TECHNET Archives

1995

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Mike McNulty <[log in to unmask]>
Date:
Fri, 13 Oct 1995 10:32:35 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (61 lines)

Robert,

At this point, it sounds like you are confined to maximizing convective 
heat transfer, which means maximizing external surface area and maximum 
draft currents within the enclosure. The first order of business is to 
spread the heat away from the component body a fast as possible meaning
large copper pad areas with as many thermal vias tying to planes as possible. 
You didn't mention the package type or material but  a thermally conductive 
epoxy bonding the component case to the pwb might be helpful, otherwise heat
transfer to the pwb will be through the leads. EG&G offers a heat sink  which
solders directly to a heat spreading pad (to which the component is mounted).
This might be a good option if you have adequate clearance. Also, as you 
mentioned, large external copper areas serve as good convective surfaces. 

Try to position the hot components at the inlet air side of the pwb and refrain 
from placing large components, which can block airflow, in front of it. Position the inlets and exhaust openings in the enclosure to maximize the chimney effect across the hottest modules. Are there air filters on the enclosure which 
restrict air flow? Are items placed in the enclosure in a manner which promotes
optimal system cooling?  It's tough solving problems like this through an E-mail forum considering the limited information available but I hope some of this was useful. Good luck!

Mike

0000000000000000000000000000000000000000000000000000
Mike McNulty
Lockheed Martin GES
Moorestown NJ
Phone:(609)722-7323
Fax:(609)722-2207
[log in to unmask]
0000000000000000000000000000000000000000000000000000


I need some help for a colleague on thermal dissapation.
 A design with one mother board and three daughter boards, ALL 
 with 10 lbs in a 5 lb container. The daughter boards are all 
 within a half inch of motherboard. Needless to say when 
 running inside the box after awhile things tend to get
 somewhat hot approx 125 degrees C. If outside in open air the
 temp drop is only about 15 degrees C. They do not want to use
 fans and conventional heatsinks seem out. A thermal plane on both
 outside layers is being considered realizing that although you 
 may be able to solder the board once rework is out. Is there
 any suggestions out there in the board material. I suggested 
 liquid cooling ha ha. I don,t now of any other means of 
 bringing the temp down that much without fans. Any help 
 would be greatly appreciated. The mechanical designer asking
 me this is supposed to do something with the enclosure, but no 
 fans are being considered. He has no other ideas either. 
 Thanks up front. I could not believe anyone would try and cram
 that much electronics in such a small area with no thought for 
 heat untill after the design was done. The desity is double sided
 surface mount with thru hole jammed on all four boards.
 P.S. I could really use this help ASAP so E-Mail or Phone would be great.
 
 Robert M. Wolfe
 203-882-6405
 
 



ATOM RSS1 RSS2