clarify units for U1 and Uv temperature model coeff

[mikofski]

• closes BUG:units for Uv & U1, convective thermal coefficient for Faiman and PVsyst, m/s should be in denominator #945
• group denominator together, same way it was done for U0 and Uc, ie: W/(m^2 C) instead of (W/m^2 C) which is ambiguous because it could mean W/m^2 * C = W * C / m^2 which is not the same thing
• add missing division symbol before 2nd denominator (m/s) to get W / (m^2 C) / (m/s) instead of (W/m^2 C) (m/s) which looks like WmC/ (m^2 s) so it's ambiguous
• pvsyst help docs (https://www.pvsyst.com/help/thermal_loss.htm) say W/m²·k / m/s which is also ambiguous but you can't win em all

• Closes #xxxx
• I am familiar with the contributing guidelines
• Tests added
• Updates entries to docs/sphinx/source/api.rst for API changes.
• Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
• New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• Pull request is nearly complete and ready for detailed review.
• Maintainer: Appropriate GitHub Labels and Milestone are assigned to the Pull Request and linked Issue.

[kandersolar]

In this and other cases with combined units, I wonder if it might be better to use a :math: tag to improve the rendering. The plaintext display is a little difficult to parse:

[mikofski]

In this and other cases with combined units, I wonder if it might be better to use a :math: tag to improve the rendering.

Everywhere? what about units that are simpler to parse? Probably best to be consistent, right?

pvlib-python/pvlib/temperature.py

Line 46 in 3ce8e1a

Ambient dry bulb temperature [C].

I'm okay with making the shift to use :math: to render all units, but maybe that should be a separate issue? It would be a bit of an undertaking, good first timer maybe?

[cwhanse]

Everywhere? what about units that are simpler to parse? Probably best to be consistent, right?

Perhaps in the few instances where the unit involves a compound fraction. I'm -1 for using :math: for units otherwise.

[mikofski]

Quick poll, so we've decided to use :math: to render these ugly confusing units here only, EG:

in :eq:`pvsyst` :math:`\left[\frac{W/{\left(m^2 C\right)}}{m/s}\right]`.

Looks like this

[adriesse]

Could we do:

 W/m2
------
C(m/s)

This would be more easily associated with the variables going in.

[mikofski]

Could we do: W/m^2 / (C(m/s))

I'm fine with

:math:`\left[\frac{W/{m^2 }}{C \left( m/s \right )}\right]`

which looks like this:

Powered by CodeCogs

But this makes me wonder if the U0 and U_c should also be changed to look the same?

Powered by CodeCogs

What do y'all think?

[mikofski]

Also should we be using Kelvin instead of Celsius?

[adriesse]

Sure, but I'd also be ok with the plain text:[W/m^2/C] and [W/m^2/C/(m/s)].
My preference is for Celsius.

[cwhanse]

My preference is for Celsius.

+1

[mikofski]

sorry stickler but inline math don't render if it's on two lines 😕

[mikofski]

What do you think @CameronTStark? Is this good?

[CameronTStark]

LGTM @mikofski, thanks!