Posted by Morris Dovey on November 4, 2011, 5:22 pm
On 11/4/11 11:20 AM, CWatters wrote:
His output temperature versus time plot appears to indicate that his
table-top reactor ignited at about 60°C - and one is left to infer that
power was removed when/before the output temperature reached 100°C.
To figure the actual reactor operating temperature we'd need to know the
input temperature and flow rate, but it's safe to assume that it was in
excess of 100°C - it may have been double or triple that.
He had that little reactor wrapped in a heat exchanger - fiberglass wool
- foil - lead jacket, so it shouldn't have been too awfully lossy in
that temperature range.
Yes, that's what I expect - but I think it's possible to avoid having to
wait for the heating element to cool to the same temperature as the
reaction chamber. See below...
Absolutely. I think this can be streamlined somewhat by experimenting to
determine the heat transfer parameters; and once established it
shouldn't be necessary to wait for a temperature equilibrium to be
established - only for the heating element temperature to drop enough to
accurately measure dT/dt (rate of change in element temperature per unit
Posted by Jim Wilkins on November 5, 2011, 12:02 pm
I misunderstood your intention.
Heater resistance is the voltage across it divided by the current through
it. You can easily measure both while the power is on with a four-terminal
(Kelvin sense) current sense resistor. You wouldn't need a separate
resistance measuring circuit protected from full heater power.
The simpler approach puts the resistor in the return leg to reduce common
mode voltage. The heater voltage should also be sensed as close as practical
to both ends of the Nichrome.
There are also inexpensive isolated Hall-effect current sensors. I haven't
tested the samples I have and don't know it they are accurate enough to
measure the small resistance change in hot Nichrome.
Inductances in the heater circuit will give you false readings immediately
after current changes. I've measured the settling time for a machine that
performed automated tests on memory chips and saw low-level effects like
dielectric absorption in wiring and relay insulation that lasted up to 5mS.
The worst case was a transistor tester that needed 15 Seconds to settle
because the engineer neglected relay coil diodes.
Posted by daestrom on November 5, 2011, 7:01 pm
On 11/5/2011 8:02 AM, Jim Wilkins wrote:
In large commercial AC generators, that is exactly how they measure the
rotor winding temperature. Measure the applied voltage and resulting
current. As far back as the 1940's (maybe even earlier), companies like
GE and Westinghouse built a sort of analog computer that was just wired
into the circuit that applies power to the rotor slip-rings and provided
continuous rotor winding temperature indication on a servo-driven pen
Posted by eric gisse on October 31, 2011, 2:50 pm
Fusion is not self sustaining without immense, directed, energy input.
Posted by vaughn on October 30, 2011, 3:27 pm
I would urge you to reconsider that approach. If you use a standard temperature
probe (perhaps right inside that pipe with the heater), you not only gain the
advantage of using some standard "pre engineered" temperature sensing circuit,
you also can dispense with that relay and easily control the heater with a
standard SSR (Solid State Relay) The varying contact resistance of that relay
would always introduce some "noise" into your temperature readings..