speaker-measurement/speaker-measurement.html
2021-05-17 23:57:47 +02:00

221 lines
7.9 KiB
HTML

<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
<!-- 2021-05-17 lun. 23:57 -->
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<title>Measurement of Loudspeaker</title>
<meta name="author" content="Dehaeze Thomas" />
<meta name="generator" content="Org Mode" />
<link rel="stylesheet" type="text/css" href="https://research.tdehaeze.xyz/css/style.css"/>
<script type="text/javascript" src="https://research.tdehaeze.xyz/js/script.js"></script>
<script>
MathJax = {
svg: {
scale: 1,
fontCache: "global"
},
tex: {
tags: "ams",
multlineWidth: "%MULTLINEWIDTH",
tagSide: "right",
macros: {bm: ["\\boldsymbol{#1}",1],},
tagIndent: ".8em"
}
};
</script>
<script id="MathJax-script" async
src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-svg.js"></script>
</head>
<body>
<div id="org-div-home-and-up">
<a accesskey="h" href="../index.html"> UP </a>
|
<a accesskey="H" href="../index.html"> HOME </a>
</div><div id="content">
<h1 class="title">Measurement of Loudspeaker</h1>
<div id="table-of-contents">
<h2>Table of Contents</h2>
<div id="text-table-of-contents">
<ul>
<li><a href="#orgb7dcd56">1. Equipment</a></li>
<li><a href="#orga8b11ea">2. Semi-Anechoic Outdoor measurement</a></li>
<li><a href="#org5f8fade">3. Ground Plane Technique</a></li>
<li><a href="#orgd4ac4d2">4. In room measurement</a></li>
</ul>
</div>
</div>
<hr>
<p>This report is also available as a <a href="./speaker-measurement.pdf">pdf</a>.</p>
<hr>
<p>
Because of the lack of a large anechoic room or a near field scanner system (e.g. the <a href="http://www.klippel.de/products/rd-system/modules/nfs-near-field-scanner.html#:~:text=The%20Near-Field%20Scanner%203D,move%20during%20the%20scanning%20process.">Klippel NFS</a>), the low frequency and high frequency response of the speaker have to be measured separately.
</p>
<ul class="org-ul">
<li>Section <a href="#orge26b894">1</a>: the tools used for the measurements are listed</li>
<li>Section <a href="#orge527a4d">2</a>: the high frequency behavior of the speaker is measured with a semi-anechoic measurement performed outdoors</li>
<li>Section <a href="#orga1313c1">3</a>: the low frequency behavior of the speaker is measured using the &ldquo;ground plane technique&rdquo;</li>
<li>Section <a href="#org44863a6">4</a>: the response of the speaker in a typical room is estimated from the two above measurements and compare with in-room measurements</li>
</ul>
<div id="outline-container-orgb7dcd56" class="outline-2">
<h2 id="orgb7dcd56"><span class="section-number-2">1</span> Equipment</h2>
<div class="outline-text-2" id="text-1">
<p>
<a id="orge26b894"></a>
</p>
<ul class="org-ul">
<li>REW software (<a href="https://www.roomeqwizard.com/">link</a>)</li>
<li>Audiobox USB (<a href="https://www.presonus.com/products/audiobox-usb">link</a>)</li>
<li>Behringer ECM 8000 Measuring microphone</li>
<li>TOPPING D70 DAC</li>
<li>Power amplifier: Amplifier</li>
</ul>
</div>
</div>
<div id="outline-container-orga8b11ea" class="outline-2">
<h2 id="orga8b11ea"><span class="section-number-2">2</span> Semi-Anechoic Outdoor measurement</h2>
<div class="outline-text-2" id="text-2">
<p>
<a id="orge527a4d"></a>
</p>
<p>
The frequency resolution is limited by the time difference between the direct sound and the first reflection.
</p>
<p>
If we note \(\tau\) this time difference, the frequency resolution \(\delta f\) is:
</p>
\begin{equation}
\delta f = \frac{1}{\tau}
\end{equation}
<p>
with \(\delta f\) in hertz and \(\tau\) in seconds.
</p>
<p>
A schematic of the setup is shown in Figure <a href="#orgd15c0ba">1</a>.
</p>
<div id="orgd15c0ba" class="figure">
<p><img src="figs/outside_meas_setup.png" alt="outside_meas_setup.png" />
</p>
<p><span class="figure-number">Figure 1: </span>Schematic of the setup</p>
</div>
<p>
We usually want to measure the speaker with a resonable distance.
Let&rsquo;s take \(L = 2\,m\).
</p>
<div class="org-src-container">
<pre class="src src-matlab">L = 2; <span class="org-comment">% Distance speaker/microphone [m]</span>
v = 340; <span class="org-comment">% Speed of sound [m/s]</span>
</pre>
</div>
<p>
We can then compute the time delay \(\tau\) between the direct sound and reflected sound as a function of the height \(H\).
Similarly, we can compute the frequency resolution \(\delta f\) as a function of \(H\).
Both are shown in Figure <a href="#org70c02e3">2</a>.
</p>
<div class="org-src-container">
<pre class="src src-matlab">H = [0<span class="org-type">:</span>0.01<span class="org-type">:</span>4]; <span class="org-comment">% [m]</span>
D = 2<span class="org-type">*</span>sqrt(H<span class="org-type">.^</span>2 <span class="org-type">+</span> L<span class="org-type">^</span>2<span class="org-type">/</span>4); <span class="org-comment">% [m]</span>
tau = (D <span class="org-type">-</span> L)<span class="org-type">/</span>v; <span class="org-comment">% [s]</span>
</pre>
</div>
<div id="org70c02e3" class="figure">
<p><img src="figs/required_height_wanted_reflection_delay.png" alt="required_height_wanted_reflection_delay.png" />
</p>
<p><span class="figure-number">Figure 2: </span>Time delay \(\tau\) between the direct sound and (first) reflected sound as a function of the height \(H\). The resulting frequency resolution \(\delta f\) is also shown.</p>
</div>
<div class="important" id="orge571119">
<p>
It is shown than even for high heights (\(H = 3\,m\)), the frequency resolution will be quite poor for low frequency characterisation of the speaker (\(\delta f \approx 70\,Hz\)).
However, it is much sufficient for high frequency characterisation of the speaker (say above 500Hz).
</p>
</div>
<p>
The speaker can then be tilted horizontally and vertically as shown in Figure <a href="#orgd7a99c8">3</a> and <a href="#org98fec49">4</a>.
</p>
<div id="orgd7a99c8" class="figure">
<p><img src="figs/outside_meas_setup_tilt_vert.png" alt="outside_meas_setup_tilt_vert.png" />
</p>
<p><span class="figure-number">Figure 3: </span>Tilting the speaker around a vertical axis</p>
</div>
<div id="org98fec49" class="figure">
<p><img src="figs/outside_meas_setup_tilt_hor.png" alt="outside_meas_setup_tilt_hor.png" />
</p>
<p><span class="figure-number">Figure 4: </span>Tilting the speaker around a horizontal axis</p>
</div>
</div>
</div>
<div id="outline-container-org5f8fade" class="outline-2">
<h2 id="org5f8fade"><span class="section-number-2">3</span> Ground Plane Technique</h2>
<div class="outline-text-2" id="text-3">
<p>
<a id="orga1313c1"></a>
</p>
<p>
The idea here is to put both the microphone and the speaker on the ground.
</p>
<p>
The ground must be relatively stiff.
</p>
<div id="org69a3268" class="figure">
<p><img src="figs/ground_plane_meas_setup.png" alt="ground_plane_meas_setup.png" />
</p>
<p><span class="figure-number">Figure 5: </span>Schematic of the measurement setup</p>
</div>
<p>
It is then possible to use very large gate windows in order to identify the low frequency behavior of the speaker.
</p>
<p>
It is here not useful to perform any off-axis measurements as at the frequencies, the speaker is perfectly omnidirectionnal.
</p>
</div>
</div>
<div id="outline-container-orgd4ac4d2" class="outline-2">
<h2 id="orgd4ac4d2"><span class="section-number-2">4</span> In room measurement</h2>
<div class="outline-text-2" id="text-4">
<p>
<a id="org44863a6"></a>
</p>
<ul class="org-ul">
<li class="off"><code>[&#xa0;]</code> Estimated response</li>
<li class="off"><code>[&#xa0;]</code> Correlation with measurement</li>
</ul>
</div>
</div>
</div>
<div id="postamble" class="status">
<p class="author">Author: Dehaeze Thomas</p>
<p class="date">Created: 2021-05-17 lun. 23:57</p>
</div>
</body>
</html>