QuoteThe adjusted Temperature for Lugano is calculated as following:With Ecat T=1250C and ε1 = .41 adjusted to ε2 = .90 then E-Cat T=896C ! (Formula according Optris manual (ε1/ε2)1/3* T measured in Kelvin.)this gives roughly 2.85 times less energy than measured, which finally is corrected by the higher bandemissivity at the lower T.ε = 0.6 as you propose is to high for 900C. This would give a power reduction of less than 2. ( you should also care that the convection is reduced somewhat less..)
This derivation is I'm afraid now very confused. e=0.6 refers to the total emissivity. I've no idea what this Optris formula is except it most certainly does not address the matter of n - this number cannot be linear with T (as you can see from Optris's own graph that you posted). There will be no clarity for you until you stop trying to interpret manufacturer's data and getting this wrong, and work out yourself from first principles what is what here. Paradigmnoia, I, and (it seems from his paper) TC have all done this.
QuoteClarke used theory to correct the temperature based on a presumably better estimate of emissivity, but it was still that, an estimate. He used an exponent of 1, based on his own mathematics, referring to the variation of camera measurement with temperature (where, above, an exponent of 4 was proposed, clearly incorrect, then a value of 3 was taken from an Optris explanatory graph, clearly not intended to give a precise value, Paradhgmnoia came up with 1.6, and we have no agreement, and there are variables that are unknown, acknowledged by all, included the exact material.
It needs some care to get these things right. We are now nearly there, but...
TC did not use an exponent of 1. He said the exponent value was much lower than 4, and roughly 1 (in fact this is a minor mistake - but it does not anywhere affect his work because he does not use it). He did the numerical calculations ab initio - and if you look at his figures you get an exponent of roughly 1.6, same as Paradignnoia. Since the exponent can only be worked out from this numerical calculation it is irrelevant.
I need to reiterate: n (the exponent) as a derived quantity calculated from the Planck curve as applied in this case. TC integrates the Planck curve directly weighted by typical bolometer characteristics and iterates to get the correct temperatures form the given data. This does not use the value of n.
There is no disagreement about this theory. Except (possibly) from Wyttenbach who is still trying to make sense of what Optris says.
As for accuracy: true - this reanalysis relies on the band emissivity and total emissivity of alumina. There is some variation in both (and some other issues, see TC's paper). But don't throw out the baby with the bath water. This method is unlikely to be inaccurate by more than 30%. That is better accuracy when we have ever before had testing Rossi's devices! Of course IH will have their own much more accurate tests.