Rossi Lugano/early demo's revisited. (technical)

  • So, how do you persuade a group of people who don't like the idea to review your work objectively?


    Scientists can (and must when reviewing) deal with process and factual matters, rather than beauty context or personal feelings. There is always some room for interpretation, some aspect of personal feelings. But, if this is science, rather than Eng Lit, positive or negative comments can be justified and that justification critiqued by third parties. Also, contrary to what seems to be a prevailing opinion in many places, most scientists are genuinely interested in the truth, and when given papers to review try to do so objectively: easier for them to do that than for somone whose life work is validated by a positive experiment and unrealised by a negative one.

  • Also, contrary to what seems to be a prevailing opinion in many places, most scientists are genuinely interested in the truth, and when given papers to review try to do so objectively


    That's as maybe. Probably true. But since we are not playing the 'peer-reviewed paper' game which I told you we would not, I was talking about experimental replication in an 'away' lab. For that to happen, as you might imagine, the away team usually needs a more than ordinary degree of interest in the first place, and to have seen things when playing at the home lab that intrigue them.

  • Paradigmnoia


    Your modified e in your Dummy spreadsheet due to IRM and View Factor seems OK to me in general.


    When viewing the output power distribution Rods%/Caps%/Ribs% the Rods seem (to me) to have a larger than expected proportion of the output for the Dummy cases, both yours and Lugano, relative to the electrical input for the respective segment. (I’ll post a little summary below in a moment)


    I already was wondering about that too and think that the length of the heater coil wire inside the rods could have been more then the 4 cm you suggested.

    For the dummy run simulation it would bring the temperatures of the ribbed area possibly a little bit down and maybe even more in agreement with the reported data but would boost significantly the power in the rods.

    A possible arangement would be a coil of 9 windings (same number of turns as shown in the IH patent application drawing) with 24 cm length (so somewhat extending under the end caps), 2 cm wire in the end caps and then 6 cm in the rods.


    As far as the output power distribution is concerned : For the two sets of rods, based on the data in the Lugano report I calculated about 118 watt while the Lugano report calculated about 130 watts.

    The dummy run recalc gave (see page 1 of the spreadsheet)


    Rods 118.38 Watt

    Caps --87.33 Watt

    Ribs --281.77 Watt


    The total power is then calculated as 487.48 Watt and the distribution becomes 24.3/17.9/57.8

    That is different from the 28.89/21.05/50.06 you are giving.


    EDIT : I now see that you compare the inflated case.

    Maybe you should add that to the text so that it is clear what data you are presenting


    What range of W/mK are you using for the Rods and Caps?

    I am assigning the powers to the heater sections in the model as per your suggestion for the power division.

    A section power is then evenly distributed over the length of the heater wire in the section.

    So I am not using W/mk values.


    Please could you make two more spreadsheets in the same form factor as the Dummy one, (I realize that this quite onerous), for both the recalculated lower temperature Run 16 and your COP 5 version of Run 16 ?


    I could do that, but you can see that the time between posting calculations is often 2 or more weeks due to the time I have available.

    So yes, I can do it and am prepared to do it , but don't expect a quick response. (Also the comming two days and the weekend I am not able to spend time on it)

    If you want to continue faster you could take for example the dummy run spread sheet and modify it yourself. (they are not protected and you are in my opinion capable enough to do it)

    Let me know what you prefer.


    Then it will easier to see what the simulation is doing. The Ribs can be treated as a single section for now.


    They must be treated as a single section since the Lugano report gives only accumulated data for the active runs, not the detailed data per section.

    And for the rods no average temperatures are given.

    So we have to find a solution for that if we want to include the rods in the calculations


    I am more interested in chasing why ~ 24% of the electrical input goes into the Rods and Caps, but ~ 50% goes out, when ~76 % of the input power goes into the thin Ribs section in the Dummy cases.


    See the comment above about the wire distribution.

  • Ignoring the cables (probably a big part of the problem) I get something like this for the Rods (Cap at the far left)...

    Assumes a constant 200 C for the far left Rod 5 cm segment, conduction from that point on to towards the RH side.

    Alumina with 36 W/mK used.

    Still fiddling with it a fair bit, though.


    EDIT: Image V2 looks better. Added the cables.

  • Started over and made this.

    451 C main tube (ribs area, but no ribs yet) constant T (input), Caps started at 325 C and left to run wherever they want after the start, and Rods not quite touching the Caps. The Caps continue heating for several hours, eventually getting not too much cooler than the Ribs area.

    (The simulation hot air movement effect is cool.)

    The second one is dialing the Ribs up to 1410 C, start Caps at 325 C, and see what happens.

    For both runs, the sole input is a constant Rib area temperature. The caps and Rods have extra heat from the wires in their respective areas not yet accounted for.

    Too bad the materials library link seems to be broken on the simulator.

  • Suggestion for the rods:


    Take a 3 cm diameter, 50 cm long ceramic pipe (or from another material which can stand the temperature).

    Stick the halve of a heating element (coil, power resister) in at one side.

    The part sticking out simulating the heated environment of the end cap.

    Put the power on.

    I think that after a short time it will be acting like a horizental flue pipe in that there will be hot air current flowing in the rod from the hot heater side to the other end which is cold.

    It would possibly explain why the rods are hotter near the cold side then would be expected from conduction only.

  • LDM ,

    This test you suggest is worthwhile.

    However, I suspect that cold air will flow into the unheated end and exit at the hot end as the air is subsequently heated and the resultant buoyancy lifts it away. (If the tube is horizontal).


    I have a large thermocouple protection tube that I can cut the closed end from to try out. Only 30 cm long, much thicker, and with a much smaller ID than the Lugano Rods, though.

  • Alan Smith wrote:

    Quote

    For that to happen, as you might imagine, the away team usually needs a more than ordinary degree of interest in the first place, and to have seen things when playing at the home lab that intrigue them.

    I don't know if they are still in active business but if so, Earthtech (Scott and Marissa Little and company) could do an objective job and I suspect, given strong evidence of positive results (Eout>Ein) they would be very interested.

    http://earthtech.org/maverickinventor/

  • Where my very basic modeling seems to have trouble is with just the hot end segments of the Rods set to the ‘right’ temperature, the far ends don’t get hot enough (stay at ambient), but when the Rods start at 21 C and the Caps are set to the ‘right’ temperature, the hot Rod end gets ‘too hot’ (compared to what it ‘should’ be), and there is then enough conduction to heat the cool ends above ambient.


    Furthermore, heating only just the Rib area to a ‘correct’ temperature, the Caps get hotter than they ‘should’ be, especially at higher temperatures, which then cascades to the Rods, making them ‘too hot’. Note the Caps climbing over 1000 C when the Rib area is set to 1410 C, with no additional input heat added to the Caps or Rods besides the heated Rib area only. (Second image above).


    There are probably a dozen ways the simulator set-up I used could be flawed, besides that it is a 2D simplification, but I think the general ideas it expresses are analogous to what actually should happen. I need to spend some more time on that, but practical experiments are probably a better guide to reality and may guide the simulation adjustments closer to the real thing.

  • don't know if they are still in active business but if so, Earthtech (Scott and Marissa Little and company) could do an objective job and I suspect, given strong evidence of positive results (Eout>Ein) they would be very interested.

    http://earthtech.org/maverickinventor/


    I know Hal, Scott and Marissa, fine people. But we have another outfit on the case, possibly even more credible.

  • Where my very basic modeling seems to have trouble is with just the hot end segments of the Rods set to the ‘right’ temperature, the far ends don’t get hot enough (stay at ambient), but when the Rods start at 21 C and the Caps are set to the ‘right’ temperature, the hot Rod end gets ‘too hot’ (compared to what it ‘should’ be), and there is then enough conduction to heat the cool ends above ambient


    I'd say that if your model is within/matching 15% of reality, it's performing well... IMO there's nothing wrong with tweaking (calibrating) the alumina conductivity and radiative/surface-air transfer coefficients based on data from your earlier (physical) hot tube experiment, to get a better overall fit. These values are influenced by porosity and surface roughness - so who really knows if the book values are truly accurate. Checking against a range of temperatures should confirm if the model is stable.

  • What about using as a first approach a 30 mm stainles steel tube coated with a Durapot layer to get the emissivity in line ?

    Much cheaper.


    You would want somehow to check the transmittance of the durapot layer at different wavelengths since for stuff going through it steel and alumina will look very different. Retrofitting the Lugano reactor is difficult due to uncertainty about Durapot layer thickness, unless it turms out to be opaque at all relevant wavelenghs.

  • You would want somehow to check the transmittance of the durapot layer at different wavelengths since for stuff going through it steel and alumina will look very different. Retrofitting the Lugano reactor is difficult due to uncertainty about Durapot layer thickness, unless it turms out to be opaque at all relevant wavelenghs.

    Alumina is optically thick in the IR range at > 0.254 mm thickness (Wade, 1959) , Durapot should be very similar.