>>1920919You're calculating volume incorrectly. The other anon is right, vehicle volume does not use length, it's expressed as the number of vehicles per hour per lane that can cross a point within a given time (hour). 1,800 vehicles per hour as he mentioned is a generally and often used number in traffic engineering which is assuming vehicles moving at free flow speed, same as in your scenario assuming some kind of average headway (what you call out as 3 second rule). 1,800 however is optimized and assumes no breakdowns, congestion, or queues.
In your calculation, assuming a Tesla can pass a single point every 3 seconds, the math is: 60 minutes * 60 = 3600 seconds, divide this by 3 seconds, and you get 1,200. In other words, 1,200 Teslas per hour, each one carrying 4 people, the volume of the tunnel in persons per lane is therefore 4,800 people/hour. You would need 19 single lane tunnels to reach 90,000 people/hour and this ignores loading. I would love to see a station design that can reliably load 90,000 (technically 45k assuming bi-direction) people per hour into low occupancy vehicles. That's a whopping 750 people per minute, and at 4 people/car, that means you need 188 vehicles actively loading, per minute! The math continues, normal taxi-stall loading dimensions are around 25 ft by 10 ft, meaning you need 47,000 square feet of loading space and not accounting for sidewalks and vertical circulation.
What you calculated would be more akin to the system capacity, not the throughput. As in, either across one hour or at any given point, how many people are actually inside of the system? And as you calc'd, this is a function of how many cars can fit inside the tunnel and how many people can safely fit inside the stations? But again, this is system capacity, NOT volume or throughput. Hope that mostly makes sense.
