Copper Traces as Ballast Resistors

[HarryD]

Tom Bruhns had an interesting article published in the 11.17.08 edition of
Electronic Design. His point was that the positive tempco of copper traces
was beneficial in design over fixed resistors. Not trying to reduce his fine
article but it is interesting to note that the 10mV drop that he depicted is
approaching the limit in trace voltage drop before heating of the trace's
center hotspot will be greater than 5 degrees centigrade and the PC boys
will do their Henny Penny renditions. Copper traces fail at 1000C and we
allow resistors to reach 150C but 5C rise in a trace will bring out a flock
of H.P. bird dogs. Most designers of PC traces, less than 2 inches, worry
about max current but max voltage drop is more indicative of trace heating.
Keeping all PC traces less than 12mV guarantees all trace hotspots will have
less than a 5C rise.
Cheers,
Harry

 

[Vladimir Vassilevsky]

Tom Bruhns had an interesting article published in the 11.17.08 edition
of Electronic Design. His point was that the positive tempco of copper
traces was beneficial in design over fixed resistors. Not trying to
reduce his fine article but it is interesting to note that the 10mV drop
that he depicted is approaching the limit in trace voltage drop before
heating of the trace's center hotspot will be greater than 5 degrees
centigrade and the PC boys will do their Henny Penny renditions. Copper
traces fail at 1000C and we allow resistors to reach 150C but 5C rise in
a trace will bring out a flock of H.P. bird dogs. Most designers of PC
traces, less than 2 inches, worry about max current but max voltage drop
is more indicative of trace heating. Keeping all PC traces less than
12mV guarantees all trace hotspots will have less than a 5C rise.

Once I did a mistake of using a PCB trace as a resistor for overcurrent
protection sensing. The problem is that the copper thickness can vary
dramatically from board to board. The copper weight spec guarantees that
there is no less copper then X ounces, but the upper limit is very inexact.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[legg]

Tom Bruhns had an interesting article published in the 11.17.08 edition of
Electronic Design. His point was that the positive tempco of copper traces
was beneficial in design over fixed resistors. Not trying to reduce his fine
article but it is interesting to note that the 10mV drop that he depicted is
approaching the limit in trace voltage drop before heating of the trace's
center hotspot will be greater than 5 degrees centigrade and the PC boys
will do their Henny Penny renditions. Copper traces fail at 1000C and we
allow resistors to reach 150C but 5C rise in a trace will bring out a flock
of H.P. bird dogs. Most designers of PC traces, less than 2 inches, worry
about max current but max voltage drop is more indicative of trace heating.
Keeping all PC traces less than 12mV guarantees all trace hotspots will have
less than a 5C rise.
Cheers,
Harry

It's the temperature limit of the base material that is of concern,
regardless of the heat source, as normal operating limits for
commercial printed wiring are listed in the pertinent safety docs.

Relating hotspot rise to mV trace drops is asking for trouble. Stick
to I^2.R and cm^2, like the good book says.

As the same surface area is required to dissipate the same power for
the same rise, regardless of the source, real estate advantages are
only achieved if unpopulatable board surfaces are employed in this
fashion.

Process variations in copper thickness may not be relevant if limiting
is only intended under fault conditions, where more extended
temperatures are permitted. You'd have to be aware of the circuit's
behaviour under the continuous worst tolerance condition and be
satisfied with the results before adopting the component cost
reduction method.

Henny Penny? These things are not always optional - look in the book,
if in doubt.

RL

 

[Eeyore]

Tom Bruhns had an interesting article published in the 11.17.08 edition of
Electronic Design. His point was that the positive tempco of copper traces
was beneficial in design over fixed resistors.

POSITIVE HOW ?

Graham

 

[Eeyore]

Once I did a mistake of using a PCB trace as a resistor for overcurrent
protection sensing. The problem is that the copper thickness can vary
dramatically from board to board. The copper weight spec guarantees that
there is no less copper then X ounces, but the upper limit is very inexact.

Not to mention 'tinning' in the conventional sense including Pb where allowed
by species who are anti-dendritic.

Graham

 

[Jasen Betts]

POSITIVE HOW ?

most metals have a slightly positive temperature coefficient of resistance
highschool physics.

 

[Eeyore]

most metals have a slightly positive temperature coefficient of resistance
highschool physics.

Yes I know that. In what WAY is it beneficial ? Resistors have tempcos too btw.

Graham

 

[Eeyore]

+3900 PPM/K for copper. Not a lot, actually, to have much ballasting
effect.

The idea was to help current sharing was it ?

Graham

 

[Ben Bradley]

POSITIVE HOW ?

A subsequent post shows you really meant to ask "BENEFICIAL HOW?"

Let's see what the article says:

[Ideas For Design]
Using Copper Traces As Ballast Resistors Ensures Balance

Tom Bruhns | ED Online ID #20023 | November 17, 2008
http://electronicdesign.com/Articles/Index.cfm?AD=1&ArticleID=20023

"The data sheet for the LT3080 linear voltage regulator suggests using
printed-circuit-board (PCB) traces for ballast resistors. Although the
LT3080’s low offset voltage suits it well for this technique, it can
be used for other ballast applications— for instance, for a set of
bipolar transistors."

There's a link to a circuit showing two voltage regulators with
outputs paralleled through 10 milliohm resistors. The main point of
the article seems to be:

"But there’s an advantage to using PCB traces that’s not mentioned in
the LT3080 data sheet. It involves the traces’ high temperature
coefficient.

"Using copper traces to sense and measure current may not be a good
idea, because the resistance has a temperature coefficient of almost
0.4% per °C, which causes nearly a 10% change over 25°C. Although
that’s a problem for precision measurement, it’s actually an advantage
in ballast resistors, because the resistor carrying the highest
current will show the largest self-heating (assuming the same
environment for all the ballast resistors). That self-heating will
increase the resistance, thus lowering the current in that branch
compared to a branch that doesn’t self-heat as much."

Well, that's interesting, and I could see where this would also be
advantageous in emitters of paralleled power BJT's where it would
compensate for current hogging caused by lower BE voltage in a hotter
transistor. It also mentions putting each ballast-resistor trace
physically (thermally) close to the active device driving it, so a
device providing more current, and thus heating up more, will cause
its respective ballast resistor value to increase, helping balance the
current between devices.

I agree with others, using copper traces in this way is at best
questionable. A good dose of solder on the trace (which I've seen from
wave-soldered boards with no solder mask - this was decades ago, but
still, one can't always dictate a manufacturing process) would lower
the resistance to make this idea useless, and cause the current
hogging it's trying to avoid. It seems there might be a market for
"copper resistors" or other low-value resistors with similar positive
tempco for use in such an application.

Also, as lifetime US citizen who considers himself well-read, I'd
never heard of Henny Penny before, but google shows this to be an
alternative name for the story and character named Chicken Little, who
kept saying "the sky is falling!" Then again, I never saw the
abbreviations +ve and -ve for positive and negative until I was on an
international email list and saw a Brit using them.

It all makes sense to me now.