Experimental Evidence on Rossi Devices

This is indeed a complex subject, and these sometimes rough arguments are excellent for paring down possibilities.
These discussions did lead me to an idea, which may or may not work, but I'll give it a shot and see if it makes sense. The information available might just be too sloppy and calculated values might be too ... wrong.

Using the resistance figures calculated for each of the power examples given in the report, the coefficient of change of resistance over the temperature range might give a clue to the actual coil wire type, or the true temperature range. Maybe both, if done carefully.

I have also to add that I looked into Thomas' comments above in more detail. I found a chart of "withstand" ratings for copper wire. Actual testing has shown that a nick-free copper wire of 14 Ga will handle up to 795 A for 30 cycles (1/2 second) before melting ( 1080 C). How this translates to a higher resistance wire with a higher melting point would need some working out. The duty cycle would need to be very low if this were to be continued over an extended period of time, and the time to cool the wire between On periods would need to be considered. Actually doing this with off-the-shelf equipment is another story... I asked this of the Compact Fusion manufacturer, to see if this could be reasonably attempted in zero cross mode, or if it would trip an over current safety setting. I will report back on their answer, if they respond. They might think that I'm some sort of nut case and ignore me...

I have received a reply from the Compact Fusion manufacturers.
It is as follows:

We have reviewed your application.. Compact FUSION only goes up to 160Amps.
Our FUSION units extend as high as 1200Amps.
Your application will work with our FUSION units if you are able to wire your load in “inside-delta” or “wye” connection.

Here is our math:
Our power controllers are rated for continuous duty. The current ratings do not increase for a lower duty cycle.
Assuming 0.4 ohm load elements:

(380/0.4)*1.732=1645 amps during on period for a delta load.
We do not have a power controller big enough for this.

(380/0.4)=950amps during on period for an inside delta load.
Use a 1000A FUSION controller.
PN: FUSION-ZC-3-DDD0-0-0001-0000

(380/1.732)/0.4=548 amps during on period for wye load.
Use a 650A FUSION controller.
PN: FUSION-ZC-3-BBB0-0-0000-0000

Additional options:
SMAFUSION-RD15 (15’ remote display kit)

.....
www.ccipower.com

@Thomas Clarke,
Did you do a calculation for the temperature for the files 1- 5 in the Lugano report? I would be interested in what you think it might be.

This is not a trick question. It might lead to another (albeit weak) solution that corroborates your position if my very rough back-of envelope calculations are within a reasonable range or error(s).

To start you off with a bit more information, your comment about the specific heat earlier set me off on a set of rough calculations. The specific heat of the device can be worked out (to some sloppy degree), and the power required to reach a temperature, and increase the temperature based on the dummy run data can be ball-parked. Steady state heat temperatures/powers can be used to estimate minimum heat transfer... etc. What I am looking at is the ability of the device to reach a certain temperature purely electrically, assuming (for the moment at least) that no unknown reactions occur.

I have been looking at several charts of estimated amps required to heat a bare nichrome wire of various diameters to certain temperatures. Ignoring watts for now, it is clear that using a range of possible wire sizes, if the wire can be heated to around 450 C in open air with around 20 A, then using 50 A should easily heat the wire over 900 C. 15 Ga wire is close to ideal for this.

When considering Watts instead, (ignoring electrical resistance) using about 480 W to reach 450 C, things are much less friendly when calculating the higher temperatures possible at higher wattages, like 920 W. It seems like the device can barely exceed 700 C and would require a very long period to reach thermal equilibrium at that temperature.

Reconciling the watts, amps, and temperatures is extremely difficult.

Yikes what? I think I explained myself further down. The idea was to see if the currents reported had a physical solution that matched a reasonable range of normal wire sizes and at least the dummy temperature. And it does.

Using nichrome wire, amps tell you a lot. This wire type is very well characterized.

See: https://www.heatersplus.com/wire.html

This PDF is also very useful (1.3 Mb) : http://www.produktinfo.conrad.…T_ISACHROM_60_10METER.pdf

To explain further, my question to myself was: Can one easily build a heater using Inconel or other calibrated resistance wire that uses 20A, at any power level, and have it stabilize at 450C? Then, if it uses 20A to reach 450C, how hot could it get at 50A using the same wire? Will it only reach 780C? Could it reach 1100 C? Without any special considerations?

The answer is the wire should get very hot at 50A (or 46.9 A) if the 20A part is satisfied (or 19.7A, if you prefer). Probably much hotter than 780 C, without difficulty.

Can such a heater be made using the dimensions, mass, and suggestions from Rossi's patent with the power levels used in the report? Well, that is a bit more murky to be sure.

Anyways, continuing along with my journey that no one else cares about....
Experimenting and some handy calculators online have allowed me to configure a close approximation of the Lugano coil wiring.
(I'll make this as short as possible.)
Two twisted 15 Ga NiCr80 wires (closely approximating Inconel) can indeed be made to create a coil with 30 wraps around a 1 cm tube, with 55 cm of straight lead wires (that pass through the end caps, six leads combined for the 55 cm length) and arrive at the resistance most of us that have tried at this have come to: 1.23 Ohms total.

And so... (this should work.... http://www.steam-engine.org/coil.asp?a=true&mat=n80&p=roundmulti&tp=1.5&r=1.24&hfnw=2880&str=2&awg=15&wl=2065.571513534512&id=10&ll=550&ws=7 ) I had to shrink the wrap spacing to 4 mm to get the coil down to 20 cm.

Late edit: ... Something weird happened to the earlier image that was here. Hacked or something. But the above link should work with values inserted....

Then, we can see that the cross sectional area of the twisted wires is 3.3 mm^2 .This is nearly identical to a 12 Ga AWG wire cross section. Since these wires are all NiCr80, which is very similar to NiCr60, we can go and look up 12 Ga wire in the heat-current chart I posted a link to yesterday. This suggests that the Lugano dummy heat at (~450°C) is appropriate at 19.7 A, in a first approximation. Looking across the row, we can also see that this size of wire is expected to reach 871°C at (a very convenient) 46.7 A ,at equilibrium, in open air.

Of course these temperatures are for a bare wire in open air, so the heat transfer characteristics of wires wrapped around a ceramic tube with Ni and Li powder, and surrounded by a ceramic tube must be considered. Also, once a coil is wound, the wires can be heated to a the same temperature with less current than it takes for a straight wire. I suspect the the wires must be within a reasonable view factor of each other for that to occur, disregarding the ceramic considerations. I don't know if the calculator deals with that all.

Since these twisted wires have a larger diameter than I had figured they were earlier, then the wires are much closer to the inside of the outer tube than I had calculated earlier. I don't have any 15 Ga wire as yet, but I have been twisting up various other sizes to see how that affects the final diameter, and to see how many twists are a reasonable range (to obtain the twist spacing for the calculator). Two 18 Ga AWG wires tend to twist up to around 2.0 mm diameter. 2.9 mm, as calculated by the coil winding site (above), does seem about right for 15 Ga.

Those of us that have worked out the resistance for each row in the Lugano report, and have accepted the Wye to Delta hypothesis, may have noticed that the resistance still drops a small amount from the dummy to active runs. From roughly 0.412 Ohms (dummy) to around 0.375 Ohms, depending whether the corrections for including the R2 leads within the delta correctly (not done in the report) have been applied or not. (otherwise I think you get about 0.381 Ohms in the active runs).

It might interest some to know that the resistance coefficient of Nichrome alloy wires with the highest Ni contents (IE: Nichrome 80 and Nikrothal 70) have a small section of reduced coefficient between about 500°C and 1000°C. In other words, the resistance increase is not totally linear, there is a small dip in the curve. This seems to have something to do with surface loading of the wire and it's particular composition. The NiCr 80 resistance coefficient increases steadily until about 500°C, then drops slightly towards about 825°C, then slowly climbs up again towards the temperature limit of the wire. Nikrothal 70 has a higher temperature before dropping off (~600°C), and a slower overall rate of change towards higher temperatures.

This curve may be a useful indicator of the maximum temperature of that the Lugano coil wire could have reached, and also is a possible confirmation that the Inconel (or Nichrome) composition described in the report is probably at least 70% Ni, and is correctly described in the report, constraining the possible maximum temperature obtained by virtue of the melting point of these wires.

Edit: Image location hacked again... removed

Thanks, Thomas.
At the moment, I am simply pulling loose threads, to see if they quickly end in a knot, or if they unwind more than expected.

I am looking for corroborative evidence (even if flimsy) and a reasonable level of internal and external self-consistency.

One thing is fairly certain: there was a functioning heater. I intend to determine its performance by building it as exactly as possible.

Continuing along with the resistance topic for a moment...
We know that most materials demonstrate a positive increase in resistance over the temperature range from 20°C to 1300°C. This means that they can be used as a thermocouple, at least in a grossly relative sense. We don't have any information on the Lugano device (in the report) for the room temperature resistance. This makes things much more difficult. (The patent application does suggest a resistance: a 15 Ga wire with 2.65 Ohms/ft ... but that is one part of the information which I have been unable corroborate with the report details, calculations derived from the report, or external information. I have no good explanation for that as yet).

Anyways, the resistance should change relative to heat changes in a somewhat predictable way, within reason, even if the values are not quite exact. The resistance for each power level, compared to the other power levels in the Active Runs are self-consistent, suggesting that the calculations used to derive them are reasonable. The proportion of "wrongness" (error, uncertainty), if any, seems to be universally applied. The numbers might not be exact, but they show a consistent trend, regardless. This suggests that the methods used in the report were consistently applied to something real; even if errors were introduced in the report, they were applied universally. The numbers are unlikely to be made up (although they could have been generated by a formula of some type with a limited number of arbitrary inputs. Lets not go there for now.).

To summarize, the Joule Heat vs Power vs Resistance seems to hold together, so the calculated resistance should indicate something of the heat applied to the wire, regardless if the wire was heated electrically or by a reaction. Aside from ideas such as wires made of inobtanium being used, this should narrow down the choices of wire types. The resistances do not show a sharp rise with temperature, which limits the types of wire used and the temperature range . The electrical calculations I have done demonstrate that twisted Nichrome/Inconel wires of 15 Ga can be made to fit the electrical performance of both resistance, input power specifications, and physical dimensions of the Lugano device (albeit a mix of calculated and measured values from details in the report and selected patent application data). This further puts limits on the resistance vs temperature range that could be expected of real wires.

From this I conclude that a wire type similar to the composition of Nikrothal 70 was used, or at least can be used, as a reasonable substitute for the unknown wire type in a reasonable simulacrum of the device. This wire type can explain the rather weak increase in resistivity over the active range, and a drop from the resistance at the temperature of the dummy run. It is not exact, and there may be better choices. I would welcome alternate suggestions of wire choice. Particularly since 15 Ga Nikrothal 70 is a special order item, and I may need to buy considerably more than I want in order to get any at all.

I am starting to suspect that such a simulacrum could repeat the Dummy Run performance, Active Run input current and power levels, and reach about 900°C without any special reaction at all. I also think it would heat up rather slowly.

Which brings us to the thermal lag of about 400 seconds when the power was increased in the Active Run.... that is an interesting problem. It can be shown that approximately 100% of the input power increase would be needed to raise the temperature 120°C over 400 seconds (using the claims from the report). This neglects any heat loss (over and above that at the lower temperature) whatsoever during that time period.

@Antoine10FF
The Inconel data I had mentioned earlier can be found here. The file is big, so I found the link again. Lots of other neat stuff too.

(10 Mb PDF)

http://www.osti.gov/scitech/servlets/purl/1022709/

or this (not sure which link will work) https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=38&cad=rja&uact=8&ved=0ahUKEwiRt7200IrLAhVJ-GMKHTaSAUI4HhAWCDgwBw&url=https%3A%2F%2Ftheses.lib.vt.edu%2Ftheses%2Favailable%2Fetd-111997-13206%2Funrestricted%2Fch6.PDF&usg=AFQjCNF1vO5HmbdMfA0knxWSVU9MlVGGxA&bvm=bv.114733917,d.cGc