Ben and Don;

Attached is a bit of dialogue I had with Ralph Hersey, a colleague involved
with my design committee. The discussion involves the interpretation of the
current carrying tables in IPC-D-275. In addition to the attached dialog,
Ralph is doing some experimentation with intent to help us revise the tables
for clarity and to better reflect current technology. However, the data
currently in the standard is good and works for many applications, as long
as one interprets it correctly.

Regards'

Gary Ferrari
Chairman - IPC-D-275

>Ralph;
>
>I believe that the table mentioned serves interelated purposes. The current
>through the conductor will create a rise in ambient temperature. So if one
>were looking to determine a conductor width, based on a specific current
>carrying requriement, he would use the maximum temperature rise as a
>limiting factor. On the onther hand if one were seeking to know the maximum
>temperature rise a current through a specific conductor creates, he can use
>the table to find the answer. 
>
>Another area of controversey is note three, which states that "For general
>use the permissible temperature rise is defined as the difference between
>the maximum safe operating temperature of the laminate and the maximum
>ambient temperature in the location where the panel will be used." I think
>that the note is concerned with the heat produced and it's effects on copper
>adhesion to the laminate material. What I believe it omits is a fuse effect
>of the copper conductor.
>
>Ralph, before I respond to Ben's message, What are your thoughts on the
>subject. Am I missing something or misinformed?
>
>Thanks,
>
>Gary
>---------------------------------------------------

>Return-Path: [log in to unmask]
>Date: 18 Apr 1996 09:05:53 -0700
>From: "Ralph Hersey" <[log in to unmask]>
>Subject: Re: Re[2]- heavy copper...
>To: "Gary Ferrari" <[log in to unmask]>
>
>        Reply to:   RE>>Re[2]: heavy copper...
>
>Gary,
>
>You and I understand the chart the same way.
>
>For thermal rise -vs- current carrying capacity both notes and subtitle are
>correct, it just depends on whether you want the cart before the horse or not.
>
>I've a general concern for both the Figures 3-4a and 3-4b, in that the graphs
>are based on "free air circulation", which we generally do not have around the
>traditional printed boards, though many of today's applications use forced air
>(fans).  As I mentioned previously, the graphs present "adjusted/compensated"
>thermal rise above ambient for vertically mounted printed boards in a test
>chamber, having four conductors on 25 mm center lines about a foot long with
>lots of space between the vertical surface of the printed board and the wall
>of the test chamber.  Current is applied to one conductor at a time and the
>base materials were single-sided metal clad in the range of 0.8-1.5 mm base
>material thickness.  Not exactly the type of designs most people are doing
>today (or even then).  The statement in Note 1 is a requirement for more/less
>conformance to requirements for the graphs to be more/less completely
>applicable; "It is assumed that, for normal design, the conductor surface area
>is a relatively small compared to the adjacent free panel area.", and we know
>this doesn't happen with today's component mounting densities.  Figure 3-4a is
>for "uncoated" conductors "thin" solder resist and conformal coatings will add
>a little to the thermal rise.  "Thick" polymeric coating will thermally
>insulate the conductor from air and increase the operating temperature of the
>conductor.
>
>You're right about the thermal limits of the base material (and polymeric
>coatings) the goal is to limit the printed board's maximum operating
>temperature so as not to destroy the base material with continuous operation
>of the assembly under those conditions.  The design/regulatory requirements is
>not to "burn up" the base material (or polymeric coatings), or have the
>conductors delaminate from the base material (keep them attached and [bonded]
>in place).  Some of the concerns (mine) are that the "ambient" temperature is
>no longer the air temperature, instead it is now the "board" temperature
>because we have a lot of components attached to the printed board that
>transfers some heat to the board.  In fact with thermal lands/vias we
>intentionally use the printed board as a cooling sink and thermal dissipator. 
>The additional heat from compoents, plus the increased component mounting
>density, plus polymeric coatings, plus significant variations in components
>heights does not result in the classic "textbook" for "free air circulation"
>for convective heat transfer.  This is why I'm a little nervous (cautious)
>about using Figure 3-4 as a "broad based" design reqirement when we
>significantly change (could be called undermining) the baseline requirements
>used to develop the data.
>
>My feeling is that we've been so conservative that we've lucked out (not be
>design --- by default).  One of the reasons is that we have other conductors
>(metals) that tend to spread out the heat and thereby somewhat lowers the
>thermal density and in a lot of applications we use "forced" air, which all
>helps.
>
>You're also right about the "fusing currents", they definately are not a
>design intent for Figure 3-4, instead, that's why the graphs are so "derated"
>so as to ensure (insure) that we never subject a printed board to a fusing
>current because the designer/manufacturer does not want to pay the insurance
>and replacement costs for somebody's home or other facility.
>
>Gary, I agree with your interpretations, provided some additional thoughts to
>support your arguements, and added some concerns about how free is "free air
>circulation".
>
>I think this has got-it for now -- Gary have a good one, Ralph
>
>--------------------------------------

>> ----------
>>>From: Walker, Don
>>>To: DesignerCouncil
>>>Subject: Re[2]: heavy copper...
>>>Date: Wednesday, April 17, 1996 9:55AM
>>>
>>>> ...you also have to know at what temperature (over ambient) that the 
>>board
>>>> will be running at...
>>>
>>>Remember, what the chart is saying is that if you apply a specified current
>>>to a specified trace size, the temperature of the trace will rise by a
>>>specified amount.
>>>
>>
>>Don, I've not heard this interpretation of the chart.  In IPC-D-275 Figure 
>>3-4 there seems to be a conflicting definition of the charts.  Above the 
>>first chart it says "For use in determining current carrying capacity....for 
>>various temperature rises above ambient".  We have always interpreted this 
>>to mean take the expected temp. rise above ambient of the board 
>>(environment) and select trace size.  However the notes for the Figure (Note 
>>1) indicates "The design chart has been prepared as an aid in estimating 
>>temperature rises (above ambient) vs. current for various cross-sectional 
>>areas of etched copper conductors."  This seems to support your 
>>interpretation, Don.  Admittedly I haven't been using the IPC chart, rather 
>>one that has been posted on the wall (a copy from the old Bishop Graphics 
>>Design Aid).  The old BG chart doesn't have the Note 1, but does have the 
>>first label above.
>>
>>So now I'm confused.  Does the chart indicate the rise of the trace 
>>temperature or does it indicate the rise above ambient of the environment? 
>> Perhaps someone from the IPC committee that worked on the 275 (Gary are you 
>>there?) could comment and/or elaborate on the charts.
>>
>>>> ...As an example, say I need 5amps and 20 degrees over ambient is a safe
>>>> assumption. The chart gives ABOUT 100 square mils, which corresponds to
>>>> ABOUT a 75mil trace width on the 1oz copper curve.
>>>
>>            <<<<<< cut >>>>>>>
>>>
>     ^^^^^^     [log in to unmask]       "It's a damn poor mind that can
>     |o  o|     voice: (512)251-2341          think of only one way to spell
>---000-()-000---fax:   (512)251-0498          a word."   --Andrew Jackson
>>>
>>>
>>
>>Ben Davis
>>[log in to unmask]>>
Regards,

Gary Ferrari
Tech Circuits
(203)269-3311
[log in to unmask]