I don’t buy this stealing, and here is why:
If they aren’t actually using the energy, and just storing it in a buffer, it would take less energy to re-charge the car again than it would otherwise. I feel like you can’t have it both ways. I can’t think of a way you can continually store energy (rather than lose it) in a buffer for every mile driven, yet somehow it constantly takes more energy to restore that energy than came out.
The only explanation that I can come up with is that it is actually lost.
I’m not suggesting that everything is exact - everything is an estimate of course. But I can’t think of an explanation for a difference in the iWh and oWh constants (I like that terminology) other than that difference is lost (or miscounted as I have previously suggested - but that is a minor semantic difference - miscounted or lost actually are effectively the same thing).
If you can explain with an example how it could work perhaps I could be convinced otherwise. I feel like there should be a way to prove/disprove it one way or another in any case.
I feel like there should be a way to prove/disprove it too, but I can't think of a way
What follows is my best shot at explaining how it MIGHT potential work (or not)
Also to be clear, I’m not suggesting “buffer theft” to explain the difference between kiWh and koWh. I think that’s probably a real ~4.5% loss they are factoring in to estimate loss going in and then out of the battery. I think other weirdness or deviations from koWh we see could be explained by the “stealing” for the buffer.
The buffer fudging can be hidden behind charging or driving inefficiency.
Say the car starts at 100%, you consume 10 koWh of energy driving, but they subtract 10.26 koWh of energy from the battery gauge & range, and increment a variable called "energy buffer" from 0 to 0.26 kWh (targeting say 2 kWh buffer by dashboard 0%).
You've only used 10 koWh of ‘real’ energy, so to replace that you only need to charge 10.47 kiWh.
As you charge, they ‘steal’ from the input to replace what they stole from the battery gauge.
It looks like you are accepting charge to the battery more efficiently than you actually are. 10.26/10.47 = 98% efficiency instead of the new ‘o/i model's ~95.5% (10/10.47) efficiency.
They just keep track of the theft. As long as they put the money back in the register by the end of the day nobody will notice
Analogy:
You are a cashier. For every $100 customers pay you and you put in the till, steal 4 quarters from the till to put in your pocket. If someone looks closely at the till through out the day they will be wondering why it is low -- but when you cash out at the end of the day, for every $24.75 you count from the till, you slip a quarter from your pocket back into the till. By the end of the cashout, the receipts and the money balance and all is well.
I think the reconcilliation with the buffer while driving doesn't start happening unless you approach near-zero on the dashboard and/or go below zero. Otherwise it's reconciled to zero as you charge towards 100%.
You could also have a big recalibration/re-estimate event where the SoC wants to drop 2% or 6 miles. Instead of just instantly dropping it 6 miles, they slowly drop it an
extra 6 miles over the next 18 miles so you don't see the big jump. They can do this by instantly moving 6 miles from the energy buffer they've built up, back into the battery gauge, and then stealing it back out slowly so you don't notice. (I mean, most people won’t notice —
we are noticing
)
If someone glances at your cash register totals throughout the day, they might notice a $10 drop, but if you are stealing quarters at a steady rate, they won't see a drop. If you make a mistake and give someone too much change, you can hide it (temporarily) by taking money out of your pocket and putting it back in the till. You just can't cover too big of a mistake
Say the “register” is a big glass jar of quarters ... your manager might notice a $10 drop at once, but not a quarter at a time.
...
I dunno ... I feel like even flat-earthers come up with some crazy 'convincing' arguments that fool people too so this could all be a cockamamie model explaining observations in a naive and ridiculous way. Or it could be on the right track
Or the meter just "reads low"
