All this talk about getting rid of heat sounds scary. If I remember right,
these bus type frames, of which VME is only one, have a limited amount of
power available per card slot. If you exceed that, the other slots go
hungry, or you find yourself with a long-term power supply reliability
problem. To paint a blacker picture, most of the power supplies in these
boxes have specs that indicate they can deliver lots of power and they can
work at the resulting higher temperatures, but often they cannot do both
simultaneously!!
And a worse result is that all that input power turns into output heat. From
my experience heat is a serious problem in these boxes. It is a problem in
19" rackmount card cages in the first place to get the heat away from the
VME card, secondly to get it out of the VME box and in the third place to
get it out of the 19" rack (or where ever you put it).
A last thought: heat is often not a problem for today, it lets its presence
known in the future, so you end up with field failures: RELIABILITY.
Good luck with your investigations,
Ahne.
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of <Peter George Duncan>
Sent: Tuesday, January 22, 2002 20:51
To: [log in to unmask]
Subject: Re: [TN] VME stuff again...
Hi, Steve,
Sounds like you're having fun with this one, and I would be most interested
in being allowed to share your evolving experiences with trying to use
Stablcor. I flagged this material up to our designers when I first heard of
it (being simple souls, they'd designed our boards with good ol' FR4), but
we haven't had the excuse to look into using it yet.
The specs are now aging, and you're right about how densely packed the
boards have become, with folks trying to pack as much functionality into
each as humanly, or inhumanly, possible. I look at it as a cocoon within
which, and from which, any number of beautiful butterflies can grow and
emerge. (What DID that girl put in my coffee?) We have two boards on the go
that are a distillation of the functionality of five slots previously.
They've given me a folicular race between going grey or bald (it's neck and
neck at the moment).
The one thing we had to give serious priority to after the component
selection stage was layout to balance thermal mass around the board,
placing hotter components closest to the best thermal path through the
board, and particularly putting the hottest ones on the component side
where there was Max Headroom (remember him? I loved that guy). That way, if
the thermal path through the board is still not enough, you have some scope
for taking the heat out over the top with nice chunks of metal and thermal
pads. Pity there's no room for Pentium 4 style heatsinks here! There's no
room at all for overhead heat sinks on the solder side.
My sincerest sympathies!
Peter
Stephen
Gregory To: [log in to unmask]
<SteveZeva@AO cc: (bcc: DUNCAN Peter/Asst
Prin Engr/ST
L.COM> Aero/ST Group)
Sent by: Subject: Re: [TN] VME stuff
again...
TechNet
<[log in to unmask]
ORG>
01/23/02
12:19 PM
Please
respond to
"TechNet
E-Mail
Forum.";
Please
respond to
SteveZeva
Hi Peter!
The spec that you quoted, is what is being followed. I'm going to purchase
the spec so that we have it here...since it appears that we're getting into
the business of "ruggedizing" commercial VME assemblies.
>From a personal view, I think that VITA established the standards
prematurely without considering how the the technology was going to
evolove. As you know, VME assemblies are extremely dense, basically single
board computers.
The assembly that I'm looking at is double-sided SMT, fine pitch on both
sides, 0402's, 18+ BGA's on the top side (313 balls, the max I've seen so
far from the BOM).
Not sure if establishing these restrictions are a good thing, so that
everything will fit as established, or if it's better to stay fluid so that
you can adjust to adapt to changes in the technology...I suppose there's
arguments on both sides...
Just know now that we have an assembly that can take advantage of some new
substrate technology, that sounds very promising as far as heat
dissipation, and CTE issues, but can't use it because of the VME
spec...this is one of what came first, "The chicken or the egg?" huh?
-Steve Gregory-
Hi, Steve,
We ran into the same problem (more or less) - thermal management
requirements and solutions conflicted with the VME spec (which spec did
you
go with, by the way?).
One solution we considered, but didn't implement owing to expense, is to
"step" the PCB so that it's the right thickness at the card guide edges
and
at the P1/P2 connector areas. Is that possible with Stablcor? - I don't
know. Makes routing traces to the connectors a bit circuitous too (pun
intended). The board thickness outwith these areas is not important, as
long as it, plus the component height, doesn't exceed the envelope max.
You'll need some pretty light-weight copper and very thin separation
layers
to fit 16 layers into the spec'd thickness. Good luck!
Peter
Stephen
Gregory To: [log in to unmask]
<SteveZeva@AO cc: (bcc: DUNCAN Peter/Asst
Prin Engr/ST
L.COM> Aero/ST Group)
Sent by: Subject: [TN] VME stuff
again...
TechNet
<[log in to unmask]
ORG>
01/23/02
10:41 AM
Please
respond to
"TechNet
E-Mail
Forum.";
Please
respond to
SteveZeva
Hey all ya'll!
VME stuff again. Pretty strict standard. Gotta new technology out there
called Stablcor from Thermalworks, problem that we're seeing is that board
thickness is going to increase, which violates the VME spec's. Not just
from
the card-guide aspects, but will change the dimension where the P1 and P2
connectors interface with the backplane on a standard VME card cage.
Just wondering if anybody has anything out there (connectors, etc.), that
will allow you to "shoe-horn" ten-pounds, inna 1-pound bag, that will
comply
with VME spec's.
This board is 16-layers, within .063" without Stablecor, if we use
2-layers
Stablcor, it will increase .020-.025", with 3-layers Stablcor (the best,
as
far as thermal dissapation and CTE solutions), it will increase
.030-.035".
This Stablcor material sounds like a really good solution for thermal and
CTE
solutions, but has a hard time staying within the dimensional spec's that
are
called out in the VME spec's...trying to change these spec's I know is an
act
of God...it's there, and it's done.
Just wondering if anybody else has explored this material, and have come
to
a
dead-end like I have....because of the reasons like I have.
-Steve Gregory-
[This e-mail is confidential and may also be privileged. If you are not the
intended recipient, please delete it and notify us immediately; you should
not copy or use it for any purpose, nor disclose its contents to any other
person. Thank you.]
----------------------------------------------------------------------------
-----
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET
Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search previous postings at: www.ipc.org > On-Line Resources & Databases >
E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
ext.5315
----------------------------------------------------------------------------
-----
---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------
|
