Temperature units for Celsius

[mdeceglie]

The temperature units are currently expressed as C, which is technically the SI symbol for coulomb. For parameters like temperature_cell Celsius is natural, so I suggest using °C. For units where temperature appears in the denominator, kelvin may be more appropriate. For example, in Table 4 on this NIST site derived units use kelvin.

[cwhanse]

Are you suggesting that the documentation text include the degree symbol, or that the unit be changed to K? I'm not in favor of the K option.

The degree symbol is awkward to type (ALT+0176 on a Windows keyboard) but could be added to the documentation file.

[mdeceglie]

Yes, for temperatures such as temperature_cell I am suggesting that the documentation text and definitions.csv include the degree symbol. The units as they are now, without it, specify coulombs.

For units where the temperature appears in the denominator such as gamma_pmp I suggest changing to K, but could go either way on this point as long as the degree symbol is included along with C.

[toddkarin]

Sounds good to me. For gamma_pmp, do we want W/°C or W/K? I'd vote for W/°C for the engineering crowd.

[cwhanse]

We can try adding the degree symbol to a line in definitions.csv and see how it renders. @mdeceglie @toddkarin care to make a PR?

[mdeceglie]

I'll give it a shot and open a PR. Thanks!

[steve-ransome]

If "°" <alt + 0176> causes problems I've seen degC used before.