<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
HI Tom,<br>
<br>
OK, limit set at 2MB.<br>
<br>
Please try it again.<br>
<br>
73,<br>
Scotty WA2DFI<br>
<br>
<div class="moz-cite-prefix">On 2019-04-24 12:17, Tom McDermott via
TangerineSDR wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CACO3nRR9qArWh6N6s5APS06OtWELWC_YaiaCR09+CXzaO3eciw@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div><br>
</div>
<div>Apparently email with the in-line figures does not get
processed correctly.</div>
<div>This is an attempt to handle them as an attachment. The
zip attachment has</div>
<div>three figures. 223 Kbytes.<br>
</div>
<div><br>
</div>
<div>
<div>I measured the Allan Deviation of the Silicon Labs
de-jitterizer / synthesizer on the Silicon Labs<br>
</div>
<div>evaluation board. This part is a candidate for
generating the ADC clock and FPGA DDC NCO</div>
<div>from a GPS receiver. <br>
</div>
<div><br>
</div>
<div>Tests:<br>
</div>
<div><br>
</div>
<div>1. Allan Deviation vs. de-jitterizer bandwidth. The ADEV
degrades a lot for loop bandwidths</div>
<div>less than 100 Hertz (as the datasheet cautions).<br>
</div>
<div><br>
</div>
<div>2. Allan Deviation vs. Synthesizer output frequency.<br>
</div>
<div><br>
</div>
<div>For most frequencies, the curves lay on the same line at
10.0 MHz output (see chart attached).</div>
<div>I was able to find one frequency especially picked to
give the synthesizer a tough time, and the</div>
<div>Allan Deviation of that is substantially worse (see
chart). That frequency is 48.0 M * 31/149,</div>
<div>or 9.986571812 MHz. rounded to the nearest milliHertz.<br>
</div>
<div><br>
</div>
<div>My test setup: Rubidium oscillator at 10.00000 MHz
(undisciplined) feeding both the synthesizer</div>
<div> and the reference input to a TICC. TICC is thus
measuring the synthesizer output vs. the synthesizer input.<br>
</div>
<div><br>
</div>
<div>3. John Ackermann N8UR has measured the phase noise vs.
loop bandwidth, and there are some tradeoffs.</div>
<div>John will have to describe his test setup, I believe it
was done at 144.195 MHz.<br>
</div>
<div><br>
</div>
<div>-- Tom, N5EG</div>
<div><br>
</div>
<div>Attachments:</div>
<div><br>
</div>
<div> ADEV vs. Loop BW. The magenta line is 100 Hz loop
bandwidth (default value). All measurements at 10.00000000
MHz output.</div>
<div>ADEV vs. 10.00000000 and 9.986571812 MHz output</div>
<div>(from John Ackermann) Phase noise vs. loop BW.</div>
<div><br>
</div>
<div><br>
</div>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
</blockquote>
<br>
</body>
</html>