<br><br><div class="gmail_quote">On Thu, Mar 14, 2013 at 1:20 PM, Mike Monnier <span dir="ltr"><<a href="mailto:w8bac@comcast.net" target="_blank">w8bac@comcast.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
***** High Performance Software Defined Radio Discussion List *****<br>
<br>
<br><u></u>
<div style="PADDING-LEFT:10px;PADDING-RIGHT:10px;PADDING-TOP:15px" name="Compose message area">
<div><font face="Calibri"><font face="Times New Roman">My Apache Labs
Hermes/Anan-10 has, what looks like, frequency instability or<br>drift that
seems to be caused by temperature changes.<br></font></font></div></div></blockquote><div><br></div><div>Yes, you are right. I discovered the problem a month or so ago and worked with Phil and the folks at Apache Labs (with whom I am quite impressed and with whom I will happily continue doing business) about it. </div>
<div><br></div><div>First, yes, you will see what appears to be temperature instability on the internal 10MHz reference. If you look at the absolute numbers you will find that this "instability" is well within the 1ppm specification. The worst-case error I ever saw was just short of 2Hz at 10MHz, or 0.2ppm. Quite a bit better than spec -- but still annoying. My first inclination was the that there was a problem with the PLL that locks the 122.88 MHz clock to the 10MHz reference. That was not the case. The problem turns out to be the way that the 10MHz internal reference is temperature compensated. Apparently it is digitally compensated and therefore the corrections are quantized, hence the sudden jumps. Yes, it is annoying, especially if you are a frequency measurement geek like me. But that does not mean it doesn't meet spec or that it can't be used, even for applications like WSPR.</div>
<div><br></div><div>The other issue is the use of external reference input. </div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="PADDING-LEFT:10px;PADDING-RIGHT:10px;PADDING-TOP:15px" name="Compose message area">
<div><font face="Calibri"><font face="Times New Roman"><br>I use an external GPS
disciplined 10 MHz reference oscillator. The internal<br>Hermes reference
oscillator is not functional due to jumper selection. The </font></font></div>
<div><font face="Calibri"><font face="Times New Roman">GPSDO is turned on 24/7 and
all diagnostic software shows the unit is</font></font></div>
<div><font face="Calibri"><font face="Times New Roman">functioning flawlessly. It
is an HP 58503A. The disciplined 10 MHz </font></font></div>
<div><font face="Calibri"><font face="Times New Roman">oscillator is so close to
standard, the difference is measured in parts</font></font></div>
<div><font face="Calibri"><font face="Times New Roman">per billion. I am quite
sure the 10 MHz signal from the GPSDO is </font></font></div>
<div><font face="Calibri"><font face="Times New Roman">on the money. I peat this
with WWV and watch the phase. <br><br>I first noticed that the receiver
calibration was off each time I checked it so<br>I studied the
problem.</font></font><font face="Calibri"><font face="Times New Roman"> I
use WWV as my calibration standard. Calibration</font></font></div>
<div><font face="Calibri"><font face="Times New Roman">is off when cold
by .100 Hz roughly.</font></font><font face="Calibri"><font face="Times New Roman"> To begin I note the external
(skin)</font></font></div>
<div><font face="Calibri"><font face="Times New Roman">temperature of the case
using an infrared thermometer. The tests </font></font><font face="Calibri"><font face="Times New Roman">start at switch </font></font></div>
<div><font face="Calibri"><font face="Times New Roman">on, skin temperature 66
degrees F and a frequency correction factor of</font></font></div>
<div><font face="Calibri"><font face="Times New Roman">.9999977100000. I view
Phase 2 and adjust the dot rotation until it<br>slows to a crawl. Almost
instantly the calibration starts to shift.<br><br>After one hour the skin temp
rises to 92 F and correction factor is now<br>1.000003125.<br><br>After two
hours skin temp is 101 F and correction factor is 1.00000327.<br><br>At hour 3
and 4 the skin temperature stabilizes at 103 F and the correction<br>factor
stays steady at 1.00000327 nominal.<br><br>All tests taken with one receiver and
no transmit activity. Highest skin temp<br>area found on the right side of the
box facing the front.<br><br>I have a small fan for peace of mind when using a 4
watt carrier. No fan was<br>used during these tests. No air obstructions. No
heat source below the radio.<br><br>My questions are, what might be causing this
problem and can it be
corrected?<br>Thanks!<br><br>Mike</font><br><br><br></font></div></div>
<br>_______________________________________________<br>
HPSDR Discussion List<br>
To post msg: <a href="mailto:hpsdr@openhpsdr.org">hpsdr@openhpsdr.org</a><br>
Subscription help: <a href="http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org" target="_blank">http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org</a><br>
HPSDR web page: <a href="http://openhpsdr.org" target="_blank">http://openhpsdr.org</a><br>
Archives: <a href="http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/" target="_blank">http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/</a><br></blockquote></div><br><br clear="all"><div><br></div>-- <br>
Brian Lloyd, WB6RQN/J79BPL<br>3191 Western Dr.<br>Cameron Park, CA 95682<br><a href="mailto:brian@lloyd.com" target="_blank">brian@lloyd.com</a><br>+1.767.617.1365 (Dominica)<br>+1.916.877.5067 (USA)<br>