<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
Richard Lacasse wrote:<br>
<blockquote type="cite"
cite="mid2778.10.12.97.224.1132598077.squirrel@webmail.gb.nrao.edu">
<blockquote type="cite">
<pre wrap=""><!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
</pre>
</blockquote>
<pre wrap=""><!---->Randy, Brian,
The polarity inversion on the cables does seem to explain why we got
readings in the neighborhood of 14000 on the telescope during our initial
tests. However, I seem to recall that the CCB readings seemed to vary
quite a bit from one day to the next, 14,000s one day, 10,000s another
etc. Also, there was a fair spread in the values. I'm having a hard time
believing that this could be caused by polarity inversion in the cables.
Maybe you've found only one of a few problems, or can you explain the
variations???
Rich
</pre>
<blockquote type="cite">
<pre wrap="">Brian Mason wrote:<br>
<blockquote type="cite"
cite="midPine.LNX.4.62.0511191724080.25874@mimas"><br>
Thanks Randy, Jason.
<br>
<br>
My last hot/cold in the lab, with ccb+receiver, was 19oct. At that
point I believe that we did *not* have the new cables between the
ccb+receiver-- the new cables were completed and installed only after
the Ka receiver had gone back up for spectral line commissioning. The
19oct levels showed the expected (4500 count vs cold, 12000 count vs
hot) power levels, and are summarized in a file called
ccbTsys19oct05.pdf that I think I sent to the mailing list the
following week.
<br>
<br>
Brian
<br>
<br>
--------------------------------------------------------------------
<br>
Brian
Mason
| office: +1(304)456-2338
<br>
Associate
Scientist
| fax: +1(304)456-2229
<br>
National Radio Astronomy Observatory | mail: PO Box 2
<a class="moz-txt-link-abbreviated"
href=<a
class="moz-txt-link-rfc2396E" href="mailto:bmason@gb.nrao.edu">"mailto:bmason@gb.nrao.edu"</a>><a
class="moz-txt-link-abbreviated" href="mailto:bmason@gb.nrao.edu">bmason@gb.nrao.edu</a></a>
| Green Bank, WV 24944
<br>
<a class="moz-txt-link-freetext"
href=<a
class="moz-txt-link-rfc2396E" href="http://www.gb.nrao.edu/%7Ebmason/">"http://www.gb.nrao.edu/~bmason/"</a>><a
class="moz-txt-link-freetext" href="http://www.gb.nrao.edu/%7Ebmason/">http://www.gb.nrao.edu/~bmason/</a></a>
|
<br>
--------------------------------------------------------------------
<br>
<br>
On Sat, 19 Nov 2005, Randy McCullough wrote:
<br>
<br>
<blockquote type="cite">All,
<br>
<br>
Jason and I spent some time today sorting out the proximate cause for
the
<br>
anomalous readings seen last week with the Ka Band Receiver/CCB setup.
<br>
<br>
As you'll recall, Brian reported that, when looking at the sky, he
would expect
<br>
to see total power counts of about 2,000 to 2,500; but, instead, all
channels
<br>
seem to be setting somewhere around 14,000 to 14,500 counts.
<br>
<br>
After going through many gyrations, Jason and I determined that the new
<br>
TWINAX cable sets had actually swapped the polarity of our differential
<br>
signals going into the CCB's input ports...
<br>
<br>
Since this would make our readings appear to be "upside down" (i.e.,
<br>
"full scale" would read "zero", while "zero" would read "full scale"),
this
<br>
could easily explain the anomaly which Brian reported...
<br>
<br>
What I DON'T understand though, is how we could have possibly missed
<br>
this while conducting our "final" lab tests... Brian, is it possible
that we failed
<br>
to conduct any hot/cold tests after we introduced the new cables? I
realize
<br>
that, by then, our focus had moved to sorting out phase switch
glitches, etc...
<br>
<br>
Anyway, come Monday morning, we'll adjust the polarity of the TWINAX
<br>
cables, check (and/or adjust) the calibration of the 16 amplifier
boards; then
<br>
run through a series of hot/cold tests to verify our changes...
<br>
<br>
INSTRUMENT STATUS...
<br>
<br>
At this point, the CCB is running in the Receiver Lab (as ccblab) with
no input
<br>
ports connected; although the receiver control cable is still attached.
<br>
The "demo_client" was disabled at the end of our testing session; so
the CCB
<br>
should now be available on the network...
<br>
<br>
Randy
<br>
<br>
<br>
<br>
<br>
<br>
<br>
</blockquote>
_______________________________________________
<br>
gb-ccb mailing list
<br>
<a class="moz-txt-link-abbreviated"
href=<a
class="moz-txt-link-rfc2396E" href="mailto:gb-ccb@listmgr.cv.nrao.edu">"mailto:gb-ccb@listmgr.cv.nrao.edu"</a>><a
class="moz-txt-link-abbreviated"
href="mailto:gb-ccb@listmgr.cv.nrao.edu">gb-ccb@listmgr.cv.nrao.edu</a></a>
<br>
<a class="moz-txt-link-freetext"
href=<a
class="moz-txt-link-rfc2396E"
href="http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb">"http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb"</a>><a
class="moz-txt-link-freetext"
href="http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb">http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb</a></a>
<br>
<br>
</blockquote>
Hi Brian,<br>
<br>
If this is truly the case, then NEARLY everything makes sense (i.e.,
it<br>
would explain the "upside-down" nature of our readings on the GBT).<br>
<br>
However, regardless of this "swapped" condition on the input signals,<br>
we STILL measured output voltages (while on the GBT) from all 16<br>
amplifier boards which were extremely widespread (as opposed to<br>
being reasonably tightly grouped as we would have expected). With<br>
this one point alone in mind, once the cables are adjusted, we need<br>
to look carefully at the performance of the system as a whole while<br>
still in the lab, check and/or adjust the calibration of the 16
amplifier<br>
boards, etc., etc., etc....<br>
<br>
Thanks,<br>
<br>
Randy<br>
<br>
<br>
</body>
</html>
_______________________________________________
gb-ccb mailing list
<a class="moz-txt-link-abbreviated"
href="mailto:gb-ccb@listmgr.cv.nrao.edu">gb-ccb@listmgr.cv.nrao.edu</a>
<a class="moz-txt-link-freetext"
href="http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb">http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb</a>
</pre>
</blockquote>
<pre wrap=""><!---->
_______________________________________________
gb-ccb mailing list
<a class="moz-txt-link-abbreviated"
href="mailto:gb-ccb@listmgr.cv.nrao.edu">gb-ccb@listmgr.cv.nrao.edu</a>
<a class="moz-txt-link-freetext"
href="http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb">http://listmgr.cv.nrao.edu/mailman/listinfo/gb-ccb</a>
</pre>
</blockquote>
Rich (& All),<br>
<br>
If you re-read my last post to this particular e-mail thread (see
above),<br>
you'll see that I share your concern that the inverted cables can
account<br>
for only PART of the questions raised by last week's testing on the GBT.<br>
<br>
This morning, just prior to Galen disassembling the Ka Band Receiver,<br>
I characterized the differential output voltages from all 16 amplifier
boards<br>
under a "hot load" in the lab. These voltages were fairly widespread
with<br>
the minimum being +0.333Vdiff and the maximum being +3.35Vdiff (which<br>
is, of course, well above the Full Scale output level).<br>
<br>
One thing that I DID learn Saturday is that bonding detector #1 to
Galen's<br>
temperature-controlled plate introduced a VERY small offset in
amplifier #1's<br>
output voltage. This shift in output voltage was readily adjusted back
out<br>
using the amplifier's "zero" pot (with plenty of room to spare); so I'm
not too<br>
worried about it at this time... Further, the amplifier's differential
output<br>
voltage exhibited no appreciable change in noise levels when the
detector<br>
was alternately bonded and floated... which is another good sign!<br>
<br>
Once Galen has the receiver back together, our plan is to re-examine
the
RF<br>
levels into all 16 detectors, re-calibrate all 16 amplifiers (this time
using CCB<br>
counts), then conduct a series of hot/cold tests, linearity tests, etc.
as Brian<br>
has described in his latest e-mail response...<br>
<br>
Randy<br>
</body>
</html>