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

I just did a Spice simulation and it looked sketchy still.

Scrounging around for a 2:1 transformer...

Quote from LDM

What about using Kantal wire to make a series resistor ?

Also a bit sketchy, having another hot (electrically and thermally) wire in the experiment area.

There is a center tap on my isolation transformer output side, so good to go there. Luckily the current should be fairly low, so hopefully it won't protest too much.

It looks to be about 56 to 57 V true RMS maximum output, which should get me to 'COP" 4.5, and keep the cylinder in the safe-ish range of around 820 C external. In theory.

(Assuming it doesn't explode as the coil wire expands)... safety goggles and leather apron on, fire extinguisher at ready...

I might fire it up tonight. I'll skip the rectifier since it isn't helping anyways.

Quote from Alan Smith

Very sensible- it serves little purpose.

I thought briefly that the bridge rectifier would make 60V DC (with a 1mF cap) from 120V AC, but it won't really work that way without the isolation transformer feeding the AC into the rectifier.

Quote from THHuxleynew

Just an additional complication which you hint at here but do not explicitly say. What also matters is the translucency of the paint. At frequencies where (for the used thickness) it has low opacity the emission characteristics revert back to those of the underlying material. And opacity can vary greatly with frequency just as emmissivity can.

I recall discussing this idea in 2015. At that time, before any idea of paint had settled upon the Lugano conversational milieux, the discussion rotated around the effect of the alternately opaque or luminous coil wires through the semitransparent alumina wavelengths.

The optical thickness of alumina becomes important to the discussion, and as it turns out, the amount of alumina to reach optical thickness, generally, is very little. Otherwise special emissivity paints could scarcely be a thing. I think the papers of Rozenbaum as well as Manara et al. get into the optical thickness measurements of alumina, SiO2, and MgO. The thickness required to affect the transmissive, rather than opaque parts of the spectral emissivity, I do imagine, would need to be greater for the former than the latter, as typically the optical thickness of ceramics is correlated to porosity more than other properties. (Increases in density are effectively decreases in porosity, and increasing porosity increases light scattering). The scattering effect of porosity is such that the opaque region of emissivity increases in emissivity with increased porosity. The surface coating therefore only needs sufficient thickness and porosity to scatter light effectly in the semitransparent range enough to reset the emissivity profile to that of the coating. This critical thickness can be as thin as 8 to 10/1000 of an inch (0.2 to 0.25 mm) (Wade, 1959, figure 17, showing thickness of zirconia to block inconel emissivity at 1200 F.)

Manara, J., et al, 2008: Bavarian Center for Applied Energy Research (ZAE Bayern), Wuerzburg, Germany, "Determining the Transmittance and Emittance of Transparent and Semitransparent Materials at Elevated Temperatures", 5th European Thermal-Sciences Conference, The Netherlands, 2008.

Rozenbaum, O., et al., 1999: Centre de Recherche sur les Materiaux a Haute Temperature-CNRS, 45071 Orleans Cedex 2, France, "A spectroscopic method to measure the spectral emissivity of semi-transparent materials up to high temperature", Review of Scientific Instruments, Volume 70, Number 10, October 1999.

Rozenbaum, O. et al, 2009: "Texture and Porosity Effects on the Thermal Radiative Behavior of Alumina Ceramics". International Journal of Thermophysics, Springer Verlag, 2009, 30 (2), pp.580-590

Wade, R., 1959: Langley Research Center, Langley Field, VA, "Measurements of Total Hemispherical Emissivity of Several Stabily Oxidized Metals and Some Refractory Oxide Coatings", National Aeronautics and Space Administration Memorandum 1-20-59L

[Note that Wade, 1959, is the source for the "alumina on Inconel" Total emissivity that appears in many IR camera and pyrometer emissivity tables, often (and incorrectly) suggested for the in-band IR detector emissivity setting]

Well, test one of the cylinder was completed late last night.

A few issues to sort out, and a few surprises.

The apparent IR pyrometer emissivity started out similar to the the Slab... 0.91 from 20 C to 100 C, then at 144.8 C it jumped up to 0.96 V, remained there to 372.1 C where it went to 0.97, then dropped smoothly from there to 0.93 at 658.7 C , when I shut it off. So the apparent emissivity is higher than previously found. I'll rotate the cylinder 180 degrees to check for a hot spot where I was aiming the pyrometer for this test.

The transformer is unhappy, and was getting real hot. Way too hot at a cylinder temperature of 658.7 C, which is why I shut it down. The transformer also consumes too much power relative to the cylinder, (and so it gets hot). This makes the cylinder power consumption difficult to reconcile with the output power. Probably the triac-like waveform doesn't agree with the transformer, and using the center tap and one lead for output possibly is another problem. So a better power supply solution needs to be worked out.

The cylinder is cooler on one end than the other for a greater distance. This is obvious when glow heat begins, and with the pyrometer also. Maybe when the Durapot was being poured in it floated-stretched the bottom end of the coil in the mold a bit. A stiffer wire should be used when another one gets cast, but that messes up the length of the coil... I'll have to work through some permutations. It is also feasible to pass some very thin stiff wires through the silicone mold (to be removed immediately after pouring) that will hold the coil wraps from up-down movement when pouring in the Durapot. I did something similar to that to keep the Slab coils from moving around, since there were four continuous parallel coils that wanted to twist themselves to odd angles at the slightest provocation.

On the other hand, it didn't explode into a shower of red hot fragments, so that was good.

Overall, not too bad for a first version of a new design.

I might try disks of high temperature insulation on the cylinder ends, with a thicker one on the cooler end, to see if the cylinder temperature can be evened out a bit.

I also measured the resistance of the coil after the power was shut off, so I have the resistance from 610 C down to ambient. That should make It possible, with the coil voltage and total input power I recorded, to work out what the transformer losses were.

Quote from Robert Horst

Why not use a Variac to drive your coil? You can get a 1000 VA Variac for ~ $70 on Amazon. That will let you set any voltage (power) input and does not create EMI like a Triac-based dimmer circuit.

I probably will do exactly that. I wasn't originally anticipating that the coil was going to be so voltage sensitive until I ran the numbers a little later than perhaps I should have, so I was improvising with what was on hand.

I can remove the transformer altogether and just be careful with the SSVR dial for another test. The steps will be more coarse, but that is probably better than the unimproved set-up I used last night.

Although the data is not sufficient at present to really confirm it, it looks as though the Durapot total emissivity is quite a bit higher than pure alumina. I'll clean up the experiment and see if this still looks like the case once better data comes in.

Alan Smith ,

This was a bit more of a challenge compared to the way I usually make things, which although infuriating at times, was worth the learning curve.

Normally I design a coil to max out right about the maximum voltage I can easily supply using my favourite wire size. Other than adjusting the coil diameter, the coil size ends up being whatever size it needs to be.

This time I had a final size range in mind, and worked out whatever needed to happen to get it there, within a limited final length range to eat up the problems that wouldn't easily go away some other way.

From past experience, I was able to anticipate and attempt to offset a bunch of problems that I expected. But until it all comes together, the true range of difficulties is hard to assess on a fresh build.

For example, I preferred doing a two-part mold, horizontally, but getting the thermocouples where I wanted and keeping them there was problematic, as well as keeping the coil centered without any support, so I did the vertical one-piece mold instead, which had other difficulties.

However, despite some power supply uncertainties, total emissivity uncertainty, and lack of precise input measurements, the cylinder is getting in the general range of being as hot as the input power and convection-radiation formulas predicted. That in itself is a useful outcome. With some improvements suggested by the last test, it should be possible to predict much more closely the performance of an improved version before it is constructed.

Cylinder cooking away again.

It turned almost grey after cooling with fine, sooty coating, probably pyrolusite.

Pyrometer emissivity setting decreased from last test. Possibly crystalization change after cooking to 650 C previously, resulting in MnO2 rejection. If the black coating does not burn off tonight, I will see about removing the coating after it cools to see if it interferes with the emissivity. Might be able to turn it into Mn3O4 tonight? (If it is Mn anyways)

He he he...

1712 C Lugano Style

Off to bed.

E 0.94 , 894.5 C ----> E 0.46 --> 1510 C ----> E 0.39 --> 1712 C ----> E 0.39 --> 1712 C

The coolest part of the device at the time of the above was 727.2 C ---- ... ----> ~1300 C , leading to a 'COP' of 4.0, ignoring that 3/5 of the device was greater than 850 C (to be conservative).