I am not sure just how it might be accomplished, but I think the best
experiment would be to melt or freeze a decent size sample with a
thermocouple imbedded and a recording indicator. I think there would be a
temperature "knee" or plateau to define the phase changes.

Don Young

"Tom Quackenbush" wrote in message
...
Specifically, how do you know when a lead alloy has reached it's
liquidus temperature?

On another newsgroup, a poster mentioned that the solder attaching
his LED to the PCB melted, therefore he knew that the temperature must
have reached 720 degrees F. I responded, asking what kind of solder he
was using, since 60/40 melts somewhere around 370 deg. F.

I know what the published values for the melting points of various
solders are, bear with me.

Original poster then says, I just measured it, my 60/40 solder melts
at 600 deg. F., what temperature does yours melt at and I don't want
to know what the charts say, I want to know what you measure.

Ok, I've got nothing better to do tonight and how can hard this be?
I've got the ingredients - various lead solders, toaster oven, decent
thermometer.

Problem #1 - my little toaster oven doesn't't like to reach 380. deg
F., let alone the 450 deg indicated on its dial. OK, I don't really
trust dials like that anyway.

Problem #2 - The big one. How do you know when your lead alloy
melts? What are the visual indications?


Here's a copy of the results that I posted on the other newsgroup
(references to the other poster are removed):

My sample melts at 600 degrees Farenheit. Yours? Dunno. Do the
test. But let's skip the bull****. Do the ****ing test and tell me
what you come back with. Don't give me a spec sheet result. Give me
a Real World result.

Let me begin by saying that I don't doubt the published figures for
the melting points of lead alloys. But, as you say, seeing is
believing, and I believe that personal observation is always valuable
(in more ways than one - you may find an error in the accepted values,
or you may learn something that your predecessors already learned).

OK, then. All degrees are Fahrenheit.

I simultaneously tested two samples each of 50/50 (solid, ~ .125
dia )and 60/40 rosin core (~.0625 dia.). One piece of each was laid
flat, one piece of each was bent into the shape of a coiled snake,
ready to strike (I'm thinking that it may be easier to detect "slump"
in the vertical section).

I'm using my toaster oven for heating. The solder pieces are
resting on a piece of cast iron (about 3/16" thick) which rests on an
aluminum tray. The surface of the cast iron is a combination of
grease and burnt grease(carbon). I fashioned a cap from heavy aluminum
foil to shield the whole shebang from the upper heating element.

I'm using a Fluke 16 to measure the temperature. The thermocouple
is clipped to the center of the cast iron piece.

It took about 30 min to raise the temperature to 350 deg. From 350
degrees the temperature was increasing at (_very_ rough approximation)
1 degree per 10 seconds. At 370 deg., the temperature was increasing
at less than 1 deg. per minute. At about 372 deg., I plugged the gaps
around the glass door with aluminum foil, and the temperature began
increasing at a tolerable rate (roughly 1 degree in 30 seconds), until
it reached another plateau at about 381 degrees. BTW, the temperature
dial on my toaster oven goes up to 450 degrees, then broil. Even at
the "broil" setting, I'm unable to reach 390 degrees in this thing.

The results? Somewhat inconclusive:
(I've decided not to follow your example and won't bother listing
temperatures where nothing changed)

Deg. F.

364 - 60/40 Vertical sample begins to slump
366 - 60/40 Vertical sample collapsed (vertical portion fell across
horizontal portion)
368 - 50/50 Vertical sample slumped and collapsed (vertical portion
did not fall across the horizontal portion)
369 - 60/40 Vertical - where the vertical portion fell across the
horizontal portion, the intersection has now fused into a smooth
lump.
371 - All undisturbed portions of samples (both 50/50 and 60/40)
have developed a sheen and appear distorted (elliptical rather than
circular cross-section).
377 - 50/50 samples clearly deformed & flowing downhill. 60/40
appears much as it did at 371 degrees.
382 - I think my toaster oven has peaked out. Solder samples appear
the same as at 377 degrees.

After waiting 45 minutes of steady 381 to 382 deg. temperatures I
stopped the experiment. I opened the oven door and dragged a steak
knife through all the samples. The effect was similar to that as when
you drag your finger through molten candle wax - liquid, but
solidifying very quickly.

Conclusions:

1. Passed solidus temperature for both alloys at 368 degrees.
References claim solidus temperature is 361 degrees. I assume that the
discrepancy lies in my crude setup.

2. I've realized that I'm not sure how to identify when the alloy has
reached its liquidus temperature. The cylinders (solder clippings)
never pooled into a ball, but does that mean that they didn't reach
their liquidus temperature or that they _did_ reach their liquidus
temperature and surface tension caused them to retain an approximation
of their original shape (anyone that has soldered knows something of
the effects of surface tension on solder).

3. An oven that reach 500 degrees would be nice.

R,
Tom Q.

p.s. I'll be asking on r.c.m for advice.