PhysicsProf Steven E. Jones circuit shows 8x overunity ?

[TinselKoala]

Ah... if your data is dealing with deltas of 0.2 degrees F... that kind of difference can be caused by tiny air currents or even ambient illumination from room lighting. I don't know the precise physical details of your setup, obviously, but we are down in the realm where experimental controls to rule out thermal artifacts become very difficult to implement properly. Would it be too much to suggest that you buy a couple of nesting styrofoam ice chests and make a nice "calorimetry qualified" enclosure for your future thermal measurements?

[nul-points]

 
lol - you'll have to come up with something more exotic than 'tiny air currents', TK!!

as i mentioned in a post above, the 0.2 degF temperature drop only relates to the new core (solenoidal) - the original toroidal core has produced approx 0.7 degF drop

when taking spot readings at night, i'm either using a flashlight in the unlit room - or only switching on light (low-wattage, cold-cathode fl. tube) for a few seconds whilst making next reading/change probe location -  - probe temperature allowed to settle for between 10-30 mins - i vacate room asap, to avoid adding body heat

device-under-test is in an enclosed region within a reasonably well-sealed, unheated room

the temperature deltas seem reasonably consistent regardless of day/night/ambient conditions

i will undoubtedly take more account of thermal isolation when i finish my other tests and start to focus instead on this temperature issue, but i will be leaving any more thorough treatment to those who are better equipped, with equipment, time, money and experience

no scientists are queing up to do my job for me, and i don't presume to do theirs - but we can all help in the process in a small way by calling out when we find something interesting

unless we get our kicks in some other way!

cheers
np

[MarkE]

One thing that would be nice would be if you would publish a null run measurement just to establish your noise baseline.

[TinselKoala]

While you may have eliminated the tiny air currents by your setup... good job so far by the way, usually I do not see or expect such care taken... there are still many other sources of artifact that might account for your near-noise-level data fluctuations.
One that has me very concerned with my own temperature measurements is the issue of thermocouple/instrumentation amp response sensitivity to ambient EM RF or even static magnetic or electric fields. It is a pain in the butt and is hard to track down. Readings may seem stable but are subtly off accuracy, or precision suffers, or total garbage comes from the probes. The fact that this can happen even when the source of the disturbing field is "off" or powered down is an incredibly frustrating feature of this kind of artifact in sensitive instrumentation operating near the noise floor.
It is in these situations especially that our own observer biases start having profound effects. We see fluctuating readings or we realize that some trials are garbage so we discard them. But we are more likely to discard trials or noisy data that don't support our hypotheses and keep trials that do, and this is completely unconscious on our part.
Paraphrasing Richard Feynman, the easiest person to fool is ourselves.
Or was that Pogo...

[nul-points]

i'm planning to check at work this week if i have a suitable value, hi-precision, low-noise resistor which i can substitute for the probe on the data-logger

i don't have any info about the input circuitry of the temperature channel and i'm not about to assume that i can safely short its input terminals just so that i can profile the noise-floor component of the datalogger for these informal observations

gentlemen - your interest is appreciated - your comments and suggestions are appropriate and relevant if and when any effect which i may be observing now is taken to the next level of serious inspection and measurement

any such more-controlled level of investigation won't be undertaken by someone with my level of resources - i shall be somewhere else, playing bass, accompanied by Dick (Feynman, RIP) on bongos - or is that on Pogos?

cheers
np