PhysicsProf Steven E. Jones circuit shows 8x overunity ?

[MarkE]

I believe you said that you are using thermocouples.  A thermocouple presents a short circuit (milliOhms) to the data logger input.  The thermocouple signal voltage is on the order of 10uV/C.  If I am mistaken and you are using a thermistor then the sensor could present typically 2K - 10K Ohms at room temperature, and you can substitute an 1/8 W resistor of similar value to get your noise floor.

If you are no convinced then as an alternate:  Feed the the thermocouple lead through the side of a styrofoam cup about 1/4 the way up from the bottom so that the sensor is suspended in air inside the cup. Cover the opening with tape.   Put a couple of steel 1/4" or 1/2" nuts in the bottom of the cup as weights.  You can use a paper cup in a pinch. Stack another cup inverted on top of the first cup and seal the edge with tape.  Hook up the sensor to your data acquisition system and wait half an hour.  Then capture 10s - 30s of readings.  That will be your noise floor, sensor included.

[nul-points]

 
no problem - i don't believe that i referred to the probe type in my previous posts


i found the high-precision resistor i was looking for at work (5KR) - i wanted something suitable to replace the thermistor sensor in the datalogger temperature probe

the graph included here shows the data for 2 channels on the datalogger: ambient temperature (degC) and an indication of the noise baseline on the temperature probe channel with a passive resistor connected

sampling-rate period is 5s, duration of data is approx 35 mins (Y-axis scale has been selected to give a similar range to previous graph)

[MarkE]

Great, so the noise floor of the electronics is about +/-0.025 ^oC, +/-0.045 ^oF.  I think the next step is to perform the styrofoam (or paper) cup test so that all the noise contributions are captured.  If it is a good thermistor, then you should not be able to see any increase in noise from the test that you just ran.

[nul-points]

Great, so the noise floor of the electronics is about +/-0.025 ^oC

No - not quite - we don't yet know the relative noise contributions of the electronics and the external resistor, separately

I think the next step is to perform the styrofoam (or paper) cup test so that all the noise contributions are captured

while we're waiting for that to happen....

what are some of your own personal goals, investigating novel energy sources, and what sort of challenges were you able to overcome in your last experiment in this area?

[MarkE]

No - not quite - we don't yet know the relative noise contributions of the electronics and the external resistor, separately


while we're waiting for that to happen....

what are some of your own personal goals, investigating novel energy sources, and what sort of challenges were you able to overcome in your last experiment in this area?

A thermistor bridge detector operates with a swing > 1mV/C.  the Johnson noise of the resistor is in the uV.

My goals are to see clean experiments.  I try to help in that direction where I can.