Maker Pro
Maker Pro

electrolytic caps in toaster oven

J

Jamie Morken

Jan 1, 1970
0
Hi all,

How well do SMT electrolytics (the silver can type) stand up to heat
during soldering in a toaster oven? I have had to rework a part on a
small board and it has been in the toaster oven 4 times now :) The
board is looking a little charred but everything still works, but I am a
bit concerned about the internals of the caps.. thanks,

cheers,
Jamie Morken
 
R

Robert Baer

Jan 1, 1970
0
Jamie said:
Hi all,

How well do SMT electrolytics (the silver can type) stand up to heat
during soldering in a toaster oven? I have had to rework a part on a
small board and it has been in the toaster oven 4 times now :) The
board is looking a little charred but everything still works, but I am a
bit concerned about the internals of the caps.. thanks,

cheers,
Jamie Morken

One could get the manufacturers spec sheet on the particular type and
model of capacitor in question.
That is to say, each manufacturer has their own manufacturing and
quality standards, and ther are different models within a give type.
For example, there are many models of aluminum capacitors, and each
one has their own temperature rating, recommended SMT temperature
profile, etc.
BUT.
The best way is to actually test some in the environment you use.
Take ten or more and mount them on a PCB with test leads (say common
ground (negative side) and seperate test points; maybe even guild the
lilly and put a diode and resistor in series to common charging voltage.
Heat them, cool them, charge them to rating, let them sit for 10
hours, discharge each one with resistor and test capacitance and ESR.
repeat 20 times or until 2-3 die.
Make a record of all readings, heat times, etc.

Ig you change brands, types or models, start again.
 
T

Terry Given

Jan 1, 1970
0
Robert said:
One could get the manufacturers spec sheet on the particular type and
model of capacitor in question.
That is to say, each manufacturer has their own manufacturing and
quality standards, and ther are different models within a give type.
For example, there are many models of aluminum capacitors, and each
one has their own temperature rating, recommended SMT temperature
profile, etc.
BUT.
The best way is to actually test some in the environment you use.
Take ten or more and mount them on a PCB with test leads (say common
ground (negative side) and seperate test points; maybe even guild the
lilly and put a diode and resistor in series to common charging voltage.
Heat them, cool them, charge them to rating, let them sit for 10
hours, discharge each one with resistor and test capacitance and ESR.
repeat 20 times or until 2-3 die.
Make a record of all readings, heat times, etc.

why only 20 times - repeat until dead. automated test rigs are great for
this. Its funny to see light switches being tested in a standards lab -
a test finger (literally) is used to turn the switch off and on again
with a known load, until the switch dies.

A have often obtained a production version of a psu, and set it up with
some timed relays to power up into a load, shut down, discharge all
caps, repeat in some suitably short time interval; then just left the
darn thing in a corner

The real problem with the test-and-see approach is of course the lack of
repeatability in terms of temperature, gradients etc. I suspect that
from job to job your toaster oven will have wildly different temperature
profiles (for example, I betcha it varies strongly with AC line voltage).

Ig you change brands, types or models, start again.

If you look at smt capacitor datasheets they ought to have reflow
profiles, which would be a good start. Its been a few years since I did
any significant work on capacitor lifetime, but as I recall there is an
upper temperature limit, beyond which the dielectric (usually a modified
carboxylic acid, but I'm no chemist) is buggered. I have seen a LOT of
cap data from Hitachi AIC (including, for example, the distribution of
cap lifetimes - they used -3 Sigma as their rated lifetime), and they
specifically mentioned the maximum tolerable hot-spot temperature.

rate-of-change of temperature (dT/dt) is usually quite important, to
minimise mechanical stresses due to CTE mismatch in materials. Probably
less of a concern with electro's than say ceramics.

Cheers
Terry
 
R

Robert Baer

Jan 1, 1970
0
Robert said:
One could get the manufacturers spec sheet on the particular type and
model of capacitor in question.
That is to say, each manufacturer has their own manufacturing and
quality standards, and ther are different models within a give type.
For example, there are many models of aluminum capacitors, and each
one has their own temperature rating, recommended SMT temperature
profile, etc.
BUT.
The best way is to actually test some in the environment you use.
Take ten or more and mount them on a PCB with test leads (say common
ground (negative side) and seperate test points; maybe even guild the
lilly and put a diode and resistor in series to common charging voltage.
Heat them, cool them, charge them to rating, let them sit for 10
hours, discharge each one with resistor and test capacitance and ESR.
repeat 20 times or until 2-3 die.
Make a record of all readings, heat times, etc.

Ig you change brands, types or models, start again.

I mentioned 20 times, because i think the caps will die long before
then.
 
Top