TECHNET Archives

September 1998

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:
Ralph Hersey <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Fri, 4 Sep 1998 13:04:45 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (93 lines)
Hi Scott--

There are several problems when using multiple plated-through holes for high current applications.

1 - for "simple" designs, as long as the copper's cross-sectional area of the via is >> than the
cross-sectional area of the ingress/egress conductors, then the design charts for thermal rise above
ambient are applicable.  This is because if the via's cross-sectional area is >> than the conductor's, then
the resistance of the via will be proportionally less than the conductor's, heating will be less, and the
end result will be the thermal rise will be less.

2 - if the cross-sectional area of the plated-through hole is the same as the conductor's, then there may
be a local increase due to the length of the via (board thickness) and this will result in a local increase
in temperature, so if this is the case then you should consider using the "internal" charts for doing the
design.

3 - if the "signal" currents are dc (direct current) or  low frequency (like power line) then the "routing
direction of the conductors ingress/egress directions becomes a  concern, especially with high pulsed
energy applications.  You need to do a series/parallel resistance/inductance network analysis to determine
where the maximum current will flow -- if by a "fluke" all currents in the plated-through holes are equal
(under all current applications), then you've got an acceptable design as the currents are evenly
distributed. An example of this would be a "parallel rail ladder" where the current enters one "rail" end,
passes through the "rungs" to the "other rail" and exits the network on the opposite end of the ladder from
where the current entered.  A worst case is when the currents enter and leave the ladder on the opposite
rails on the same end of the ladder.  In this example the "loop resistance/inductance" of the first ladder
rung (via) will carry the most current, the additional ladder rail resistance/inductance reduces the
current in the second rung/via, etc.  What can/will happen is that the first via will "blow-out", then the
next, etc.  This is not a desirable design feature.

One way to get out of it is to use a "star" like pattern, especially with mounting fasteners (such as
screws bolts, stud-mounted semiconductors, etc) with the fastener in the "center" and the via's located on
equal radii.

To my knowledge, there is no design data/requirements  for your application, if there is it's all held
"company proprietary".  We used the above for about 30+ years in my previous life's employment without any
known failures in applications of dc currents of <200 A, and pulsed currents of <2 kA with a few ns pulse
widths.

Bottom line -- treat is as a complex electrical interconnection network, and go for uniform voltage drops
and current distribution over the range of application frequencies.

Hope this helps, other wise contact me and I'll provide some additional information.

Ralph


Scott Holthausen wrote:

> Hey TechNeter's
>
> Does anybody have an equation that could help a person determine the
> current handling capability of through hole vias?  Our practice has
> normally been to use multiple vias to connect components with high current
> paths to plane layers.  However, that isn't good enough for one of our
> engineers who wants more than an approximation, and wishes to find out
> whether we are putting more vias on the board than we need to.
>
> Thanks,
> Scott Holthausen
>
> ################################################################
> TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
> ################################################################
> To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
> To subscribe:   SUBSCRIBE TechNet <your full name>
> To unsubscribe:   SIGNOFF TechNet
> ################################################################
> Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
> For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
> ################################################################



--
Ralph Hersey

Ralph Hersey & Associates
3885 Mills Way
Livermore, CA 94550-3319
PHN/FAX: 925.454.9805
e-mail: [log in to unmask]

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################


ATOM RSS1 RSS2