Automotive manufacturing plants are notoriously high energy consumers… more so than one would expect. Its not just the heavy manufacturing body shop or the final assembly but often the paint shop uses the MOST energy overall. And that is just the energy that goes into direct costs for manufacturing. In a giant large usually open space bigger often than giant aircraft hangers, just HEATING the place for the employees is a massive energy suck. Oddly, paint shop is often the overall highest consumer since air handing is such a priority in order to keep air particles out for quality spray and having no impurities in the paint jobs. It’s also very cavernous (like a manufacturing facility overall) and that too has to be quite precisely temperature controlled - again like the big open manufacturing facility space. All large consumers of energy. Much of this is high voltage electricity but a plant will certainly use nat gas to create the heat required for plant operations during the colder months. Freemont certainly has more stable temperature for the winter months, compared to say southern or east Germany, but laws of thermodynamics say - bigger open volume of air require more energy to change temp, up or down. up is actually harder and requires more energy than down.
This is not to say that Tesla energy couldn’t put in some large megapack array (might not be able to replace the air heating scenarios unless one had simply true heat exchangers) but could certainly keep ops running, certainly final assembly for some period of time. But, is it worth it, for the maybe half day or shift of downtime that might occur once every 18 months? I’m sure (or at least I HOPE) that someone in Tesla vehicle manufacturing and Tesla energy has done this thought exercise with some real numbers from their factory. Something tells me they probably have.