Hihi, one or two things to consider but this doesn’t majorly contradict anything said above I just think it’s useful background info
On the flex vs non-flex point:
People referred to brick packs, and if you’re not familiar with them they are often vertical arrangements (cells stand on the narrow circular end rather than lie on the long cylindrical side), and almost always packed in a pelican case style box. This has a couple of notable effect, firstly the longest dimension of the pack is significantly reduced and geometric rigidity (ie natural stiffness from the shape of the pack, disregarding construction methods) is increased quite a lot. Secondly it’s encased in a fairly stiff material but more importantly in a body that’s not subject to much load at all. As compared to a long(er) under board pack, where having one dimension much longer gives less rigidity and (IMO more importantly) having the enclosing material subject to very high loads.
The way that humans notice or experience flex in a board is mostly the slow or sustained bending, ie you jump on a board and it bows for half a second and springs back. What we don’t notice or think of the same are the really fast vibrations because if something moves half a millisecond instead of half a second you’re not gonna feel it, but there’s a lot of energy in a fast movement like that. So in short: vibrating is very similar to flexing, lots of stuff vibrates, under board batteries are a flexible shape enclosed in material that both vibrates and flexes (vibrates on uneven surfaces, flexes when you step on or drop a 3cm curb or whatever), and nickel welds should not be a big structural component of anything if you want it to last. IMO then you’re better off making P groups and wiring them together because wire can flex, but it’s not the end of the world. Plenty of packs have been made that lasted a goodly long time with nickel series connection, but it gives me the heeby jeebies. I’m open to being corrected on how relevant all of this is by a mechanical engineer, I’m outside of my electrical comfort zone here, also here’s some fun videos that extract data on otherwise-invisible vibrations from video. It takes short contrast patterns and blur artefacts and uses them to pull out information on very quick deflections that would just look like vibrations
On the battery current:
Bunch of good answers above about batteries only drawing what you ask, also consider this to get a reference point for how much current you need. I was answering someone asking about using high battery current and linked them to the ride logs of a biiiig fuck off board, where you can see both the maximum current ever used and importantly, the normal currents. It barely ever passed 60A, ie less than 3kW, and never in the life of the board passed 3.5kW as far as I can tell.
And this
To be extra clear: this should be a bit comforting that you can have a truly exceptionally high performance build without ever worrying about that 85A battery current range
Cell selection:
Discussion has already been mentioned but I’m gonna leave this link here
As well as mention that it’s not a huuge difference tbh. Definitely seems like a better pick, and P28A cells from a batch in the last 12-18 months perform a little better again, but it’s not going to be the difference between a usable board and an unusable one, 30Q has been a decent choice for a long time because it’s pretty decent.