that involve playing sound over the speaker and using a
measurement mic to analyze its output are completely ridiculous for
they are very much influenced by acoustics. Stick to the multimeter and
just play music you know very well and judge their quality
subjectively. Even then there is a lot you can tell about their imaging
and freq response etc.
If you want to experience your room's acoustic coloration just connect
one speaker, play a sinesweep through it and listen to all the changes
in loudness; then realize that all the changes in loudness are
different for every position of the loudspeaker and the
listener/measurement mic. Loudspeaker placement is very important and
you can change its sound drastically. There are certain simple
guidelines for setting up speakers in a 'colored' room to get a nice
sound. Let me know it you want to hear.

regards Henk

[Andrew Haley]

wrote:
Speaker Tests are usually done in an anechoic chamber.

Not these days. MLSSA measures the anechoic frequency response of
loudspeakers without an anechoic chamber.

"To measure the anechoic frequency response of a loudspeaker MLSSA
uses cursors to select only the initial portion of the measured
impulse response before the arrival of any room reflections (see
figure below). MLSSA then applies an FFT to the selected segment to
yield the anechoic frequency response of the loudspeaker (see figure
right). Because MLSSA measures directly in the time domain, it is a
simple matter to discover any room reflections that might contaminate
the measurement and then window them out of the results."

More at http://www.mlssa.com/pdf/MLSSA-Brochure.pdf

Andrew.

[Arny Krueger]

"Jim" wrote in message
oups.com
Sorry if this is a repeat question, but I can't seem to
find what I'm looking for. I need a cheap, easy to use,
but effective software to test some speakers I've built.
I need to test the frequency response, the cross over,
distortion, phase. . . that sort of thing.

Can anyone recommend some software?

There's a simple and surprisingly effective loudspeaker test facility in
this free softwa

http://audio.rightmark.org/downloads...d-2.2-xram.exe

Eventually you might want something better, but experience with this
software will get you started.

[Arny Krueger]

"Jim" wrote in message
oups.com

Tim Perry wrote:

RMAA seems like it will fit your needs
http://audio.rightmark.org/index_new.shtml

assuming you have adequate sound cards, measurement mic, etc.


The more I read about this the more I wonder if I can get
any useful data running these test at home with a pc,
even with a descent mic, etc.

This is my problem: I built a pair of MTM 2 way speakers
using a kit from Zalytron. I bought a kit because I am a
novice speaker builder at best. I've never been able to
test the speakers after I put them together and I've
always wondered if I am getting the optimum performance
out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively,
they sound very good to me but there is "something"
missing.

I realize that testing them in my living room with a pc
is not the best condition but can I get some usable data?

Yes, by all means.

Of course, once I get the data I need to know what to do
with it.

Among other things the recommended Audio Rightmark program will provide a
frequency response curve generation facility. You need to look at both
on-axis and off-axis response.

There are lots of sites online that have posted frequency response test
results for speaker projects. Compare your speakers to them. Also, beg or
borrow comparable commercial speakers and test them.

IMO Tim Perry gave you really good advice. The Audio Rightmark program can
give you insights into your situation. This software uses a measurement
methodology that gives some immunity to irrelevant room effects.

To complete the test setup, you need a windows PC with a halfways-decent
audio interface, a mic preamp, and a measurement mic. The mic preamp can
be a cheap Behringer mixer such as the UB502 or Xenyx 502. The measurement
mic can be the inexpensive but widely-respected Behringer ECM 8000. A good
PC and $100 or so will set you up.

[[email protected]]

wrote:
Speaker Tests are usually done in an anechoic chamber.

Not any more, they're not. The only advanatge these days to
doing measurements inside an anechoice chamber or noise
isolation and ambient environmental control.

These days and, actually, for the last couple of decades, the
field has been more and more predominated by the use of
various "gated" techniques where the direct arrival signal is
isolated from reflections and reverberation in time, thus
effectively allowing one to do anechoic measurements in
a reflective and reverberent environment.

There are several predominant means of accomplishing this.

1. The first, pioneered by Dick Heyser in the early 1970's,
is a technique called by him "time delay spectrometry."
A sine wave is swept rapidly (and linearly) in frequency.
At the same time, a narrow band-pass filter is swept
synchonously, but with a delay. If the delay corresponds
to the direct-arrival delay from the speaker to the mic-
rophone, then the bandpass filter will aloow that signal
to pass unchanged. But any other signal with a different
delay will be that of a different frequency, and the filter
will, at that point, have moved away form that frequency
and it will be rejected.

2. The second, which encompasses a wide variety of
systems differing primarily in the stimulus signal, is
referred to as "gated" measurement. Basically, a switch,
in effect, turns on the microphone just as the direct
arrival signal reaches the microphone, and turns it off
immediately before the first reflected signal arrives,
thus isolating the direct (anechoic) signal from the
indirect (reflected, reverberent) signal. KEF in GB
pioneered its use for loudspeaker measurement
in the mid 1970's.

There are several subcategories of the technique:

a. Gated sine, where a tone burst is sent out at
discrete frequencies, and the microphone is
gated accordingly. While not intrinsically required,
most systems using this technique are computer-
based. Examples include the original Linear-X's
LMS system and one mode of Audiomatica's CLIO.
The primary advantages are low scost and relatively
simple analysis techniques. The disadvantages are
poor noise immunity, ambiguous phase information,
and slow throughput.

b. Direct impulse techniques, where a narrow impulse
is sent out, the rturend signal is gated, and the result
is run through an FFT to derive the system's complex
transfer function. CLIO can run in this mode as well.
advantages are high-speed data acquisition, complete
complex transfer function information (amplitude and
phase) disadvantages include low noise immunity,
very high crest factors that can lead to overload issues,
and requirements for fairly meaty computational horse-
power (not a real big issue, these days).

c. Maximal Length Sequence and other spectrally dense
systems based on cross-correlation techniques. The
basic principle is that given a spectrally dense signal,
the cross correlation between the output of a system
and the input to the system is the impulse response of
the system. Once the impulse response is in hand,
FFT gets you system transfer function. Gating is still
required to isolate the direct from reverberent data.
Big advantges are high speed, high noise immunity,
complete transfer function information, disadvantages
relatively few, complexity and horsepower being among
them. Examples include the original DRA MLSSA system
and CLIO. There is also something called WinMLS.

Now, what's interesting is that any of these systems, in
a room of given dimensions, will generall, in terms of
bandwidth, perform about as well as an anechoic chamber
of roughly the same overall dimensions. The reason being
is that the chambers cease being anechoic at a low frequency
determined by their physical size, thus setting a low frequency
limit to the measurement.

The open time of the gating window in the gating systems is
ALSO determined by the size of the room they're in, since the
size of the room sets the upper limit to the time delay between
the direct arrival and the first reflection. And, by the time-
frequency uncertainty relation:

1/2 = delta t * delta f

the length of the window in time corresponds, reciprocally,
to the low frequency limit of the measurement

This is not to say that anechoic chambers are never used
anymore, hell, a number of universities have them, left over
from the days when there were no other choices, so they
might as well use them. But as to being the predominant
means of measuring loudspeakers, no.

[Rudi Fischer]

"Jim" wrote:

Sorry if this is a repeat question, but I can't seem to find what I'm
looking for. I need a cheap, easy to use, but effective software to
test some speakers I've built. I need to test the frequency response,
the cross over, distortion, phase. . . that sort of thing.

Can anyone recommend some software?

ARTA
http://www.fesb.hr/~mateljan/arta/index.htm

Rudi Fischer
--
....and may good music always be with you

[[email protected]]

Jim wrote:
The more I read about this the more I wonder if I can get any useful
data running these test at home with a pc, even with a descent mic,
etc.

This is my problem: I built a pair of MTM 2 way speakers using a kit
from Zalytron. I bought a kit because I am a novice speaker builder at
best. I've never been able to test the speakers after I put them
together and I've always wondered if I am getting the optimum
performance out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively, they sound
very good to me but there is "something" missing.

I realize that testing them in my living room with a pc is not the best
condition but can I get some usable data? Of course, once I get the
data I need to know what to do with it. Maybe the speakers will be
just fine and its my ears that need tweeking (after all they are more
than 50 years old), but I wouldn't be surprised to find that the
crossovers need a bit of work.

I guess if you could show simulated results then measure and compare,
that would work out nicely.

While these all these other instruments are very valuable, one should
first start out by using pink noise and a basic spectrum analyzer.
Listening to pink noise is invaluable. It also trains you for things to
listen for. Listening to pink noise at all angles from the speaker is
also a necessary step. Comparison is also a good step. Use other
speakers for reference. After doing instrument testing one should go
back to the beginning and listen.

greg

[Jim]

Whoa . . . lots of great replies. Much to mull over. Since I don't
have access to a chamber I'll try the next best thing: some software
and a descent mic set up.

Thanks for the advise.
Jim

Rudi Fischer wrote:
"Jim" wrote:

Sorry if this is a repeat question, but I can't seem to find what I'm
looking for. I need a cheap, easy to use, but effective software to
test some speakers I've built. I need to test the frequency response,
the cross over, distortion, phase. . . that sort of thing.

Can anyone recommend some software?

ARTA
http://www.fesb.hr/~mateljan/arta/index.htm

Rudi Fischer
--
...and may good music always be with you

[GregS]

In article . com, wrote:

Jim wrote:
The more I read about this the more I wonder if I can get any useful
data running these test at home with a pc, even with a descent mic,
etc.

This is my problem: I built a pair of MTM 2 way speakers using a kit
from Zalytron. I bought a kit because I am a novice speaker builder at
best. I've never been able to test the speakers after I put them
together and I've always wondered if I am getting the optimum
performance out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively, they sound
very good to me but there is "something" missing.

I realize that testing them in my living room with a pc is not the best
condition but can I get some usable data? Of course, once I get the
data I need to know what to do with it. Maybe the speakers will be
just fine and its my ears that need tweeking (after all they are more
than 50 years old), but I wouldn't be surprised to find that the
crossovers need a bit of work.

I guess if you could show simulated results then measure and compare,
that would work out nicely.

While these all these other instruments are very valuable, one should
first start out by using pink noise and a basic spectrum analyzer.
Listening to pink noise is invaluable. It also trains you for things to
listen for. Listening to pink noise at all angles from the speaker is
also a necessary step. Comparison is also a good step. Use other
speakers for reference. After doing instrument testing one should go
back to the beginning and listen.

With an MTM you have to know the intended application or what the vertical
dispersion was designed for, so you can plot response in the vertical plane.
I imaging the kit should have complete set of graphs.

greg

[[email protected]]

Jim wrote:
Whoa . . . lots of great replies. Much to mull over. Since I don't
have access to a chamber I'll try the next best thing: some software
and a descent mic set up.

Thanks for the advise.

That all being said, let me point out that as one who has spent
the last 30 or so years measuring hundreds and thousands of
loudspeakers in and out of chambers, using pretty much all of
the techniques outline here and more, that measuring loud-
speakers consistently, reliably and well is a VERY difficult task.

That's why, after reviewing all the replies, including my own,
I would recommend that you take some else's advice and
consider getting a pink noise source and a 1/3 octave real-
time analyzer. YOu can do it entirely in software on a PC
equipped with a decent microphone and soundcard using
John Murphy's TrueRTA (www.trueaudio.com). It's inexpensive
for the full-featured version, has settable bandwidth from 1/1
octave to 1/24 octave, and has a built in sine, sweep, white
and pink noise generator.

It;s not going to give you detailed phase information or
anything like that, but what would you do with it if you had
it?

Simple to use, reliabl;e, foolproof, and even though it's
a relatively low resolution tool, it avoids many of the pitfalls
of using far more sophisticated, far higher resolution, far
more difficult to use and interpret tools.

[TimPerry]

Jim wrote:
The more I read about this the more I wonder if I can get any useful
data running these test at home with a pc, even with a descent mic,
etc.

This is my problem: I built a pair of MTM 2 way speakers using a kit
from Zalytron. I bought a kit because I am a novice speaker builder
at best. I've never been able to test the speakers after I put them
together and I've always wondered if I am getting the optimum
performance out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively, they sound
very good to me but there is "something" missing.

I realize that testing them in my living room with a pc is not the
best condition but can I get some usable data? Of course, once I get
the data I need to know what to do with it. Maybe the speakers will
be just fine and its my ears that need tweeking (after all they are
more than 50 years old), but I wouldn't be surprised to find that the
crossovers need a bit of work.

So, if there is some kind sole out there who wouldn't mind giving me
the benefit of his or her vast knowledge, I would greatly appreciate
it.

Thanks, Jim

based on the above i think the very best thing you could do is invest in or
build a stereo A/B switch (if your amp does not already have one.) you hook
you kit up to 'A' and you buddies speakers (which he paid megabucks for) to
'B'. then while consuming the beverages of your choice and listening to
you favorite tunes you THINK "wow my hand made speakers sound better then
his expensive boxes" while you SAY: "mine sounds pretty good to me but
yours sound way better"

then you do the same thing next week as HIS place only this time HE buys the
beverages.

you still may be confused as to exactly how good you speakers are but you
will have LOTS more fun.

[bear]

Arny Krueger wrote:
To complete the test setup, you need a windows PC with a halfways-decent
audio interface, a mic preamp, and a measurement mic. The mic preamp can
be a cheap Behringer mixer such as the UB502 or Xenyx 502. The measurement
mic can be the inexpensive but widely-respected Behringer ECM 8000. A good
PC and $100 or so will set you up.

As the Behringer ECM 8000 mic requires phantom power, use the Behringer
UB802...

[Jim]

I may take your advise. However, I doubt that I'll be in any shape to
evaluate anything before too long.
JH
TimPerry wrote:
Jim wrote:
The more I read about this the more I wonder if I can get any useful
data running these test at home with a pc, even with a descent mic,
etc.

This is my problem: I built a pair of MTM 2 way speakers using a kit
from Zalytron. I bought a kit because I am a novice speaker builder
at best. I've never been able to test the speakers after I put them
together and I've always wondered if I am getting the optimum
performance out of them. I'd like to test them to get some sort of
objective "opinion" of their performance. Subjectively, they sound
very good to me but there is "something" missing.

I realize that testing them in my living room with a pc is not the
best condition but can I get some usable data? Of course, once I get
the data I need to know what to do with it. Maybe the speakers will
be just fine and its my ears that need tweeking (after all they are
more than 50 years old), but I wouldn't be surprised to find that the
crossovers need a bit of work.

So, if there is some kind sole out there who wouldn't mind giving me
the benefit of his or her vast knowledge, I would greatly appreciate
it.

Thanks, Jim

based on the above i think the very best thing you could do is invest in or
build a stereo A/B switch (if your amp does not already have one.) you hook
you kit up to 'A' and you buddies speakers (which he paid megabucks for) to
'B'. then while consuming the beverages of your choice and listening to
you favorite tunes you THINK "wow my hand made speakers sound better then
his expensive boxes" while you SAY: "mine sounds pretty good to me but
yours sound way better"

then you do the same thing next week as HIS place only this time HE buys the
beverages.

you still may be confused as to exactly how good you speakers are but you
will have LOTS more fun.

[Arny Krueger]

"bear" wrote in message
ups.com
Arny Krueger wrote:
To complete the test setup, you need a windows PC with a
halfways-decent audio interface, a mic preamp, and a
measurement mic. The mic preamp can be a cheap
Behringer mixer such as the UB502 or Xenyx 502. The
measurement mic can be the inexpensive but
widely-respected Behringer ECM 8000. A good PC and
$100 or so will set you up.

As the Behringer ECM 8000 mic requires phantom power, use
the Behringer UB802...

Thanks for the correction - I didn't realize that the 502 lacked phantom
power. But the online doc for the 502 and 802 make that fact quite clear.