A Lesson in Vibrations

In engineering school, we had a class called Intro to Vibrations.  In this class, we studied how real life systems will vibrate and how to predict and control them.  We had one very interesting group project that went from a 1 month project to an over-nighter in the space of 29 days.  It ended up being a 12-hour marathon for three mechanical engineering students and a 20-some page paper, all but 6 of which were calculation appendixes.

The problem went something like this: using a given simplified model of a car, supply spring rates and damping rates in order to make the car bounce (vertical vibration) at a rate between X and Y and pitch at a rate between U and W (two different given rates).  The idea is that a car is uncomfortable if the car bounces or pitches too slowly or rapidly when the car goes over a bump.  The car will be subjected to bumps modeled by a given sine curve.  Assume the weight distribution of the car is unknown.

In short, we select a car's springs and shocks based on comfort.  As complex as this was (36 man-hours plus were spent on this), it ignores the VAST, VAST majority of real automotive engineers' design goals (handling, stability, etc.).  That being said, this sort of thing is still important to automakers because cars should be comfortable.



Allow me to diverge for a minute to describe something that happened to me more than once.  The 2nd car that I drove in life was a 2005 Pontiac G6.  I admit, this was not a car I liked very much, its only redeeming factor being that if I drove it like a grandpa I could manage to get the gas mileage up around 36/46 (city/highway).  That and people seemed to think it was a nice car.

Anyway, here was a situation.  I'm driving down a curvy road through south Raleigh, very fun, often taken at or above the speed limit of 45 mph by most traffic despite the 35 mph suggested speed limit.  I was probably doing 48-50, nothing unusually fast.  While going over a certain rock or rumble in the road, my front tires gave up traction, my steering wheel went limp and the wheel snapped into the direction I was pushing it.  I, after my years and years of video game training, immediately straightened the wheel and braked until I was slow enough to complete the turn.  My decision was the correct one, though the unintuitive one, saving me from plowing into a tree (or five).



I got into this situation doing nothing wrong.  The suspension in the Pontiac just didn't like that particular series of waves or bumps in the road and the car lost traction.  My strong suspicion, having been through some of the basics of auto engineering, is that Pontiac didn't spend the engineering hours on this car to prevent this sort of thing.  I believe that many companies cut corners in any way that doesn't effect the bottom line, and they know that the last decision made before a purchase is related to the test drive.  The trouble is that test drives do not cover most driving circumstances, and in practice rarely include any driving that reveal the strength of the underlying car.  Instead, most test drivers focus on how comfortable and nice and rich a car is.  Therefore, the best way to sell a car to someone who is test driving it, is to make the car as smooth and comfortable and generally nice as possible.

This means that the most important thing to a suspension engineering team run by a company that wants to sell lots of cars is the extended version of my college project.  I don't mean to say that comfort is not important, but it's not what's most important.  What's most important is that the car safely and reliably gets you from A to B (in saying this I'm quoting every person who tells me "I don't need a race car, I just need something to get me from A to B").  Comfort and style are, at best, a close second.

I'm hoping my statement is already becoming clear here, but as I'm likely to ramble, I'll simplify the purpose of this blog entry to one statement (good luck to me).  Car companies often cheap out on engineering hours that do not lead to the sale of a vehicle and/or only spend time and money on making a car seem nice.

It's just something for you to think about.  The kinds of car companies that I and many other enthusiasts often support are companies that are willing to go the extra mile to make a car a better car.  Yeah, often we've got our heads stuck up our own butts about "true sports cars", but often we have a reason.  I love and support Mazda because they put a great chassis under their car, and I believe that this is more important than having sat-nav standard.

Unofficial Statistical Research

Alright, back to the blog.  Sorry for those who have read this and wished I had kept better care, I'll try again to meet my 1/week goal.

As for my non-statistically accepted research as discussed in a previous post, my purpose is this: to study the number of what I loosely describe as "dip-s**t" errors made, broken down by category.  These are usually mild offenses, ones that likely do not cause accidents, but ones that usually aggravate (me) situations, or are at least a general sign that the driver is, in some way, not on his/her game.  The rest of this post is defining my study such that it's at least almost kinda-close-ish  to resembling a scientific study.  More of a guidelines than anything interesting.

Offenses include (but are subject to addition):

• Attempting to enter a highway at 5 mph or more lower than the flow of traffic.   This offense is particularly grievous to me as it is the cause of the majority of traffic I encounter on the way to work, often doubling my commute.  Gotta love the accordion effect.
• Letting off the gas/braking before the highway exit.  Going 55 mph on the highway before the exit is going to cause traffic problems.  Again, accordion effect.
• Non-use of a turning lane or not fully entering a turning lane before slowing down and turning.  Especially common on 2-lane roads with a center turning lane.
• Cutting corners unnecessarily short and end up running the rear tire over the curb, usually found in parking lots.
• Often in fear of committing the above offense, turning WAY too far out into the road/isle and winding up blocking both lanes.
• Parallel parking at an angle (nose in), parking more than a foot away from the raised portion of the curb, or parking outside of the lines by more than an inch or two.
• Parking in normal spaces over lines, at an angle (usually cars who's fronts are centered but forget that the back doesn't line up automatically), or parking with the butt of the car sticking way out into the isle.
• Driving unnecessarily slowly.  This comes in many forms including slow parking, overly slow reactions to green lights, stopping at a green protected turn or other right of way/clear yield turn.
• Leaving more than 3/4 car length in front of your car when traffic is thick enough for some cars to be stuck in an intersection.  I am picky, yes, but this is a thoughtless action that can result in an intersection being blocked or partially blocked.
• Blocking the box (entering an intersection when there are stopped cars in front of you already at least halfway in to the intersection.
• Pulling up to a drive-through window more than a foot away from the curb, or scraping tires against the curb. 
• Not moving/adjusting to allow someone to merge smoothly onto the highway
• Merging into the wrong lane at a double left or double right turn
• Improper use of left/right lanes on the highway

These are all I can think of right now.  I've picked these as they appear to be in the same classification of thoughtless errors.  Many are less thoughtless and more a lack of spacial knowledge (thinking ones car is larger/smaller than it is) but I will still classify them together as they both come from a lack of thought in driving.

I will conduct this study by bringing a tape recorder with me in the car and calling out pertinent information when seen.  Yes, this is a horrible way to perform a survey, but I'm not going to take this any farther.  I will report back after the end of April.

What do I call pertinent information?  The categories of data to be collected.  As stated, I'm mostly concerned with make of vehicle, however I also will be calling out (only when it happens) a few particular classes of cars.  These include small SUV or SUV-like in stature vehicles that I deem to be ill designed for off-roading (think Hyundai Santa-Fe and Scion XB).  Also included are mini-vans and pickup trucks.  Admittedly, I only know what I already expect in regards to style of vehicle, also that this is, again, horribly un-scientific.  I hope that my point 1 month from now will show itself.

I admit to a bias going in.  I will reveal the answers when I complete the study and I plan to be honest as to whether or not my bias was correct then and only then.

Again, all very unscientific and picky.  Lets meet back in a month.