Sounds promising- those rules are a lot more demanding (0.02" PCB width for 2 conductors) than my assumption, and I think I used a number for thermal conductivity that was higher.
AWG 36 phosphor bronze has a resistance of around 10 ohms per meter and about 1/6 the thermal conductivity of a similar diameter copper wire, so around the equivalent of AWG 48 copper wire.
Yup, without thinking, I made the mistake of using a several Watt film resistor in a circuit to limit ignition current when starting a mercury short-arc lamp with a spark gap. The film blew off the resistor on the first spark. Replacing it with a composition resistor fixed the problem.
Jon
Yes, I specifically used 1/4 W 10 Ohm carbon film resistors on my gas tube arrestors on the phone lines into my house. I have had to replace the resistors several times over the last 20 years when a large induced lightning current delivered a big spike. But, they have protected my modem, answering machine and PBX from any damage.
Jon
If you put traces on opposite sides of a thin flex substrate, the magnetic loop area will be so small it might as well be twisted pair.
John
cooled
=A0I
second
at
=46or electrical conductivity check this:
=46or thermal conductivity there is this:
Aluminum, copper, silver, and gold are the most conductive in both cases.
And for alloys such as brass, bronze, kovar, constantan etc., can be looked up just as easily.
?-)
om
rote:
a 5m-ohm
figure out
stor in
close*
5mm,edicted
he
the
e?
6.5PCB
t is
ate.
ts,
into
with
web
from
nto
on.
led
=A0I
cond
text -
ng
to
xt -
Product
=A0 =A0 =A0 =A0 =A0 649
0 =A0 =A0 =A0 =A0 =A0 =A0 650=A0 =A0 =A0 =A0 =A0 1200
=A0 =A0 =A0 =A0 =A0 =A0708
=A0 =A0 =A0 =A0 =A0 =A0 677
=A0 =A0 =A0 =A0 =A0 =A0 =A0735
67 =A0 =A0 =A0 =A0 =A0 =A0 =A0737=A0 =A0 =A0 =A0 =A0 =A0 1065
=A0 =A0 =A0 =A0 864
=A0 =A0 =A0 =A0 =A0 1056
Phosphor bronze 10.9 a bit
Here's a better link (maybe)
Interesting conclusion.. copper might be better than phosphor bronze. George H.
5m-ohm
figure out
at
For cryogenic applications it might be different--the thermal conductivity isn't anything like constant, so you have to use the thermal conductance integral instead of just the thermal conductivity times the temperature drop. The best source I know of for that is the Lakeshore catalogue.
Cheers
Phil Hobbs
com
a 5m-ohm
u figure out
istor in
close*
.5mm,
redicted
the
r the
te?
6.5e PCB
at is
rate.
ots,
t into
e with
er web
t from
into
son.
oled
. =A0I
econd
text -
ing
t to
xext -
=A0 Product
=A0 =A0 =A0 =A0 =A0 =A0 649
=A0200 =A0 =A0 =A0 =A0 =A0 =A0 650
=A0 =A0 =A0 =A0 =A0 =A0 1200
=A0 =A0 =A0 =A0 =A0 =A0 =A0708
=A0 =A0 =A0 =A0 =A0 =A0 677
=A035 =A0 =A0 =A0 =A0 =A0 =A0 =A0735
=A067 =A0 =A0 =A0 =A0 =A0 =A0 =A0737
5 =A0 =A0 =A0 =A0 =A0 =A0 1065r=A0 =A0 =A0 =A0 =A0 864
=A0 =A0 =A0 =A0 =A0 =A0 1056
t -
Yeah, the thermal conductivity of Copper increases for a while as you go below room temperature. The paper talks about tapered conductors, which sounds expensive, till you think about copper traces on flex or fiber glass.
George H.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required