Man, that’s embarrassing.
In order of importance:
- All body parts and car parts remain attached.
- It’s a piece of cake to recover from this error.
- I was, in hindsight, being a complete dumbass.
The dumb-assity of it was clear the instant the car hit the ground, but it took a while to figure out the “complete” part. This post explains why the car inevitably fell off the jack. Because it wasn’t obvious to me, even after it happened.
1: Easy recovery.
Just jack it up by the other end.
On a modern unibody car (with no separate frame), with a standard scissor jack, you have to place the jack under the car, directly below a ridge of sheet metal. Then make sure the jack end straddles that ridge, and proceed to jack up the car.
That means that if your car falls off the jack, with the wheel off, the brake rotor ends up on the ground, and there isn’t enough room to put the jack back under the car.
I literally couldn’t figure out what to do. Pry it up with a 2×4? Buy a floor jack and put it under the front suspension?
Then I looked up this YouTube video:
The answer is, jack the car up by the other end. Turns out, that will raise the entire side of the car. In my case, I was changing a front tire. After chocking the wheels, I jacked the car up at the rear jack point. This raised the front frame enough for me to put a brick under it. I then let the car down, moved the jack to the front, and proceeded as if nothing had happened.
No harm done.
But why did it fall?
The parking brake wasn’t set. I should have checked that, but … stuff happens. Edit 1/27/2024: In hindsight, the parking brake wasn’t working, something I only found out after yet another flat on that car. I had the good sense to chock the wheels the second time around.
Even so, the car was in park, on level smooth pavement.
Why didn’t putting the car in park keep it from rolling?
I had the car in park, and one front tire was firmly on the ground. Why didn’t this hold the car in place?
Turns out, on a front-wheel-drive car, park locks the output shaft of the transmission. But it doesn’t lock the differential — the gizmo that splits the power between the two front wheels, and allows those wheels to spin independently as you go around a curve.
Because of that, if one front wheel can turn, then both front wheels can turn. They turn in opposite directions, through the differential.
The upshot is, on a front-wheel-drive car, once you have one front wheel off the ground, there is nothing about “park” that holds the car in place.
Apparently, in northern climates, it’s well-known that if you park with one wheel on ice, in a front-wheel-drive car, the car may slide even though the other wheel is on dry pavement. That happens for the same reason. As long as one tire can rotate, both tires can rotate, even if the car is in park.
My bad. Lesson learned.
But how did I generate enough force to move the car?
I mean, I placed the jack very nearly vertical. To within a few degrees, anyway. How could that possibly have generated enough sideways force to get the car to roll off the jack.
This car was in my garage, on a concrete slab that is very nearly level. It weighs about 3000 pounds. My gut feeling is that it should take quite a bit of force to get that moving. Right?
Wrong. Based on a quick check of several sources, the coefficient of rolling resistance of car tires is about 0.01 to 0.015. That means it should take somewhere between 1% to 1.5% of the weight of the car, applied horizontally, to get the car rolling on a smooth surface.
Or in this case, somewhere between 30 and 45 pounds of force, horizontally, could be enough to get the car rolling on level pavement.
How far off did the angle of the jack have to be, to generate that amount of force? Assuming I recall my elementary physics right, that goes as the sine of the angle away from vertical. In this case:
The upshot is that I’d have had to eyeball the angle of the jack to less than two degrees of vertical. Off by any more than that, and the geometry and physics of the situation would generate more than enough force to roll the car.
(On top of this, the garage floor has a slight slant, such that the car already has an estimated 15 pounds of horizontal force on it, just from sitting on that slant. And, by picking up (say) 1000 pounds of weight off the car, and onto the jack, I actually only had 2000 pounds on the wheels, so I only needed 20 to 30 pounds of horizontal force to get the car to move.)
Conclusion
Before this, I had the vague notion that the car might roll off the jack if I didn’t take preventive action.
Now, having done the math, I can say that my front-wheel-drive car, in park, on level pavement, will roll off the scissor jack if I don’t take preventative action when changing a front tire.
In hindsight, it wasn’t a slightly risky, maybe-I’ll-get-away-with-it maneuver. Falling off the jack was a near-certainty, under these circumstances.
FWIW, I’ve added a pair of folding metal wheel chocks to the bag holding the jack and tire iron. Presumably, if I literally have those in my hand, the next time I get a nail in a tire, I will have the good sense to use them.
I think.
Illustrations are from the Gencraft and Freepik AI sites.
