828 lines
34 KiB
HTML
828 lines
34 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>
|
|
<!-- 2021-02-20 sam. 23:08 -->
|
|
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
|
|
<title>Determination of the optimal nano-hexapod’s stiffness for reducing the effect of disturbances</title>
|
|
<meta name="generator" content="Org mode" />
|
|
<meta name="author" content="Dehaeze Thomas" />
|
|
<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">Determination of the optimal nano-hexapod’s stiffness for reducing the effect of disturbances</h1>
|
|
<div id="table-of-contents">
|
|
<h2>Table of Contents</h2>
|
|
<div id="text-table-of-contents">
|
|
<ul>
|
|
<li><a href="#org1d00c3b">1. Disturbances</a></li>
|
|
<li><a href="#orgd8c957a">2. Effect of disturbances on the position error</a>
|
|
<ul>
|
|
<li><a href="#org35d2f91">2.1. Initialization</a></li>
|
|
<li><a href="#org094b5a7">2.2. Identification</a></li>
|
|
<li><a href="#org7e3e79d">2.3. Sensitivity to Stages vibration (Filtering)</a></li>
|
|
<li><a href="#org19c4b17">2.4. Effect of Ground motion (Transmissibility).</a></li>
|
|
<li><a href="#org0890188">2.5. Direct Forces (Compliance).</a></li>
|
|
<li><a href="#orgea2d185">2.6. Conclusion</a></li>
|
|
</ul>
|
|
</li>
|
|
<li><a href="#orgf6c1d0f">3. Effect of granite stiffness</a>
|
|
<ul>
|
|
<li><a href="#org9b72d9e">3.1. Analytical Analysis</a>
|
|
<ul>
|
|
<li><a href="#orgcb7fd26">3.1.1. Simple mass-spring-damper model</a></li>
|
|
<li><a href="#orgb9ac8f4">3.1.2. General Case</a></li>
|
|
</ul>
|
|
</li>
|
|
<li><a href="#orge9846e9">3.2. Soft Granite</a></li>
|
|
<li><a href="#org699ef0a">3.3. Effect of the Granite transfer function</a></li>
|
|
<li><a href="#org26b072f">3.4. Conclusion</a></li>
|
|
</ul>
|
|
</li>
|
|
<li><a href="#org92ed7b0">4. Open Loop Budget Error</a>
|
|
<ul>
|
|
<li><a href="#orgf8bb217">4.1. Noise Budgeting - Theory</a></li>
|
|
<li><a href="#org0a97ffb">4.2. Power Spectral Densities</a></li>
|
|
<li><a href="#org44eafed">4.3. Cumulative Amplitude Spectrum</a></li>
|
|
<li><a href="#org091b440">4.4. Conclusion</a></li>
|
|
</ul>
|
|
</li>
|
|
<li><a href="#org5bb46c4">5. Closed Loop Budget Error</a>
|
|
<ul>
|
|
<li><a href="#org2cb5e9a">5.1. Approximation of the effect of feedback on the motion error</a></li>
|
|
<li><a href="#org15ee341">5.2. Reduction thanks to feedback - Required bandwidth</a></li>
|
|
</ul>
|
|
</li>
|
|
<li><a href="#orgf3b0566">6. Conclusion</a></li>
|
|
</ul>
|
|
</div>
|
|
</div>
|
|
|
|
<p>
|
|
In this document is studied how the stiffness of the nano-hexapod will impact the effect of disturbances on the position error of the sample.
|
|
</p>
|
|
|
|
<p>
|
|
It is divided in the following sections:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>Section <a href="#orge637f3d">1</a>: the disturbances are listed and their Power Spectral Densities (PSD) are shown</li>
|
|
<li>Section <a href="#org8db9681">2</a>: the transfer functions from disturbances to the position error of the sample are computed for a wide range of nano-hexapod stiffnesses</li>
|
|
<li>Section <a href="#orge639dbd">3</a>:</li>
|
|
<li>Section <a href="#org5714572">4</a>: from both the PSD of the disturbances and the transfer function from disturbances to sample’s position errors, we compute the resulting PSD and Cumulative Amplitude Spectrum (CAS)</li>
|
|
<li>Section <a href="#org6ed8184">5</a>: from a simplistic model is computed the required control bandwidth to reduce the position error to acceptable values</li>
|
|
</ul>
|
|
|
|
<div id="outline-container-org1d00c3b" class="outline-2">
|
|
<h2 id="org1d00c3b"><span class="section-number-2">1</span> Disturbances</h2>
|
|
<div class="outline-text-2" id="text-1">
|
|
<p>
|
|
<a id="orge637f3d"></a>
|
|
</p>
|
|
<p>
|
|
The main disturbances considered here are:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>\(D_w\): Ground displacement in the \(x\), \(y\) and \(z\) directions</li>
|
|
<li>\(F_{ty}\): Forces applied by the Translation stage in the \(x\) and \(z\) directions</li>
|
|
<li>\(F_{rz}\): Forces applied by the Spindle in the \(z\) direction</li>
|
|
<li>\(F_d\): Direct forces applied at the center of mass of the Payload</li>
|
|
</ul>
|
|
|
|
<p>
|
|
The level of these disturbances has been identified form experiments which are detailed in <a href="disturbances.html">this</a> document.
|
|
</p>
|
|
<p>
|
|
The measured Amplitude Spectral Densities (ASD) of these forces are shown in Figures <a href="#org3baaaea">1</a> and <a href="#orga5ce540">2</a>.
|
|
</p>
|
|
|
|
<p>
|
|
In this study, the expected frequency content of the direct forces applied to the payload is not considered.
|
|
</p>
|
|
|
|
|
|
<div id="org3baaaea" class="figure">
|
|
<p><img src="figs/opt_stiff_dist_gm.png" alt="opt_stiff_dist_gm.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 1: </span>Amplitude Spectral Density of the Ground Displacement (<a href="./figs/opt_stiff_dist_gm.png">png</a>, <a href="./figs/opt_stiff_dist_gm.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="orga5ce540" class="figure">
|
|
<p><img src="figs/opt_stiff_dist_fty_frz.png" alt="opt_stiff_dist_fty_frz.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 2: </span>Amplitude Spectral Density of the “parasitic” forces comming from the Translation stage and the spindle (<a href="./figs/opt_stiff_dist_fty_frz.png">png</a>, <a href="./figs/opt_stiff_dist_fty_frz.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orgd8c957a" class="outline-2">
|
|
<h2 id="orgd8c957a"><span class="section-number-2">2</span> Effect of disturbances on the position error</h2>
|
|
<div class="outline-text-2" id="text-2">
|
|
<p>
|
|
<a id="org8db9681"></a>
|
|
</p>
|
|
<p>
|
|
In this section, we use the Simscape model to identify the transfer function from disturbances to the position error of the sample.
|
|
We do that for a wide range of nano-hexapod stiffnesses and we compare the obtained results.
|
|
</p>
|
|
</div>
|
|
<div id="outline-container-org35d2f91" class="outline-3">
|
|
<h3 id="org35d2f91"><span class="section-number-3">2.1</span> Initialization</h3>
|
|
<div class="outline-text-3" id="text-2-1">
|
|
<p>
|
|
We initialize all the stages with the default parameters.
|
|
</p>
|
|
<div class="org-src-container">
|
|
<pre class="src src-matlab"> initializeGround();
|
|
initializeGranite();
|
|
initializeTy();
|
|
initializeRy();
|
|
initializeRz();
|
|
initializeMicroHexapod();
|
|
initializeAxisc();
|
|
initializeMirror();
|
|
</pre>
|
|
</div>
|
|
|
|
<p>
|
|
We use a sample mass of 10kg.
|
|
</p>
|
|
<div class="org-src-container">
|
|
<pre class="src src-matlab"> initializeSample(<span class="org-string">'mass'</span>, 10);
|
|
</pre>
|
|
</div>
|
|
|
|
<p>
|
|
We include gravity, and we use no controller.
|
|
</p>
|
|
<div class="org-src-container">
|
|
<pre class="src src-matlab"> initializeSimscapeConfiguration(<span class="org-string">'gravity'</span>, <span class="org-constant">true</span>);
|
|
initializeController();
|
|
initializeDisturbances(<span class="org-string">'enable'</span>, <span class="org-constant">false</span>);
|
|
initializeLoggingConfiguration(<span class="org-string">'log'</span>, <span class="org-string">'none'</span>);
|
|
</pre>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org094b5a7" class="outline-3">
|
|
<h3 id="org094b5a7"><span class="section-number-3">2.2</span> Identification</h3>
|
|
<div class="outline-text-3" id="text-2-2">
|
|
<p>
|
|
The considered inputs are:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li><code>Dwx</code>: Ground displacement in the \(x\) direction</li>
|
|
<li><code>Dwy</code>: Ground displacement in the \(y\) direction</li>
|
|
<li><code>Dwz</code>: Ground displacement in the \(z\) direction</li>
|
|
<li><code>Fty_x</code>: Forces applied by the Translation stage in the \(x\) direction</li>
|
|
<li><code>Fty_z</code>: Forces applied by the Translation stage in the \(z\) direction</li>
|
|
<li><code>Frz_z</code>: Forces applied by the Spindle in the \(z\) direction</li>
|
|
<li><code>Fd</code>: Direct forces applied at the center of mass of the Payload</li>
|
|
</ul>
|
|
|
|
<p>
|
|
The outputs are <code>Ex</code>, <code>Ey</code>, <code>Ez</code>, <code>Erx</code>, <code>Ery</code>, <code>Erz</code> which are the 3 positions and 3 orientations errors of the sample.
|
|
</p>
|
|
|
|
<p>
|
|
We initialize the set of the nano-hexapod stiffnesses, and for each of them, we identify the dynamics from defined inputs to defined outputs.
|
|
</p>
|
|
<div class="org-src-container">
|
|
<pre class="src src-matlab"> Ks = logspace(3,9,7); <span class="org-comment">% [N/m]</span>
|
|
</pre>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org7e3e79d" class="outline-3">
|
|
<h3 id="org7e3e79d"><span class="section-number-3">2.3</span> Sensitivity to Stages vibration (Filtering)</h3>
|
|
<div class="outline-text-3" id="text-2-3">
|
|
<p>
|
|
The sensitivity the stage vibrations are displayed:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>Figure <a href="#org0073408">3</a>: sensitivity to vertical spindle vibrations</li>
|
|
<li>Figure <a href="#org0999f05">4</a>: sensitivity to vertical translation stage vibrations</li>
|
|
<li>Figure <a href="#org6337962">5</a>: sensitivity to horizontal (x) translation stage vibrations</li>
|
|
</ul>
|
|
|
|
|
|
<div id="org0073408" class="figure">
|
|
<p><img src="figs/opt_stiff_sensitivity_Frz.png" alt="opt_stiff_sensitivity_Frz.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 3: </span>Sensitivity to Spindle vertical motion error (\(F_{rz}\)) to the vertical error position of the sample (\(E_z\)) (<a href="./figs/opt_stiff_sensitivity_Frz.png">png</a>, <a href="./figs/opt_stiff_sensitivity_Frz.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org0999f05" class="figure">
|
|
<p><img src="figs/opt_stiff_sensitivity_Fty_z.png" alt="opt_stiff_sensitivity_Fty_z.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 4: </span>Sensitivity to Translation stage vertical motion error (\(F_{ty,z}\)) to the vertical error position of the sample (\(E_z\)) (<a href="./figs/opt_stiff_sensitivity_Fty_z.png">png</a>, <a href="./figs/opt_stiff_sensitivity_Fty_z.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org6337962" class="figure">
|
|
<p><img src="figs/opt_stiff_sensitivity_Fty_x.png" alt="opt_stiff_sensitivity_Fty_x.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 5: </span>Sensitivity to Translation stage \(x\) motion error (\(F_{ty,x}\)) to the error position of the sample in the \(x\) direction (\(E_x\)) (<a href="./figs/opt_stiff_sensitivity_Fty_x.png">png</a>, <a href="./figs/opt_stiff_sensitivity_Fty_x.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org19c4b17" class="outline-3">
|
|
<h3 id="org19c4b17"><span class="section-number-3">2.4</span> Effect of Ground motion (Transmissibility).</h3>
|
|
<div class="outline-text-3" id="text-2-4">
|
|
<p>
|
|
The effect of Ground motion on the position error of the sample is shown in Figure <a href="#org671bbd6">6</a>.
|
|
</p>
|
|
|
|
|
|
<div id="org671bbd6" class="figure">
|
|
<p><img src="figs/opt_stiff_sensitivity_Dw.png" alt="opt_stiff_sensitivity_Dw.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 6: </span>Sensitivity to Ground motion (\(D_{w}\)) to the position error of the sample (\(E_y\) and \(E_z\)) (<a href="./figs/opt_stiff_sensitivity_Dw.png">png</a>, <a href="./figs/opt_stiff_sensitivity_Dw.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org0890188" class="outline-3">
|
|
<h3 id="org0890188"><span class="section-number-3">2.5</span> Direct Forces (Compliance).</h3>
|
|
<div class="outline-text-3" id="text-2-5">
|
|
<p>
|
|
The effect of direct forces/torques applied on the sample (cable forces for instance) on the position error of the sample is shown in Figure <a href="#org022a37b">7</a>.
|
|
</p>
|
|
|
|
|
|
<div id="org022a37b" class="figure">
|
|
<p><img src="figs/opt_stiff_sensitivity_Fd.png" alt="opt_stiff_sensitivity_Fd.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 7: </span>Sensitivity to Direct forces and torques applied to the sample (\(F_d\), \(M_d\)) to the position error of the sample (<a href="./figs/opt_stiff_sensitivity_Fd.png">png</a>, <a href="./figs/opt_stiff_sensitivity_Fd.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orgea2d185" class="outline-3">
|
|
<h3 id="orgea2d185"><span class="section-number-3">2.6</span> Conclusion</h3>
|
|
<div class="outline-text-3" id="text-2-6">
|
|
<div class="important" id="orgb58fd33">
|
|
<p>
|
|
Reducing the nano-hexapod stiffness generally lowers the sensitivity to stages vibration but increases the sensitivity to ground motion and direct forces.
|
|
</p>
|
|
|
|
<p>
|
|
In order to conclude on the optimal stiffness that will yield the smallest sample vibration, one has to include the level of disturbances. This is done in Section <a href="#org5714572">4</a>.
|
|
</p>
|
|
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orgf6c1d0f" class="outline-2">
|
|
<h2 id="orgf6c1d0f"><span class="section-number-2">3</span> Effect of granite stiffness</h2>
|
|
<div class="outline-text-2" id="text-3">
|
|
<p>
|
|
<a id="orge639dbd"></a>
|
|
</p>
|
|
<p>
|
|
In this section, we wish to see if a soft granite suspension could help in reducing the effect of disturbances on the position error of the sample.
|
|
</p>
|
|
</div>
|
|
<div id="outline-container-org9b72d9e" class="outline-3">
|
|
<h3 id="org9b72d9e"><span class="section-number-3">3.1</span> Analytical Analysis</h3>
|
|
<div class="outline-text-3" id="text-3-1">
|
|
</div>
|
|
<div id="outline-container-orgcb7fd26" class="outline-4">
|
|
<h4 id="orgcb7fd26"><span class="section-number-4">3.1.1</span> Simple mass-spring-damper model</h4>
|
|
<div class="outline-text-4" id="text-3-1-1">
|
|
<p>
|
|
Let’s consider the system shown in Figure <a href="#org210f3bb">8</a> consisting of two stacked mass-spring-damper systems.
|
|
The bottom one represents the granite, and the top one all the positioning stages.
|
|
We want the smallest stage “deformation” \(d = x^\prime - x\) due to ground motion \(w\).
|
|
</p>
|
|
|
|
|
|
<div id="org210f3bb" class="figure">
|
|
<p><img src="figs/2dof_system_granite_stiffness.png" alt="2dof_system_granite_stiffness.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 8: </span>Mass Spring Damper system consisting of a granite and a positioning stage</p>
|
|
</div>
|
|
|
|
<p>
|
|
If we write the equation of motion of the system in Figure <a href="#org210f3bb">8</a>, we obtain:
|
|
</p>
|
|
\begin{align}
|
|
m^\prime s^2 x^\prime &= (c^\prime s + k^\prime) (x - x^\prime) \\
|
|
ms^2 x &= (c^\prime s + k^\prime) (x^\prime - x) + (cs + k) (w - x)
|
|
\end{align}
|
|
|
|
<p>
|
|
If we note \(d = x^\prime - x\), we obtain:
|
|
</p>
|
|
\begin{equation}
|
|
\frac{d}{w} = \frac{-m^\prime s^2 (cs + k)}{ (m^\prime s^2 + c^\prime s + k^\prime) (ms^2 + cs + k) + m^\prime s^2(c^\prime s + k^\prime)}
|
|
\end{equation}
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orgb9ac8f4" class="outline-4">
|
|
<h4 id="orgb9ac8f4"><span class="section-number-4">3.1.2</span> General Case</h4>
|
|
<div class="outline-text-4" id="text-3-1-2">
|
|
<p>
|
|
Let’s now considering a general positioning stage defined by:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>\(G^\prime(s) = \frac{F}{x}\): its mechanical “impedance”</li>
|
|
<li>\(D^\prime(s) = \frac{d}{x}\): its “deformation” transfer function</li>
|
|
</ul>
|
|
|
|
|
|
<div id="org82d8583" class="figure">
|
|
<p><img src="figs/general_system_granite_stiffness.png" alt="general_system_granite_stiffness.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 9: </span>Mass Spring Damper representing the granite and a general representation of positioning stages</p>
|
|
</div>
|
|
|
|
<p>
|
|
The equation of motion are:
|
|
</p>
|
|
\begin{align}
|
|
ms^2 x &= (cs + k) (x - w) - F \\
|
|
F &= G^\prime(s) x \\
|
|
d &= D^\prime(s) x
|
|
\end{align}
|
|
<p>
|
|
where:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>\(F\) is the force applied by the position stages on the granite mass</li>
|
|
</ul>
|
|
|
|
<div class="important" id="org3106f21">
|
|
<p>
|
|
We can express \(d\) as a function of \(w\):
|
|
</p>
|
|
\begin{equation}
|
|
\frac{d}{w} = \frac{D^\prime(s) (cs + k)}{ms^2 + cs + k + G^\prime(s)}
|
|
\end{equation}
|
|
|
|
<p>
|
|
This is the transfer function that we would like to minimize.
|
|
</p>
|
|
|
|
</div>
|
|
|
|
<p>
|
|
Let’s verify this formula for a simple mass/spring/damper positioning stage.
|
|
In that case, we have:
|
|
</p>
|
|
\begin{align*}
|
|
D^\prime(s) &= \frac{d}{x} = \frac{- m^\prime s^2}{m^\prime s^2 + c^\prime s + k^\prime} \\
|
|
G^\prime(s) &= \frac{F}{x} = \frac{m^\prime s^2(c^\prime s + k)}{m^\prime s^2 + c^\prime s + k^\prime}
|
|
\end{align*}
|
|
|
|
<p>
|
|
And finally:
|
|
</p>
|
|
\begin{equation}
|
|
\frac{d}{w} = \frac{-m^\prime s^2 (cs + k)}{ (m^\prime s^2 + c^\prime s + k^\prime) (ms^2 + cs + k) + m^\prime s^2(c^\prime s + k^\prime)}
|
|
\end{equation}
|
|
<p>
|
|
which is the same as the previously derived equation.
|
|
</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orge9846e9" class="outline-3">
|
|
<h3 id="orge9846e9"><span class="section-number-3">3.2</span> Soft Granite</h3>
|
|
<div class="outline-text-3" id="text-3-2">
|
|
<p>
|
|
Let’s initialize a soft granite and see how the sensitivity to disturbances will change.
|
|
</p>
|
|
<div class="org-src-container">
|
|
<pre class="src src-matlab"> initializeGranite(<span class="org-string">'K'</span>, 5e5<span class="org-type">*</span>ones(3,1), <span class="org-string">'C'</span>, 5e3<span class="org-type">*</span>ones(3,1));
|
|
</pre>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org699ef0a" class="outline-3">
|
|
<h3 id="org699ef0a"><span class="section-number-3">3.3</span> Effect of the Granite transfer function</h3>
|
|
<div class="outline-text-3" id="text-3-3">
|
|
<p>
|
|
From Figure <a href="#org5ecd9f8">10</a>, we can see that having a “soft” granite suspension greatly lowers the sensitivity to ground motion.
|
|
The sensitivity is indeed lowered starting from the resonance of the granite on its soft suspension (few Hz here).
|
|
</p>
|
|
|
|
<p>
|
|
From Figures <a href="#org71cfcc7">11</a> and <a href="#org0bb4ee0">12</a>, we see that the change of granite suspension does not change a lot the sensitivity to both direct forces and stage vibrations.
|
|
</p>
|
|
|
|
|
|
<div id="org5ecd9f8" class="figure">
|
|
<p><img src="figs/opt_stiff_soft_granite_Dw.png" alt="opt_stiff_soft_granite_Dw.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 10: </span>Change of sensibility to Ground motion when using a stiff Granite (solid curves) and a soft Granite (dashed curves) (<a href="./figs/opt_stiff_soft_granite_Dw.png">png</a>, <a href="./figs/opt_stiff_soft_granite_Dw.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org71cfcc7" class="figure">
|
|
<p><img src="figs/opt_stiff_soft_granite_Frz.png" alt="opt_stiff_soft_granite_Frz.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 11: </span>Change of sensibility to Spindle vibrations when using a stiff Granite (solid curves) and a soft Granite (dashed curves) (<a href="./figs/opt_stiff_soft_granite_Frz.png">png</a>, <a href="./figs/opt_stiff_soft_granite_Frz.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org0bb4ee0" class="figure">
|
|
<p><img src="figs/opt_stiff_soft_granite_Fd.png" alt="opt_stiff_soft_granite_Fd.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 12: </span>Change of sensibility to direct forces when using a stiff Granite (solid curves) and a soft Granite (dashed curves) (<a href="./figs/opt_stiff_soft_granite_Fd.png">png</a>, <a href="./figs/opt_stiff_soft_granite_Fd.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org26b072f" class="outline-3">
|
|
<h3 id="org26b072f"><span class="section-number-3">3.4</span> Conclusion</h3>
|
|
<div class="outline-text-3" id="text-3-4">
|
|
<div class="important" id="orge437d65">
|
|
<p>
|
|
Having a soft granite suspension greatly decreases the sensitivity the ground motion.
|
|
Also, it does not affect much the sensitivity to stage vibration and direct forces.
|
|
Thus the level of sample vibration can be reduced by using a soft granite suspension if it is found that ground motion is the limiting factor.
|
|
</p>
|
|
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org92ed7b0" class="outline-2">
|
|
<h2 id="org92ed7b0"><span class="section-number-2">4</span> Open Loop Budget Error</h2>
|
|
<div class="outline-text-2" id="text-4">
|
|
<p>
|
|
<a id="org5714572"></a>
|
|
</p>
|
|
<p>
|
|
Now that the frequency content of disturbances have been estimated (Section <a href="#orge637f3d">1</a>) and the transfer functions from disturbances to the position error of the sample have been identified (Section <a href="#org8db9681">2</a>), we can compute the level of sample vibration due to the disturbances.
|
|
</p>
|
|
|
|
<p>
|
|
We then can conclude and the nano-hexapod stiffness that will lower the sample position error.
|
|
</p>
|
|
</div>
|
|
<div id="outline-container-orgf8bb217" class="outline-3">
|
|
<h3 id="orgf8bb217"><span class="section-number-3">4.1</span> Noise Budgeting - Theory</h3>
|
|
<div class="outline-text-3" id="text-4-1">
|
|
<p>
|
|
Let’s consider Figure <a href="#orgf48ae3a">13</a> there \(G_d(s)\) is the transfer function from a signal \(d\) (the perturbation) to a signal \(y\) (the sample’s position error).
|
|
</p>
|
|
|
|
|
|
<div id="orgf48ae3a" class="figure">
|
|
<p><img src="figs/psd_change_tf.png" alt="psd_change_tf.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 13: </span>Signal \(d\) going through and LTI transfer function \(G_d(s)\) to give a signal \(y\)</p>
|
|
</div>
|
|
|
|
<p>
|
|
We can compute the Power Spectral Density (PSD) of signal \(y\) from the PSD of \(d\) and the norm of \(G_d(s)\):
|
|
</p>
|
|
\begin{equation}
|
|
S_{y}(\omega) = \left|G_d(j\omega)\right|^2 S_{d}(\omega) \label{eq:psd_transfer_function}
|
|
\end{equation}
|
|
|
|
<p>
|
|
If we now consider multiple disturbances \(d_1, \dots, d_n\) as shown in Figure <a href="#orga4616f6">14</a>, we have that:
|
|
</p>
|
|
\begin{equation}
|
|
S_{y}(\omega) = \left|G_{d_1}(j\omega)\right|^2 S_{d_1}(\omega) + \dots + \left|G_{d_n}(j\omega)\right|^2 S_{d_n}(\omega) \label{eq:sum_psd}
|
|
\end{equation}
|
|
|
|
<p>
|
|
Sometimes, we prefer to compute the <b>Amplitude</b> Spectral Density (ASD) which is related to the PSD by:
|
|
\[ \Gamma_y(\omega) = \sqrt{S_y(\omega)} \]
|
|
</p>
|
|
|
|
|
|
<div id="orga4616f6" class="figure">
|
|
<p><img src="figs/psd_change_tf_multiple_pert.png" alt="psd_change_tf_multiple_pert.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 14: </span>Block diagram showing and output \(y\) resulting from the addition of multiple perturbations \(d_i\)</p>
|
|
</div>
|
|
|
|
<p>
|
|
The Cumulative Power Spectrum (CPS) is here defined as:
|
|
</p>
|
|
\begin{equation}
|
|
\Phi_y(\omega) = \int_\omega^\infty S_y(\nu) d\nu
|
|
\end{equation}
|
|
|
|
<p>
|
|
And the Cumulative Amplitude Spectrum (CAS):
|
|
</p>
|
|
\begin{equation}
|
|
\Psi(\omega) = \sqrt{\Phi(\omega)} = \sqrt{\int_\omega^\infty S_y(\nu) d\nu}
|
|
\end{equation}
|
|
|
|
<p>
|
|
The CAS evaluation for all frequency corresponds to the rms value of the considered quantity:
|
|
\[ y_{\text{rms}} = \Psi(\omega = 0) = \sqrt{\int_0^\infty S_y(\nu) d\nu} \]
|
|
</p>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org0a97ffb" class="outline-3">
|
|
<h3 id="org0a97ffb"><span class="section-number-3">4.2</span> Power Spectral Densities</h3>
|
|
<div class="outline-text-3" id="text-4-2">
|
|
<p>
|
|
We compute the effect of perturbations on the motion error thanks to Eq. \eqref{eq:psd_transfer_function}.
|
|
</p>
|
|
|
|
<p>
|
|
The result is shown in:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>Figure <a href="#orgd8bc100">15</a>: PSD of the vertical sample’s motion error due to vertical ground motion</li>
|
|
<li>Figure <a href="#org631c8cb">16</a>: PSD of the vertical sample’s motion error due to vertical vibrations of the Spindle</li>
|
|
</ul>
|
|
|
|
|
|
<div id="orgd8bc100" class="figure">
|
|
<p><img src="figs/opt_stiff_psd_dz_gm.png" alt="opt_stiff_psd_dz_gm.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 15: </span>Amplitude Spectral Density of the Sample vertical position error due to Ground motion for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_psd_dz_gm.png">png</a>, <a href="./figs/opt_stiff_psd_dz_gm.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org631c8cb" class="figure">
|
|
<p><img src="figs/opt_stiff_psd_dz_rz.png" alt="opt_stiff_psd_dz_rz.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 16: </span>Amplitude Spectral Density of the Sample vertical position error due to Vertical vibration of the Spindle for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_psd_dz_rz.png">png</a>, <a href="./figs/opt_stiff_psd_dz_rz.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
<p>
|
|
We compute the effect of all perturbations on the vertical position error using Eq. \eqref{eq:sum_psd} and the resulting PSD is shown in Figure <a href="#org5ce09d9">17</a>.
|
|
</p>
|
|
|
|
<div id="org5ce09d9" class="figure">
|
|
<p><img src="figs/opt_stiff_psd_dz_tot.png" alt="opt_stiff_psd_dz_tot.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 17: </span>Amplitude Spectral Density of the Sample vertical position error due to all considered perturbations for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_psd_dz_tot.png">png</a>, <a href="./figs/opt_stiff_psd_dz_tot.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org44eafed" class="outline-3">
|
|
<h3 id="org44eafed"><span class="section-number-3">4.3</span> Cumulative Amplitude Spectrum</h3>
|
|
<div class="outline-text-3" id="text-4-3">
|
|
<p>
|
|
Similarly, the Cumulative Amplitude Spectrum of the sample vibrations are shown:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>Figure <a href="#orgccc4019">18</a>: due to vertical ground motion</li>
|
|
<li>Figure <a href="#orgbc68410">19</a>: due to vertical vibrations of the Spindle</li>
|
|
<li>Figure <a href="#org0bcb1fa">20</a>: due to all considered perturbations</li>
|
|
</ul>
|
|
|
|
<p>
|
|
The black dashed line corresponds to the performance objective of a sample vibration equal to \(10\ nm [rms]\).
|
|
</p>
|
|
|
|
|
|
<div id="orgccc4019" class="figure">
|
|
<p><img src="figs/opt_stiff_cas_dz_gm.png" alt="opt_stiff_cas_dz_gm.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 18: </span>Cumulative Amplitude Spectrum of the Sample vertical position error due to Ground motion for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_cas_dz_gm.png">png</a>, <a href="./figs/opt_stiff_cas_dz_gm.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="orgbc68410" class="figure">
|
|
<p><img src="figs/opt_stiff_cas_dz_rz.png" alt="opt_stiff_cas_dz_rz.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 19: </span>Cumulative Amplitude Spectrum of the Sample vertical position error due to Vertical vibration of the Spindle for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_cas_dz_rz.png">png</a>, <a href="./figs/opt_stiff_cas_dz_rz.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="org0bcb1fa" class="figure">
|
|
<p><img src="figs/opt_stiff_cas_dz_tot.png" alt="opt_stiff_cas_dz_tot.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 20: </span>Cumulative Amplitude Spectrum of the Sample vertical position error due to all considered perturbations for multiple nano-hexapod stiffnesses (<a href="./figs/opt_stiff_cas_dz_tot.png">png</a>, <a href="./figs/opt_stiff_cas_dz_tot.pdf">pdf</a>)</p>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org091b440" class="outline-3">
|
|
<h3 id="org091b440"><span class="section-number-3">4.4</span> Conclusion</h3>
|
|
<div class="outline-text-3" id="text-4-4">
|
|
<div class="important" id="org86258fc">
|
|
<p>
|
|
From Figure <a href="#org0bcb1fa">20</a>, we can see that a soft nano-hexapod \(k<10^6\ [N/m]\) significantly reduces the effect of perturbations from 20Hz to 300Hz.
|
|
</p>
|
|
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org5bb46c4" class="outline-2">
|
|
<h2 id="org5bb46c4"><span class="section-number-2">5</span> Closed Loop Budget Error</h2>
|
|
<div class="outline-text-2" id="text-5">
|
|
<p>
|
|
<a id="org6ed8184"></a>
|
|
</p>
|
|
<p>
|
|
From the total open-loop power spectral density of the sample’s motion error, we can estimate what is the required control bandwidth for the sample’s motion error to be reduced down to \(10nm\).
|
|
</p>
|
|
</div>
|
|
<div id="outline-container-org2cb5e9a" class="outline-3">
|
|
<h3 id="org2cb5e9a"><span class="section-number-3">5.1</span> Approximation of the effect of feedback on the motion error</h3>
|
|
<div class="outline-text-3" id="text-5-1">
|
|
<p>
|
|
Let’s consider Figure <a href="#orgd3c6de4">21</a> where a controller \(K\) is used to reduce the effect of the disturbance \(d\) on the position error \(y\).
|
|
</p>
|
|
|
|
|
|
<div id="orgd3c6de4" class="figure">
|
|
<p><img src="figs/effect_feedback_disturbance_diagram.png" alt="effect_feedback_disturbance_diagram.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 21: </span>Feedback System</p>
|
|
</div>
|
|
|
|
<p>
|
|
The reduction of the impact of \(d\) on \(y\) thanks to feedback is described by the following equation:
|
|
</p>
|
|
\begin{equation}
|
|
\frac{y}{d} = \frac{G_d}{1 + KG}
|
|
\end{equation}
|
|
<p>
|
|
The transfer functions corresponding to \(G_d\) are those identified in Section <a href="#org8db9681">2</a>.
|
|
</p>
|
|
|
|
|
|
<p>
|
|
As a first approximation, we can consider that the controller \(K\) is designed in such a way that the loop gain \(KG\) is a pure integrator:
|
|
\[ L_1(s) = K_1(s) G(s) = \frac{\omega_c}{s} \]
|
|
where \(\omega_c\) is the crossover frequency.
|
|
</p>
|
|
|
|
|
|
<p>
|
|
We may then consider another controller in such a way that the loop gain corresponds to a double integrator with a lead centered with the crossover frequency \(\omega_c\):
|
|
\[ L_2(s) = K_2(s) G(s) = \left( \frac{\omega_c}{s} \right)^2 \cdot \frac{1 + \frac{s}{\omega_c/2}}{1 + \frac{s}{2\omega_c}} \]
|
|
</p>
|
|
|
|
|
|
<p>
|
|
In the next section, we see how the power spectral density of \(y\) is reduced as a function of the control bandwidth \(\omega_c\).
|
|
This will help to determine what is the approximate control bandwidth required such that the rms value of \(y\) is below \(10nm\).
|
|
</p>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-org15ee341" class="outline-3">
|
|
<h3 id="org15ee341"><span class="section-number-3">5.2</span> Reduction thanks to feedback - Required bandwidth</h3>
|
|
<div class="outline-text-3" id="text-5-2">
|
|
<p>
|
|
Let’s first see how does the Cumulative Amplitude Spectrum of the sample’s motion error is modified by the control.
|
|
</p>
|
|
|
|
<p>
|
|
In Figure <a href="#org67a0258">22</a> is shown the Cumulative Amplitude Spectrum of the sample’s motion error in Open-Loop and in Closed Loop for several control bandwidth (from 1Hz to 200Hz) and 4 different nano-hexapod stiffnesses.
|
|
The controller used in this simulation is \(K_1\). The loop gain is then a pure integrator.
|
|
</p>
|
|
|
|
<p>
|
|
In Figure <a href="#orgdbc233d">23</a> is shown the expected RMS value of the sample’s position error as a function of the control bandwidth, both for \(K_1\) (left plot) and \(K_2\) (right plot).
|
|
As expected, it is shown that \(K_2\) performs better than \(K_1\).
|
|
This Figure tells us how much control bandwidth is required to attain a certain level of performance, and that for all the considered nano-hexapod stiffnesses.
|
|
</p>
|
|
|
|
<p>
|
|
The obtained required bandwidth (approximate upper and lower bounds) to obtained a sample’s motion error less than 10nm rms are gathered in Table <a href="#org16b406d">1</a>.
|
|
</p>
|
|
|
|
|
|
<div id="org67a0258" class="figure">
|
|
<p><img src="figs/opt_stiff_cas_closed_loop.png" alt="opt_stiff_cas_closed_loop.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 22: </span>Cumulative Amplitude Spectrum of the sample’s motion error in Open-Loop and in Closed Loop for several control bandwidth and 4 different nano-hexapod stiffnesses (<a href="./figs/opt_stiff_cas_closed_loop.png">png</a>, <a href="./figs/opt_stiff_cas_closed_loop.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
|
|
<div id="orgdbc233d" class="figure">
|
|
<p><img src="figs/opt_stiff_req_bandwidth_K1_K2.png" alt="opt_stiff_req_bandwidth_K1_K2.png" />
|
|
</p>
|
|
<p><span class="figure-number">Figure 23: </span>Expected RMS value of the sample’s motion error \(E_z\) as a function of the control bandwidth when using \(K_1\) and \(K_2\) (<a href="./figs/opt_stiff_req_bandwidth_K1_K2.png">png</a>, <a href="./figs/opt_stiff_req_bandwidth_K1_K2.pdf">pdf</a>)</p>
|
|
</div>
|
|
|
|
<table id="org16b406d" border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
|
|
<caption class="t-above"><span class="table-number">Table 1:</span> Approximate required control bandwidth such that the motion error is below \(10nm\)</caption>
|
|
|
|
<colgroup>
|
|
<col class="org-left" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
|
|
<col class="org-right" />
|
|
</colgroup>
|
|
<thead>
|
|
<tr>
|
|
<th scope="col" class="org-left">Nano-Hexapod stiffness [N/m]</th>
|
|
<th scope="col" class="org-right">10^3</th>
|
|
<th scope="col" class="org-right">10^4</th>
|
|
<th scope="col" class="org-right">10^5</th>
|
|
<th scope="col" class="org-right">10^6</th>
|
|
<th scope="col" class="org-right">10^7</th>
|
|
<th scope="col" class="org-right">10^8</th>
|
|
<th scope="col" class="org-right">10^9</th>
|
|
</tr>
|
|
</thead>
|
|
<tbody>
|
|
<tr>
|
|
<td class="org-left">Required wc with L1 [Hz]</td>
|
|
<td class="org-right">152</td>
|
|
<td class="org-right">305</td>
|
|
<td class="org-right">1000</td>
|
|
<td class="org-right">870</td>
|
|
<td class="org-right">933</td>
|
|
<td class="org-right">870</td>
|
|
<td class="org-right">870</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="org-left">Required wc with L2 [Hz]</td>
|
|
<td class="org-right">57</td>
|
|
<td class="org-right">66</td>
|
|
<td class="org-right">152</td>
|
|
<td class="org-right">152</td>
|
|
<td class="org-right">248</td>
|
|
<td class="org-right">266</td>
|
|
<td class="org-right">248</td>
|
|
</tr>
|
|
</tbody>
|
|
</table>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<div id="outline-container-orgf3b0566" class="outline-2">
|
|
<h2 id="orgf3b0566"><span class="section-number-2">6</span> Conclusion</h2>
|
|
<div class="outline-text-2" id="text-6">
|
|
<div class="important" id="org692b84e">
|
|
<p>
|
|
From Figure <a href="#orgdbc233d">23</a> and Table <a href="#org16b406d">1</a>, we can clearly see three different results depending on the nano-hexapod stiffness:
|
|
</p>
|
|
<ul class="org-ul">
|
|
<li>For a soft nano-hexapod (\(k < 10^4\ [N/m]\)), the required bandwidth is \(\omega_c \approx 50-100\ Hz\)</li>
|
|
<li>For a nano-hexapods with \(10^5 < k < 10^6\ [N/m]\), the required bandwidth is \(\omega_c \approx 150-300\ Hz\)</li>
|
|
<li>For a stiff nano-hexapods (\(k > 10^7\ [N/m]\)), the required bandwidth is \(\omega_c \approx 250-500\ Hz\)</li>
|
|
</ul>
|
|
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
<div id="postamble" class="status">
|
|
<p class="author">Author: Dehaeze Thomas</p>
|
|
<p class="date">Created: 2021-02-20 sam. 23:08</p>
|
|
</div>
|
|
</body>
|
|
</html>
|