Add pictures

This commit is contained in:
Thomas Dehaeze 2021-02-12 18:26:24 +01:00
parent 9a06c0995f
commit 195c5b5598
6 changed files with 128 additions and 96 deletions

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.4 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 236 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 268 KiB

View File

@ -3,7 +3,7 @@
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
<!-- 2021-02-11 jeu. 15:21 -->
<!-- 2021-02-12 ven. 18:26 -->
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<title>Encoder Renishaw Vionic - Test Bench</title>
<meta name="generator" content="Org mode" />
@ -39,21 +39,21 @@
<h2>Table of Contents</h2>
<div id="text-table-of-contents">
<ul>
<li><a href="#orgacaf822">1. Expected Performances</a></li>
<li><a href="#orgd1b48b9">2. Encoder Model</a></li>
<li><a href="#org9947f0d">3. Noise Measurement</a>
<li><a href="#orgfa3d11e">1. Expected Performances</a></li>
<li><a href="#orgf23b21b">2. Encoder Model</a></li>
<li><a href="#org9c17913">3. Noise Measurement</a>
<ul>
<li><a href="#org7dd6ce0">3.1. Test Bench</a></li>
<li><a href="#orgd61ad80">3.2. Thermal drifts</a></li>
<li><a href="#org8f23c76">3.3. Time Domain signals</a></li>
<li><a href="#orgbd6cefe">3.4. Noise Spectral Density</a></li>
<li><a href="#orgc14197f">3.5. Noise Model</a></li>
<li><a href="#orgb9429ef">3.1. Test Bench</a></li>
<li><a href="#orgd9c9c77">3.2. Thermal drifts</a></li>
<li><a href="#org8ec1ba2">3.3. Time Domain signals</a></li>
<li><a href="#org833451c">3.4. Noise Spectral Density</a></li>
<li><a href="#org71a7d07">3.5. Noise Model</a></li>
</ul>
</li>
<li><a href="#orgbc58807">4. Linearity Measurement</a>
<li><a href="#org61522ff">4. Linearity Measurement</a>
<ul>
<li><a href="#org38d4317">4.1. Test Bench</a></li>
<li><a href="#org9a6927b">4.2. Results</a></li>
<li><a href="#orge455e25">4.1. Test Bench</a></li>
<li><a href="#orgc6e5044">4.2. Results</a></li>
</ul>
</li>
</ul>
@ -63,7 +63,7 @@
<p>This report is also available as a <a href="./test-bench-vionic.pdf">pdf</a>.</p>
<hr>
<div class="note" id="org34d0504">
<div class="note" id="orge01a92a">
<p>
You can find below the documentation of:
</p>
@ -88,25 +88,25 @@ In particular, we would like to measure:
This document is structured as follow:
</p>
<ul class="org-ul">
<li>Section <a href="#orgafe2cb7">1</a>: the expected performance of the Vionic encoder system are described</li>
<li>Section <a href="#org1d1f36e">2</a>: a simple model of the encoder is developed</li>
<li>Section <a href="#orgf70a154">3</a>: the noise of the encoder is measured and a model of the noise is identified</li>
<li>Section <a href="#org3767bd5">4</a>: the linearity of the sensor is estimated</li>
<li>Section <a href="#org5ddac7d">1</a>: the expected performance of the Vionic encoder system are described</li>
<li>Section <a href="#org55cdc69">2</a>: a simple model of the encoder is developed</li>
<li>Section <a href="#orgb828c8d">3</a>: the noise of the encoder is measured and a model of the noise is identified</li>
<li>Section <a href="#org49975c3">4</a>: the linearity of the sensor is estimated</li>
</ul>
<div id="outline-container-orgacaf822" class="outline-2">
<h2 id="orgacaf822"><span class="section-number-2">1</span> Expected Performances</h2>
<div id="outline-container-orgfa3d11e" class="outline-2">
<h2 id="orgfa3d11e"><span class="section-number-2">1</span> Expected Performances</h2>
<div class="outline-text-2" id="text-1">
<p>
<a id="orgafe2cb7"></a>
<a id="org5ddac7d"></a>
</p>
<p>
The Vionic encoder is shown in Figure <a href="#org300cb52">1</a>.
The Vionic encoder is shown in Figure <a href="#orga0ecb6c">1</a>.
</p>
<div id="org300cb52" class="figure">
<div id="orga0ecb6c" class="figure">
<p><img src="figs/encoder_vionic.png" alt="encoder_vionic.png" />
</p>
<p><span class="figure-number">Figure 1: </span>Picture of the Vionic Encoder</p>
@ -134,21 +134,21 @@ Interpolation is within the readhead, with fine resolution versions being furthe
</blockquote>
<p>
The expected interpolation errors (non-linearity) is shown in Figure <a href="#org74b94f4">2</a>.
The expected interpolation errors (non-linearity) is shown in Figure <a href="#orgc38e53f">2</a>.
</p>
<div id="org74b94f4" class="figure">
<div id="orgc38e53f" class="figure">
<p><img src="./figs/vionic_expected_noise.png" alt="vionic_expected_noise.png" />
</p>
<p><span class="figure-number">Figure 2: </span>Expected interpolation errors for the Vionic Encoder</p>
</div>
<p>
The characteristics as advertise in the manual as well as our specifications are shown in Table <a href="#org12ad600">1</a>.
The characteristics as advertise in the manual as well as our specifications are shown in Table <a href="#org091f419">1</a>.
</p>
<table id="org12ad600" border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<table id="org091f419" border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<caption class="t-above"><span class="table-number">Table 1:</span> Characteristics of the Vionic compared with the specifications</caption>
<colgroup>
@ -200,11 +200,11 @@ The characteristics as advertise in the manual as well as our specifications are
</div>
</div>
<div id="outline-container-orgd1b48b9" class="outline-2">
<h2 id="orgd1b48b9"><span class="section-number-2">2</span> Encoder Model</h2>
<div id="outline-container-orgf23b21b" class="outline-2">
<h2 id="orgf23b21b"><span class="section-number-2">2</span> Encoder Model</h2>
<div class="outline-text-2" id="text-2">
<p>
<a id="org1d1f36e"></a>
<a id="org55cdc69"></a>
</p>
<p>
@ -217,65 +217,84 @@ It is also characterized by its measurement noise \(n\) that can be described by
</p>
<p>
The model of the encoder is shown in Figure <a href="#orge3dfe4a">3</a>.
The model of the encoder is shown in Figure <a href="#org4fdb73a">3</a>.
</p>
<div id="orge3dfe4a" class="figure">
<div id="org4fdb73a" class="figure">
<p><img src="figs/encoder-model-schematic.png" alt="encoder-model-schematic.png" />
</p>
<p><span class="figure-number">Figure 3: </span>Model of the Encoder</p>
</div>
<p>
We can also use a transfer function \(G_n(s)\) to shape a noise \(\tilde{n}\) with unity ASD as shown in Figure <a href="#org74b94f4">2</a>.
We can also use a transfer function \(G_n(s)\) to shape a noise \(\tilde{n}\) with unity ASD as shown in Figure <a href="#orgc38e53f">2</a>.
</p>
<div id="orgb259ef8" class="figure">
<div id="org793433f" class="figure">
<p><img src="figs/encoder-model-schematic-with-asd.png" alt="encoder-model-schematic-with-asd.png" />
</p>
</div>
</div>
</div>
<div id="outline-container-org9947f0d" class="outline-2">
<h2 id="org9947f0d"><span class="section-number-2">3</span> Noise Measurement</h2>
<div id="outline-container-org9c17913" class="outline-2">
<h2 id="org9c17913"><span class="section-number-2">3</span> Noise Measurement</h2>
<div class="outline-text-2" id="text-3">
<p>
<a id="orgf70a154"></a>
<a id="orgb828c8d"></a>
</p>
<p>
This part is structured as follow:
</p>
<ul class="org-ul">
<li>Section <a href="#org1bbddb3">3.1</a>: the measurement bench is described</li>
<li>Section <a href="#orge37ddeb">3.2</a>: long measurement is performed to estimate the low frequency drifts in the measurement</li>
<li>Section <a href="#orgbe1c0e1">3.3</a>: high frequency measurements are performed to estimate the high frequency noise</li>
<li>Section <a href="#orgfafa9fd">3.4</a>: the Spectral density of the measurement noise is estimated</li>
<li>Section <a href="#org2284feb">3.5</a>: finally, the measured noise is modeled</li>
<li>Section <a href="#org8cfb922">3.1</a>: the measurement bench is described</li>
<li>Section <a href="#orgfd5ce06">3.2</a>: long measurement is performed to estimate the low frequency drifts in the measurement</li>
<li>Section <a href="#org4df45c5">3.3</a>: high frequency measurements are performed to estimate the high frequency noise</li>
<li>Section <a href="#orgd464562">3.4</a>: the Spectral density of the measurement noise is estimated</li>
<li>Section <a href="#orgd6ec52a">3.5</a>: finally, the measured noise is modeled</li>
</ul>
</div>
<div id="outline-container-org7dd6ce0" class="outline-3">
<h3 id="org7dd6ce0"><span class="section-number-3">3.1</span> Test Bench</h3>
<div id="outline-container-orgb9429ef" class="outline-3">
<h3 id="orgb9429ef"><span class="section-number-3">3.1</span> Test Bench</h3>
<div class="outline-text-3" id="text-3-1">
<p>
<a id="org1bbddb3"></a>
<a id="org8cfb922"></a>
</p>
<p>
To measure the noise \(n\) of the encoder, one can rigidly fix the head and the ruler together such that no motion should be measured.
Then, the measured signal \(y_m\) corresponds to the noise \(n\).
</p>
<p>
The measurement bench is shown in Figures <a href="#org4037996">5</a> and <a href="#org06e2754">6</a>.
Note that the bench is then covered with a &ldquo;plastic bubble sheet&rdquo; in order to keep disturbances as small as possible.
</p>
<div id="org4037996" class="figure">
<p><img src="figs/IMG_20210211_170554.jpg" alt="IMG_20210211_170554.jpg" />
</p>
<p><span class="figure-number">Figure 5: </span>Top view picture of the measurement bench</p>
</div>
<div id="org06e2754" class="figure">
<p><img src="figs/IMG_20210211_170607.jpg" alt="IMG_20210211_170607.jpg" />
</p>
<p><span class="figure-number">Figure 6: </span>Side view picture of the measurement bench</p>
</div>
</div>
</div>
<div id="outline-container-orgd61ad80" class="outline-3">
<h3 id="orgd61ad80"><span class="section-number-3">3.2</span> Thermal drifts</h3>
<div id="outline-container-orgd9c9c77" class="outline-3">
<h3 id="orgd9c9c77"><span class="section-number-3">3.2</span> Thermal drifts</h3>
<div class="outline-text-3" id="text-3-2">
<p>
<a id="orge37ddeb"></a>
<a id="orgfd5ce06"></a>
Measured displacement were recording during approximately 40 hours with a sample frequency of 100Hz.
A first order low pass filter with a corner frequency of 1Hz
</p>
@ -286,13 +305,13 @@ A first order low pass filter with a corner frequency of 1Hz
</div>
<p>
The measured time domain data are shown in Figure <a href="#org55bfe2a">5</a>.
The measured time domain data are shown in Figure <a href="#org1454db4">7</a>.
</p>
<div id="org55bfe2a" class="figure">
<div id="org1454db4" class="figure">
<p><img src="figs/vionic_drifts_time.png" alt="vionic_drifts_time.png" />
</p>
<p><span class="figure-number">Figure 5: </span>Measured thermal drifts</p>
<p><span class="figure-number">Figure 7: </span>Measured thermal drifts</p>
</div>
<p>
@ -306,33 +325,33 @@ The mean drift is approximately 60.9 [nm/hour] or 1.0 [nm/min]
<p>
Comparison between the data and the linear fit is shown in Figure <a href="#org1085735">6</a>.
Comparison between the data and the linear fit is shown in Figure <a href="#orgfbe5f53">8</a>.
</p>
<div id="org1085735" class="figure">
<div id="orgfbe5f53" class="figure">
<p><img src="figs/vionic_drifts_linear_fit.png" alt="vionic_drifts_linear_fit.png" />
</p>
<p><span class="figure-number">Figure 6: </span>Measured drift and linear fit</p>
<p><span class="figure-number">Figure 8: </span>Measured drift and linear fit</p>
</div>
<p>
Let&rsquo;s now estimate the Power Spectral Density of the measured displacement.
The obtained low frequency ASD is shown in Figure <a href="#orgf2675d7">7</a>.
The obtained low frequency ASD is shown in Figure <a href="#org42f3fad">9</a>.
</p>
<div id="orgf2675d7" class="figure">
<div id="org42f3fad" class="figure">
<p><img src="figs/vionic_noise_asd_low_freq.png" alt="vionic_noise_asd_low_freq.png" />
</p>
<p><span class="figure-number">Figure 7: </span>Amplitude Spectral density of the measured displacement</p>
<p><span class="figure-number">Figure 9: </span>Amplitude Spectral density of the measured displacement</p>
</div>
</div>
</div>
<div id="outline-container-org8f23c76" class="outline-3">
<h3 id="org8f23c76"><span class="section-number-3">3.3</span> Time Domain signals</h3>
<div id="outline-container-org8ec1ba2" class="outline-3">
<h3 id="org8ec1ba2"><span class="section-number-3">3.3</span> Time Domain signals</h3>
<div class="outline-text-3" id="text-3-3">
<p>
<a id="orgbe1c0e1"></a>
<a id="org4df45c5"></a>
</p>
<p>
@ -340,67 +359,67 @@ Then, and for all the 7 encoders, we record the measured motion during 100s with
</p>
<p>
The raw measured data as well as the low pass filtered data (using a first order low pass filter with a cut-off at 10Hz) are shown in Figure <a href="#orgbd876dc">8</a>.
The raw measured data as well as the low pass filtered data (using a first order low pass filter with a cut-off at 10Hz) are shown in Figure <a href="#org28ad5da">10</a>.
</p>
<div id="orgbd876dc" class="figure">
<div id="org28ad5da" class="figure">
<p><img src="figs/vionic_noise_raw_lpf.png" alt="vionic_noise_raw_lpf.png" />
</p>
<p><span class="figure-number">Figure 8: </span>Time domain measurement (raw data and low pass filtered data with first order 10Hz LPF)</p>
<p><span class="figure-number">Figure 10: </span>Time domain measurement (raw data and low pass filtered data with first order 10Hz LPF)</p>
</div>
<p>
The time domain data for all the encoders are compared in Figure <a href="#org63a82cb">9</a>.
The time domain data for all the encoders are compared in Figure <a href="#org1656541">11</a>.
</p>
<p>
We can see some drifts that are in the order of few nm to 20nm per minute.
As shown in Section <a href="#orge37ddeb">3.2</a>, these drifts should diminish over time down to 1nm/min.
As shown in Section <a href="#orgfd5ce06">3.2</a>, these drifts should diminish over time down to 1nm/min.
</p>
<div id="org63a82cb" class="figure">
<div id="org1656541" class="figure">
<p><img src="figs/vionic_noise_time.png" alt="vionic_noise_time.png" />
</p>
<p><span class="figure-number">Figure 9: </span>Comparison of the time domain measurement</p>
<p><span class="figure-number">Figure 11: </span>Comparison of the time domain measurement</p>
</div>
</div>
</div>
<div id="outline-container-orgbd6cefe" class="outline-3">
<h3 id="orgbd6cefe"><span class="section-number-3">3.4</span> Noise Spectral Density</h3>
<div id="outline-container-org833451c" class="outline-3">
<h3 id="org833451c"><span class="section-number-3">3.4</span> Noise Spectral Density</h3>
<div class="outline-text-3" id="text-3-4">
<p>
<a id="orgfafa9fd"></a>
<a id="orgd464562"></a>
</p>
<p>
The amplitude spectral densities for all the encoder are computed and shown in Figure <a href="#org4b13cc6">10</a>.
The amplitude spectral densities for all the encoder are computed and shown in Figure <a href="#org7e93bb1">12</a>.
</p>
<div id="org4b13cc6" class="figure">
<div id="org7e93bb1" class="figure">
<p><img src="figs/vionic_noise_asd.png" alt="vionic_noise_asd.png" />
</p>
<p><span class="figure-number">Figure 10: </span>Amplitude Spectral Density of the measured signal</p>
<p><span class="figure-number">Figure 12: </span>Amplitude Spectral Density of the measured signal</p>
</div>
<p>
We can combine these measurements with the low frequency noise computed in Section <a href="#orge37ddeb">3.2</a>.
The obtained ASD is shown in Figure <a href="#orgec960f3">11</a>.
We can combine these measurements with the low frequency noise computed in Section <a href="#orgfd5ce06">3.2</a>.
The obtained ASD is shown in Figure <a href="#org7e54160">13</a>.
</p>
<div id="orgec960f3" class="figure">
<div id="org7e54160" class="figure">
<p><img src="figs/vionic_noise_asd_combined.png" alt="vionic_noise_asd_combined.png" />
</p>
<p><span class="figure-number">Figure 11: </span>Combined low frequency and high frequency noise measurements</p>
<p><span class="figure-number">Figure 13: </span>Combined low frequency and high frequency noise measurements</p>
</div>
</div>
</div>
<div id="outline-container-orgc14197f" class="outline-3">
<h3 id="orgc14197f"><span class="section-number-3">3.5</span> Noise Model</h3>
<div id="outline-container-org71a7d07" class="outline-3">
<h3 id="org71a7d07"><span class="section-number-3">3.5</span> Noise Model</h3>
<div class="outline-text-3" id="text-3-5">
<p>
<a id="org2284feb"></a>
<a id="orgd6ec52a"></a>
</p>
<p>
@ -412,18 +431,18 @@ Let&rsquo;s create a transfer function that approximate the measured noise of th
</div>
<p>
The amplitude of the transfer function and the measured ASD are shown in Figure <a href="#org904aecb">12</a>.
The amplitude of the transfer function and the measured ASD are shown in Figure <a href="#org5d39757">14</a>.
</p>
<div id="org904aecb" class="figure">
<div id="org5d39757" class="figure">
<p><img src="figs/vionic_noise_asd_model.png" alt="vionic_noise_asd_model.png" />
</p>
<p><span class="figure-number">Figure 12: </span>Measured ASD of the noise and modeled one</p>
<p><span class="figure-number">Figure 14: </span>Measured ASD of the noise and modeled one</p>
</div>
<p>
The cumulative amplitude spectrum is now computed and shown in Figure <a href="#orgff7d2cd">13</a>.
The cumulative amplitude spectrum is now computed and shown in Figure <a href="#org05b258c">15</a>.
</p>
<p>
@ -431,24 +450,24 @@ We can see that the Root Mean Square value of the measurement noise is \(\approx
</p>
<div id="orgff7d2cd" class="figure">
<div id="org05b258c" class="figure">
<p><img src="figs/vionic_noise_cas_model.png" alt="vionic_noise_cas_model.png" />
</p>
<p><span class="figure-number">Figure 13: </span>Meassured CAS of the noise and modeled one</p>
<p><span class="figure-number">Figure 15: </span>Meassured CAS of the noise and modeled one</p>
</div>
</div>
</div>
</div>
<div id="outline-container-orgbc58807" class="outline-2">
<h2 id="orgbc58807"><span class="section-number-2">4</span> Linearity Measurement</h2>
<div id="outline-container-org61522ff" class="outline-2">
<h2 id="org61522ff"><span class="section-number-2">4</span> Linearity Measurement</h2>
<div class="outline-text-2" id="text-4">
<p>
<a id="org3767bd5"></a>
<a id="org49975c3"></a>
</p>
</div>
<div id="outline-container-org38d4317" class="outline-3">
<h3 id="org38d4317"><span class="section-number-3">4.1</span> Test Bench</h3>
<div id="outline-container-orge455e25" class="outline-3">
<h3 id="orge455e25"><span class="section-number-3">4.1</span> Test Bench</h3>
<div class="outline-text-3" id="text-4-1">
<p>
In order to measure the linearity, we have to compare the measured displacement with a reference sensor with a known linearity.
@ -457,7 +476,7 @@ An actuator should also be there so impose a displacement.
</p>
<p>
One idea is to use the test-bench shown in Figure <a href="#org5a7f983">14</a>.
One idea is to use the test-bench shown in Figure <a href="#org177aa2c">16</a>.
</p>
<p>
@ -470,22 +489,22 @@ As the interferometer has a very large bandwidth, we should be able to estimate
</p>
<div id="org5a7f983" class="figure">
<div id="org177aa2c" class="figure">
<p><img src="figs/test_bench_encoder_calibration.png" alt="test_bench_encoder_calibration.png" />
</p>
<p><span class="figure-number">Figure 14: </span>Schematic of the test bench</p>
<p><span class="figure-number">Figure 16: </span>Schematic of the test bench</p>
</div>
</div>
</div>
<div id="outline-container-org9a6927b" class="outline-3">
<h3 id="org9a6927b"><span class="section-number-3">4.2</span> Results</h3>
<div id="outline-container-orgc6e5044" class="outline-3">
<h3 id="orgc6e5044"><span class="section-number-3">4.2</span> Results</h3>
</div>
</div>
</div>
<div id="postamble" class="status">
<p class="author">Author: Dehaeze Thomas</p>
<p class="date">Created: 2021-02-11 jeu. 15:21</p>
<p class="date">Created: 2021-02-12 ven. 18:26</p>
</div>
</body>
</html>

View File

@ -185,6 +185,19 @@ This part is structured as follow:
To measure the noise $n$ of the encoder, one can rigidly fix the head and the ruler together such that no motion should be measured.
Then, the measured signal $y_m$ corresponds to the noise $n$.
The measurement bench is shown in Figures [[fig:meas_bench_top_view]] and [[fig:meas_bench_side_view]].
Note that the bench is then covered with a "plastic bubble sheet" in order to keep disturbances as small as possible.
#+name: fig:meas_bench_top_view
#+caption: Top view picture of the measurement bench
#+attr_latex: :width 0.8\linewidth
[[file:figs/IMG_20210211_170554.jpg]]
#+name: fig:meas_bench_side_view
#+caption: Side view picture of the measurement bench
#+attr_latex: :width 0.8\linewidth
[[file:figs/IMG_20210211_170607.jpg]]
** Matlab Init :noexport:ignore:
#+begin_src matlab :tangle no :exports none :results silent :noweb yes :var current_dir=(file-name-directory buffer-file-name)
<<matlab-dir>>

Binary file not shown.