Baudline to compare speakers
raudette
I'm looking to roughly compare a variety of speakers, and I'm looking
to plot sound levels against frequency.
So I'm using the tone generator to create a sine wave sweep over a
frequency range, and then using the "average" display to get a rough
idea of how the speaker performs over a given range.
My questions is: is the average display what I really want? Is the
average screen presenting the average sound level at a given frequency
over a time period? How does the "quiet" time before and after I
produce the sweep affect results?
I understand that my results will be far from perfect, due to
uncalibrated microphones, imperfect room with sound reflections etc...,
but I'm looking at relative performance of various speakers.
As audiopenguin suggested earlier, setting the Hz scale to logarithmic
Hz would be a useful feature.
That being said, Baudline is pretty slick.
- Richard
Sig Blip
I like both White Gaussian Noise (WGN) and linear sine sweeps for device testing. They have different strengths and weaknesses so I use them both. You might be interested in reading these three baudline test and measurement application notes:
http://www.baudline.com/solutions/swept_sine_vs_wgn/index.html
http://www.baudline.com/solutions/full_duplex/index.html
http://www.baudline.com/solutions/codec/index.html
They describe some loopback techniques and procedures that should be applicable to your speaker testing.
Baudline has some useful calibration tools; a sample rate correction feature and an Equalization window. You can read more about them in the on-line documentation pages:
http://www.baudline.com/manual/input.html
http://www.baudline.com/manual/equalization.html
If you are testing the low frequencies of a sub woofer then you might be interesting in using baudline's decimation feature in the Input Devices window. Decimation is a great way of increasing the bass frequency resolution.
raudette
thanks!
Richard
> ------=_Part_14200_1058037.1146781644959
> Content-Type: text/html; charset=ISO-8859-1
> Content-Transfer-Encoding: quoted-printable
> X-Google-AttachSize: 3426
>
> Yes you are correct, the Average window is the best window to use for
> speaker frequency response measurements. The Average window is an
> accumulation of spectral traces so collection time does make a
> difference. Longer collection durations will lower the variance
> of the noise floor and the averaging of "quiet" signals will lower the
> absolute amplitude which is important if you are comparing spectral
> traces. Remember that the Average window can collect data while
> in baudline's record mode and sample data can be copy-n-pasted while in
> the pause mode.<br>
> <br>
> I like both White Gaussian Noise (WGN) and linear sine sweeps for
> device testing. They have different strengths and weaknesses so I
> use them both. You might be interested in reading these three
> baudline test and measurement application notes:<br>
> <br>
> <a href="http://www.baudline.com/solutions/swept_sine_vs_wgn/index.html">http://www.baudline.com/solutions/swept_sine_vs_wgn/index.html</a><br>
> <a href="http://www.baudline.com/solutions/full_duplex/index.html">http://www.baudline.com/solutions/full_duplex/index.html</a><br>
> <a href="http://www.baudline.com/solutions/codec/index.html">http://www.baudline.com/solutions/codec/index.html</a><br>
> <br>
> They describe some loopback techniques and procedures that should be applicable to your speaker testing.<br>
> <br>
> Baudline has some useful calibration tools; a sample rate correction
> feature and an Equalization window. You can read more about them in the
> on-line documentation pages:<br>
> <br>
> <a href="http://www.baudline.com/manual/input.html">http://www.baudline.com/manual/input.html</a><br>
> <a href="http://www.baudline.com/manual/equalization.html">http://www.baudline.com/manual/equalization.html</a><br>
> <br>If you are testing the low frequencies of a sub woofer then you
> might be interesting in using baudline's decimation feature in the
> Input Devices window. Decimation is a great way of increasing the
> bass frequency resolution.<br>
> <br><div><span class="gmail_quote">On 5/3/06, <b class="gmail_sendername">raudette</b> <<a href="mailto:raud...@organworks.com">raud...@organworks.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
> <br>I am new to signal analysis.<br><br>I'm looking to roughly compare a variety of speakers, and I'm looking<br>to plot sound levels against frequency.<br><br>So I'm using the tone generator to create a sine wave sweep over a
> <br>frequency range, and then using the "average" display to get a rough<br>idea of how the speaker performs over a given range.<br><br>My questions is: is the average display what I really want? Is the<br>average screen presenting the average sound level at a given frequency
> <br>over a time period? How does the "quiet" time before and after I<br>produce the sweep affect results?<br><br>I understand that my results will be far from perfect, due to<br>uncalibrated microphones, imperfect room with sound reflections etc...,
> <br>but I'm looking at relative performance of various speakers.<br><br>As audiopenguin suggested earlier, setting the Hz scale to logarithmic<br>Hz would be a useful feature.<br><br>That being said, Baudline is pretty slick.
> <br><br>- Richard<br><br><br>
> ------=_Part_14200_1058037.1146781644959--
Sig Blip
Also, decimation can be very valuable for weak signal infrasound monitoring. When looking at weak signals with baudline I find that some decimation gain (+dB next to the Decimate By control) and some slice integration (try the "anti-alias on spectrogram zoom" option in the Drift Integrator) can help pull the signal out of the noise a little bit more.