SpaceX And/Or Rocketry In General

A hump generally has two sets of retarders. The amount of braking applied by the master retarder is calculated based on the car weight, but the actual effect (resulting speed) is then measured and fed into the calculation for the group retarder. Besides the weight of the car, other contributing factors include the temperature and the wind speed and direction. Really strong winds have been known to actually push cars back up the hump.

Another part of the calculation is the computer's notion of how many cars are already in the destination track and how far down they are. This is NOT measured empirically. Many years ago, I observed the Burlington Northern hump at Northtown Yard when computation was not matching reality. I kept hearing WHAM! and finally noticed a box car rolling into a track and hitting the preceding car really hard. At the distance from which I was watching, it took a few seconds for the next WHAM! to reach me. I imagine there were a lot of claims for damaged lading in the next few days.

At the museum, of course, we don't have a hump, but we do try our best to couple gently. A standing car or train being coupled to would ideally not move at all, but preferably no more than a few inches. But it does have to be hit hard enough for the knuckle to close and the pin to drop, and then you stretch it to make sure.