Anatomy of a Rollover - Part 2
Part 1 explained what happened; Part 2 will try to explain why it happened.
It took me a while to figure this out and still, I can't guarantee that this is 100% accurate. Nevertheless I'm pretty sure this is why the car ultimately lost control and rolled over. It bugs me because I'm a fairly conservative poor-weather driver and the accident occurred on a straightaway - it's something I would've never imagined could happen in an AWD vehicle but I think the Honda Pilot's AWD was part of the problem...
Initial conditions:
- two-lane state highway
- straight stretch of road
- 35 degrees and snowing
- 2" unplowed snow created a "two-track" effect
- driving 40-45 mph
- going east with a 20-30 mph NE wind (left quartering headwind)
- nobody in front or behind in our lane
- one car in the oncoming lane
1. Oncoming car passes, causing an air draft that combined with the already strong crosswind, nudges our car toward the shoulder.
2. I steer slightly to the left and take my foot off the pedal, trying to re-center the car in our lane.
3. The front wheels begin to "plow" as our car enters the snowy parts of the "two track".
4. Our car starts to weather-vane into the quartering headwind; i.e. the back end is moving closer to the shoulder than the front end.
5. Sensitive to a possible oversteer condition, I return the steering wheel to neutral and lightly step on the gas. I intended to "drive" back into the snow-less two-track and then straighten out the car there, where there was more traction.
6. The back end begins to fishtail toward the shoulder, catching me by surprise. It was only later that I figured this out: The Honda Pilot AWD system operates in FWD (front wheel drive) but will direct power to the rear wheels when the front wheels start slipping. When I straightened the wheel and applied some power, the fronts were already plowing so power was directed to the rears. However, due to the weather-vane effect there was swing momentum on the back end and with the rear tires already in snow they must've slipped, causing a fishtail.
7. Still I wasn't too concerned; the front wheels hit the dry two-track and regained traction, pulling us back into the center of the lane.
8. I steered a bit to the right, intending to straighten out the car and I thought that this minor event was all but over.
9. Here's where I made my mistake: I kept my foot on the accelerator, assuming that the front wheels on dry pavement would pull the car into alignment. I didn't realize that the rear wheels had been spinning - I'm not sure when the Pilot's AWD reverts back to FWD, but since it doesn't have traction control or stability control, it's not as "intelligent" as a more modern system. The rear wheels were probably still spinning as I steered right and the front wheels caught dry pavement, allowing the rear end to fishtail strongly to the left.
10. The back end is now across the center line to the left; I assumed that I hadn't centered the steering wheel - turning it too far to the right in step 8, so I turned it to the left thinking that I was centering the wheels. In hindsight I had centered it in step 8, so when I turned it left here I inadvertently set us up for an oversteer.
11. With the car somewhere around 45 degrees sideways, turning the front wheels pointed them along the direction of the road. Since they were still on pavement and still under power, they tried to pull the back end back toward center.
12. Mistake #2: I took my foot off the gas, but too late. The back end had already started swinging back to the right by the time I removed power from the wheels. Coupled with the strong crosswind, the back end swung severely to the right.
13. The back end swings almost all the way around, not quite 180 degrees - we were coasting backwards in our lane but with the back end a bit closer to the shoulder than the front.
14. The wind continues to push, causing the car to drift onto, and off of, the shoulder.
15. This wouldn't be so bad in most places, but we happened to go off the road at a 12-foot embankment with a 45-degree slope. I was hoping we'd just coast along the slope at an oblique angle and come to a nice stop when it flattened out...
16. Alas, the back left corner of the car dug into the tall grass and soft dirt, stopping our rearward movement rather abruptly.
17. We still had some kinetic energy that had to dissipate somewhere, so the car began to roll. It wasn't perfectly sideways - when the rear left corner dug in, the car pivoted a bit around that corner and the resulting roll was somewhat back-left-bumper to front-right-windshield as we rolled down the slope, hence the damage on the passenger side of the front roof.
None of this would've happened if I had better understood how the AWD would react, and it's a good case for why an AWD without traction control can be confusing. Amanda's Subaru Forester has full-time AWD where all four wheels are getting 25% of the engine's power under normal conditions so when one wheel begins to slip, it's more predictable and controllable.
Yet most AWD systems are FWD in normal conditions, meaning that when wheels slip it's hard to conceptualize what will happen next. These systems are mostly meant for accelerating from a stop in slippery conditions, not maintaining control, which is why most of them come with traction control these days. My 2003 Honda Pilot didn't have it, but I believe all Pilots since 2006 have the more advanced AWD systems.
Comments
I'm glad you are both OK - I hope all the injuries are healing quickly. Sometimes the mental part of it takes a bit longer to mend. ;)
It's really amazing how you remember all the details. Your breakdown of the sequence of events and analysis = fascinating.
It does seem odd that the officer could cite you for something that neither he nor anyone else witnessed. So because you crashed, he assumed you were going too fast? *shrug* Maybe you were - hindsight would probably say so. But given the original conditions, it doesn't sound like you were driving recklessly at all. So the ticket sucks.
Yeah I'm still bummed about the ticket because I know that I wasn't being reckless or even careless. The one truth about the officer's observation is that had I been going slower, then the car would've fishtailed less, or spun out but not gone off the road, etc. Of course then some car behind me would try to pass if I were going too slow, and they would've caused a crash...
I found your blog googling about running socks - thanks for the review! I ended up at sockgeek.com, which is a very helpful retail site.
I appreciate your inquisitive mind, trying to figure out exactly why your Pilot ended up rolling. And you're right: not all AWD systems work the same way. Every manufacturer of course believes their technology is superior, at least in terms of achieving a couple design objectives (efficiency/wasted energy, effectiveness in maintaining the driver's intended path, cost, complexity/reliability, weight, etc).
However, you seem to have left out one of the most important variables in the assessment of the situation: the tires on the car.
Was your Pilot outfitted with four, equal-sized (and equal tread depth) snow tires? At the risk of sounding like a parent, "All Season" tires are NOT snow tires, and pale in comparison to dedicated snow tires in inclement conditions. Additionally, having AWD or 4WD doesn’t negate the fact that without adequate traction at all four tires, there isn’t going to be much improvement over a 2WD vehicle, once those tires’ limits are exceeded.
Even if you were indeed “driving too fast for conditions,” as the officer of the law deemed, you probably could have regained control of your vehicle, if your vehicle was properly equipped for the conditions.
I’m a veteran driving enthusiast (15+ years experience with autocross and track-day events in many different vehicles). I have lost control of FWD, RWD, and AWD cars in various conditions – under the controlled environment of a race course, and in the real world where dodging an unexpected obstacle or inclement conditions cause a sliding vehicle. Sometimes I can reel it in; sometimes not.
I want to mention the fact that WEIGHT TRANSFER plays an important role in vehicle dynamics, no matter what type of AWD system the car is equipped with, or if it’s FWD or RWD, or even if its inherent weight distribution is affected by the vehicle being front-engine, mid-engine, or rear-engine.
Regardless of the vehicle, when the front tires are no longer able to maintain the driver’s intended path (i.e., understeer), it may help to smoothly shift the car’s weight forward to help those tires bite in to the road surface. When the back tires of a vehicle are slipping (oversteer), a gradual increase in throttle to shift the car’s weight rearwards can help counter that condition.
With your specific accident conditions – the strong crosswind, and most importantly, the dramatically different traction conditions of the “snow tracks” versus the untracked snow – the puzzle of determining what driving actions might have saved the day is even harder to solve (other than slower overall speed).
Also, even if you did all of the “right things,” but took correction action just a touch too late or too vigorously, you might have made things worse, as opposed to just letting go of the steering wheel entirely and letting the front wheel unwind naturally to where they “want” to be (believe it or not, this can actually work in certain situations).
In any case, I’m glad your family (and dogs!) were not injured in the incident.
Thank you,
Hsun Chen
NJ native, now living in Colorado (where nobody knows how to drive in snow, and the cars spun off the road in snowstorms are typically SUVs, Subarus, and Audis)
Dog owner/lover, writer, runner, and driver (not necessarily in that order)