<!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>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
Richard Lacasse wrote:<br>
<blockquote type="cite" cite="mid434AD52B.6010704@nrao.edu">Randy,
Jason, et al., <br>
<br>
I've done a bit more thinking about power supplies for the CCB. I'd
like to run my thoughts by all of you and get whatever feedback you
deem appropriate. <br>
<br>
I think what would work best would be a system of two supplies and a
regulator mounted on a bracket just below the CCB. The two supplies
would include a Power One HN5-9/OVPA (5 volts at 9 amps and 6.6 lbs)
and an HC12-3.4-A (12 volts at 3.4 Amps and 4.25 lbs). The regulator
would be built around an LM338 5A linear regulator and would provide 8
volts at 2 amps to the linear circuitry. <br>
<br>
The following considerations drive me to this configuration
electrically. The embedded computer is quite sensitive to the power
supply level. Having the 5-volt supply mounted very close by, with
short wires will minimize the IR drop. It eliminates the need for
remote sensing. With regards to the 8-volt supply, since it is not an
industry standard voltage, we seem to have a choice between using an
adjustable lab supply or adding a regulator to an existing supply.
Post-regulating the 12-volt supply down to 8 volts does not seem too
unreasonable except for the wasted power, but hey this is analog! One
concern is the isolation of the analog supply from any noise generated
by the fan. We should look at this. One thing that goes in our favor
is the isolation gained from two regulators, one in the power supply
chassis which regulates 12 volts down to 8 and a second in the CCB
chassis which regulates 8 volts down to 5. Another advantage of this
approach is RFI. We have used Power-One supplies in the past with no
problems I am aware of. If we were to buy an adjustable lab supply the
RFI qualification could be a problem (not to mention extra weight). <br>
<br>
There are some mechanical considerations we need to think through. A
power supply chassis like I'm proposing about would add about 15 pounds
to the receiver package. Does this have any negative consequences,
e.g. pointing? I don't think it makes a whole lot of difference to the
moment generated by the supply weight, as compared to the traditional
power supply plates. The supplies are mounted at about the same radius
either way. We would need to come up with an additional bracket to
support the supplies below the CCB. I don't think this would be very
hard. Access to the supplies would be a little better with them
mounted below the CCB. Installation would be easier. <br>
<br>
Clearly, we need to get some decisions made fairly soon. These
supplies are not that expensive. The 5-volt/9 amp for instance is $78
in stock at Digikey. My feeling is that we ought to just go ahead and
buy the supplies, design and build the bracket and give it a try. If
we have to backtrack, it will not have cost us very much. What does
everybody else think? <br>
<br>
Rich <br>
<br>
_______________________________________________ <br>
gb-ccb mailing list <br>
<a class="moz-txt-link-abbreviated"
href="mailto:gb-ccb@listmgr.cv.nrao.edu">gb-ccb@listmgr.cv.nrao.edu</a>
<br>
<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>
<br>
<br>
</blockquote>
Rich (& All),<br>
<br>
Your approach sounds reasonable to me, but I'd like to add the
following comments / questions...<br>
<br>
1) I think, regardless of which supplies we select, there must be
provisions for some <u>large</u><br>
capacitors to deal with the surge currents which occur during
initial power-up of the CCB.<br>
I'm thinking that these need to be on the order of 10,000uF
(or greater?) on each of the<br>
three supplies (+5V, +12V, & +8V); but we should give due
thought to the actual sizes,<br>
ESR's, etc. which would prove sufficient to handle the
(somewhat large) surge
currents.<br>
<br>
2) It would seem that mounting the power supply system immediately
below the CCB would<br>
make it fairly accessible; a situation which carries positive
implications regarding the location<br>
and wiring of connectors in support of the REMOTE RESET,
REMOTE POWER ON/OFF,<br>
and CONSOLE PORT features.<br>
<br>
3) Are you thinking that we would build and deploy two identical
(?) power supply systems...<br>
one for use with the CCB connected to the Ka Band Receiver and
a second for use with<br>
the CCB connected to the 3mm Receiver in the future (not to
mention test fixtures and/or<br>
spares)???<br>
<br>
Randy<br>
<br>
<br>
<br>
<br>
<br>
<br>
</body>
</html>