Originally posted by XanthosNZ
That would be a horrible setup. First you'd be heating the thimble too. Second not all the heat would be directed into the thimble full of water you'd also heat up the wax of the candle and the air in the room.
Also no one in their right mind uses BTUs anymore. Joules (or perhaps kJ) makes much more sense. A unit based on a pound of water and a degree ou would end up with a smaller value as you lost heat that you can't accurately account for.
Well when I speak of BTU, I always convert to other formats, that
was just the first thing that popped up. Could have used calories
or Kilocalories too, once you have it in one regime its easy enough
to convert, just rolling numbers around. Also I was thinking about that,
I can see the problems with that setup, just a first thought
experiment. Its a matter of getting the heat available to the
device being measured. There was an interesting experiment
where the guy was not going for calories or joules or such but
convection currents and had a container closed off that was
1 mm^3, very small, filled with water and I think he did oil too.
He used a simple arangement of tiny TC gauges inside the liquid
and was able to graph the convection currents inside by graphing
the temperature changes. I did a similar experiment at work when
we had to test some photonics circuitry (Erbium doped waveguide
amplifiers) that had to have a heat stress test for 24 hours.
It was what we called the bomb. It was a pressure vessel capable of
withstanding over 10,000 PSI and got that by using DI water and
just heating it, and it had a TC probe holder, kind of like a test tube
sticking into the small chamber, a cylinder about 150 mm wide by
450 mm high, very thick SS. I was able to chart the internal
convection currents by charting the TC probe readings. It was
interesting to see, obviously liquid phase followed by gaseous phase
currents inside, we were not going for huge pressures, only about
2 atm., and about 110 degrees C. That was baby stuff for that device
though! It went on for 24 hours to 72 hours simulating lifetime checks.
Funny thing was, we thought the performance of the circuit would
have deteriorated but some of the devices got better! Go figure.