2" diameter PVC pipe gas a 2.375" OD and is cheap. Use something to take up the extra 1/16" each side.

12 nm is the size of the internal wiring. Smaller feature size means lower power, higher speeds and.greater density (smaller physical size).

The DRAM packages can be mounted to Printed circuit boards (PCBs) for laptop, desktop or server form.factors.

Very true. I think the big time arbitrage options are for vehicles that don't operate much. School buses Tractors / agricultural vehicles Delivery trucks Rental cars / fleet vehicles.

Just think about using school buses to power the grid. They generally don't operate at peak demand (6p in the summer). They are already distributed geographically and would have large batteries.

Most combines only operate 1x per year.

Suddenly these vehicles have an entire new use case.

Right concept, but there's likely better battery chemistry choices because the energy density is less critical on fixed units. It's all about cost and number of charging cycles.

What's really.interesting - if you think about the California duck.curve - peak pricing is likely overnight and midday prices on a sunny day should be very low as there could be an excess of renewable power.

Having lots of battery storage produces interesting electrical arbitrage opportunities.

I'm interested in how fast charging infrastructure gets implemented. Charging at 500 kw/vehicle means a service station with 16 vehicles could put 8MW swings on the grid. That's somewhat substantive.

Multiply times a few charge stations and that could impose maximum 100MW swings.on a grid. Now, that's worse case scenario, it's pretty unlikely you go from 200 vehicles charging to 0. The most likely number is probably 20MW swings.

I see each station having a few MW-hrs of batteries as distributed storage. That way the batteries are trickle charged continuously and the peak loading is essentially behind the meter / off grid.

Mainly because I'm lazy and hate doing rework. That being said - I only want to rework once. When the first tile fails I'd do them all.

Personally, I'd wait for the first time to fall off then re do it all. May take 2 years, may take 5 years - but I'm sure the adhesive will fail.

Agree with your comments. There is a very vocal pro-nuclear majority that seems to think this is the best technology to deploy for power. The tough part of nuclear is that it's super expensive. And even though we've been building PWR reactors for 50 years, the cost to build new reactors isn't coming down. Over building renewables / batteries is more cost effective than building fission reactors.

Maybe fusion or SMRs can change that, but nuclear is just too expensive to build as a primary power source.

You wrote the units wrong also. You want kw*hr, not kw/hr. Maybe not so obvious?

I'm saying kw is already a rate. It's units are J,/s. Kw/hr would have units of J/s-hr. That makes no sense. Another way to say you supply 3.6 MJ in an hour would be to say 1Kw-hr/hr. And hrs cancel to 1Kw.

Most people want kw*hr, not kw/hr. Kw/hr would be the rate of change of charging speed. Ie, I started charging at 1kw, an hour later I'm charging at 11kw, charging rate increased at 10kw/hr.

I think phase here means built in one investment decision. So they committed to 350MW in one go as opposed to 100MW, then 200MW then 200 MW.

Would definitely be 3 phase power, since basically all AC power distribution is 3 phase.

Kw/h is a meaningless unit. Kw has units of kilojoules/sec, it's already a rate. The unit kw-hr - or kj-hr/sec is a capacity unit cause the time "cancels". Kinda. Kw-hr units are strange in a lot of ways.

I don't think kw / kw-hr are intuitive units for most people.