Stock Car Science

After getting in late from Columbus, I made the mistake of listening to the radio as I attempted to sleep. Listening to the comments about the Red Bull Racing penalties got me slightly riled. NASCAR came down hard on Red Bull, including 150 driver/owner points, a $100 kilobuck fine and indefinite suspensions for the car chief and the crew chief. My preliminary thoughts about what might have happened and why turned out to be pretty close to what I’ve been hearing through the grapevine. But there are a couple big misconceptions running around from the sound of it. Let’s take a look at a few of them.

The penalties were so high because this is a safety issue. This is not a safety issue. We’re talking about twenty-five mils of steel versus twenty mils. (A mil is a thousandth of an inch.) The body is for aerodynamics, not strength. The strength of a race car comes from its roll cage. NASCAR does have a rule that you cannot race without a doorskin; however, that rule is probably outdated. In the new car, there is a 90 mil thick piece of steel in the passenger side door and a sheet of Tegris (the splitter material)in the driver’s side door. Those two components are there to prevent anything sharp from coming in between the door bars and hitting the driver. A twenty-five mil sheet of steel isn’t going to protect you from much and losing five mil from that isn’t going to make a noticeable difference.

The reason for the stiff penalty is that NASCAR is escalating penalties each time they catch someone messing with the chassis or the body. If you look at the penalties this year, they’ve steadily gotten worse. The exception is a 25-point penalty for the 12 car; however that was a height violation, not a chassis/body violation.

How could the driver not know that this was going on? Easy. With a very few exceptions, what the driver knows about the car is what the crew chief tells him. Some drivers ask a lot of questions about what springs and shocks and steering box are in the car. Few drivers spend a lot of time at the shop. They are usually so busy with appearances and such that they come from their motor home, spend a little while talking with the crew chief and get in the car.

There have to have been a lot of people who knew about this because acid dipping pieces of sheet metal isn’t easy to do. Acid etching has legitimate uses, some of which are applied to rather large parts. For example, if you want to weld or braze something, or coat a piece of metal with a decorative coating, an acid etch gives you a clean smooth surface. The words “acid dipping” conjure up a vision of a mad scientist with the vat of boiling green liquid. Solder flux is a type of acid etch. Stainless steel etch is usually a mixture of hydrofluoric and nitric acid. You could brush it on, leave sit for a little while and rinse it off pretty easily. It is, of course, possible that a lot of people knew, but the process is not a complicated enough thing to require that a lot of people be involved. We used a similar etch to clean stainless parts for a sputtering system. I remember as a graduate student leaving a couple shims in the etch for too long and coming back to find that they were totally gone - etched entirely away.

In my first post, I just used the side of the car as an example. Remember that you’re not just trying to make the car lighter - you’re trying to make it bottom heavy. The most obvious place to thin the metal would be the roof panel. If you were clever, you would thin only the center section of the roof because if you thin the edges, you might have problems in welding, and if you mess with the edges that can be seen (the bottom of the doorskin, for example), it would be easier to detect. If you estimate the roof at 4 foot by 4 foot (again, round numbers just as an easy estimate), and uniformly decrease the thickness to 20 mils, you’re saving maybe three pounds. But it is much more significant saving three pounds at the very top of the car compared to saving three pounds at the bottom of the car. The total numbers I’ve heard say that the car was somewhere around 12-16 lbs lighter. Where that 12-16 lbs was missing is very important.

How could the crew chief not know? This originally bothered me as well, but remember that the overall weight of the car is 3450 lbs. You’re talking about 0.4% of the total weight. Given everything else that can change on the car (and everything that has to be done before going to the track and at the track), I’m convinced it could be overlooked pretty easily.

Red Bull management must have known since they aren’t contesting the penalty. When something like this happens, teams usually find out pretty fast who was responsible for the infraction. You have two choices then: fire someone publicly, or do as Joe Gibbs Racing did when they were caught in the Nationwide Series. JGR simple said that they know how it happened, and that they believed the people involved deserved a second chance. It’s just not a NASCAR thing to identify some guy who works in the shop and throw out his name to the press. What the RBR statement says is that they know they were guilty.

I will write a little more about center of gravity in the future to explain why the location of the missing weight is so important.

The stock car science blog has been a little quiet lately, mostly because I’ve been working on a really exciting project I hope to be able to tell you all about in the very near future, but also because I’ve been traveling all over the country giving talks and because there hasn’t been a whole lot of science-related news in NASCAR lately. I was just joking in my talk yesterrday that I was sort of hoping someone would try something clever just so that I’d have something to write about. And voila…

Fox Sports reporter Lee Spencer is reporting that the No. 83 Red Bull Toyota, which was selected for random testing after Martinsville, was found to have not met the minimum thickness requirements for the body panels. Spencer anticipates that NASCAR will be levying “record-breaking” fines.

mum thickness of the body panels is 24 gauge, which translates to 0.025 inches or 1/40th of an inch thick. Let’s model the side of a car as shown below, as three rectangles with dimensions as shown. Yes, I’m using rectangles to make my calculations easier. I’m considering only one side of the car and I’m ignoring windows.

The area of the sheet metal on the side of my model car is 4771 square inches. Multiply that by the thickness of the metal and you get a volume of about 119 cubic inches of sheet metal.

The density of 1018 steel is 0.283 lbs/in³, so the weight of this much metal is roughly 33.75 lbs.

If you want to thin a material, you have options. You can mechanically polish the metal, for example, grinding away a thin layer. This tends to be difficult to do with any uniformity unless you’re really set up for it.

In the lab, I often need really clean surfaces, so instead of rubbing and sanding them, I etch them. Etching is dipping a material in something that eats away at the material. In the case of steel and other metals, the etchant is usually an acid and this is the “acid dipping or chemical milling” to which Spencer refers. You put the metal piece into a bath of acid and how much material is removed depends on how long you let the metal sit in the acid bath. The metal comes out looking like new. In fact, if you take a wedding ring to a jeweler to be cleaned, what they usually do is to dip it in a mild etchant. You lose a miniscule amount of metal from the ring, but it comes out looking shiny and new.

It would be hard to tell whether a piece of metal had been etched by looking at it by eye; however, a metallurgist can examine the metal (with the paint stripped away, of course) and can analyze the etch patterns. Certain directions in a crystal etch faster than others (see Figure 3 in this paper for example), so determining whether a piece of metal has been etched isn’t too difficult if you have the right tools.

How much weight could you actually save by etching away some of the sheet metal? Let’s say that the side of the car we calculated above was etched from 24 ga to 26 ga (which takes it from 0.025 to 0.01875 inches). The weight just on that side would be reduced by about 8.4 lbs and, making a lot of approximations, maybe by 20 lbs across the car. If the thickness were reduced to 28 ga (0.015625″), you’d save 12.6 lbs. But even someone casually familiar with sheet goods would likely notice that much of a decrease in thickness.

Why would you reduce the weight of the body? The answer is back to our old friend The Center of Gravity or CG. The new car has a higher center of gravity, and the higher center of gravity means more load transfer when the car turns. One reason the new car has a higher CG is because it is taller. More weight up higher in the car increases the CG. More load transfer makes the car harder to turn.

You can lower the center of gravity by adding weight (ballast) in the framerails of the car; however, you don’t want to make the car any heavier than the minimum 3450 lbs. Spencer suggests that the RBR team made the panels thinner so that they could save weight (she claims up to 75 pounds) and then use ballast to make up for the decreased body weight. That would lower the CG. The NASCAR R and D Center ought to be able to determine intent because they no doubt know how much ballast would be appropriate for a regulation car. If there is another 20 lbs or so of ballast in the RBR car, that would suggest that someone knew that there was a significant weight savings somewhere else.

Is this a safety issue? Probably not so much. The strength of the car is in the tube chassis, not the body. You can dent the sheet metal in the body pretty easily. Ask Carl Edwards and Kevin Harvick.

There is a potential complication in what I’ve presented you with here, which is that stock cars have curves. I don’t have an accurate estimate of the surface area of a stock car, so it’s difficult for me to calculate exactly how much weight could be saved in this manner mand whether the 75 lbs Spencer suggests is realistic. My intuition is that 75 lbs would be really difficult to shave off the car without it being somewhat obvious to knowing eyes. I am in Columbus Ohio today at Ohio State University, but I will update this post when I get back to Dallas Thursday and hopefully wheedle some more accurate numbers about the surface area.

UPDATE: Well, that didn’t take long. From what I’ve been told if you figure losing about 5 mil from the thickness, that leaves you with about 6.75 lbs for the side I showed above, so maybe 13 lbs or so is about the weight saving you might expect to see. I can’t see any way you could get 75 lbs.

Juan Pablo Montoya’s pole run last Friday at Kansas was disqualified when his shock absorbers failed tech inspection. The shocks and springs are important components of the supension. A car without a suspension would bounce all over the track. When you hit a bump, the springs compress. When you go over the bump the springs extend back, which keeps the wheels in contact with the track.

The problem is that springs are, well, springy. When you compress a spring and let it go, it extends, then compresses, and just keeps bouncing up and down. Hence the need for shock absorbers.

The shock absorber (like the spring) connects the wheel to the chassis. When a spring compresses, it stores energy. That energy is what is dissipated by friction when the car bounces up and down. You’d like to damp out the spring’s oscillations more quickly than friction allows.

A monotube shock absorber (the type used in NASCAR) has a piston (a disc with holes in it, shown below) moving up and down in oil. The holes in the piston are very small, which means it requires quite a bit of force to move the piston. Punch tiny holes in a piece of plastic and then try moving it through pancake syrup. This movement of an object through oil introduces an interesting way to control force.

The force exerted by a spring is proportional to how far you compress or extend it. (That’s Hooke’s law). In a shock, the force exerted by the shock is proportional to the speed at which the piston moves through the oil. So the force depends on how fast you move, not how far you move. You pick the shock to match the spring and the two work together.

A spring can compress and extend. So can a shock. When the shaft is pushed into the shock, it’s called compression. When the shaft is pulled out of the shock, it’s called rebound. You’d like to be able to adjust the rebound and the compression independently. One way you do this is that the two sides of the piston have different hole geometries. A hole may be large on one side and small on the other, which affects how the oil flows through the piston.

Another way you can tailor the compression and rebound response is by changing the shims (very thin washers) that bend when the piston moves up and down. The shims bend more when the piston moves faster, uncovering more of the holes and allowing more oil to move from one side of the piston to the other. (At very low speeds, a valve in the shaft allows direct flow.)

The key to how a shock works is the resistance to the piston’s motion. The resistance is provided by the oil. Most shock oils are between 2-5 wt. More viscous oil provides more resistance to motion.

Look back to the first picture of the shock. There’s an area there marked ‘gas’. Ignore that for a moment and assume you just screwed the two pieces of the shock together without doing anything special. If you want to try this experiment at home, get some cooking oil and put it in your blender. Turn the blender on - that will mimic what happens when the piston moves up and down. (If you have a French press coffee maker, that would be a more accurate analogy, but a blender has the same effect and is less likely to engender spousal irritation.)

Whirl the oil in the blender for a few seconds and you’ll notice that you have foam. Foam is gas bubbles trapped in a liquid or solid. Stryfoam, for example, is air bubbles in a polymer. The foamy mess you have in your blender was created when the whirling motion incorporated air bubbles into the oil. (When you have bubbles of one liquid trapped in another, that’s called an emulsion, but it’s the same idea.)

There’s a problem when shock oil foams. The principle on which a shock works is that the oil provides resistance to the piston’s motion. Air trapped in the oil makes it much easier for the piston to move. You use oil in a shock because it is incompressible, which means that when you press on it, it doesn’t change volume. When the oil foams, you push on it and the air that’s dissolved in the oil doesn’t provide much resistance. A marshmallow (which is a foam) is a combination of sugar and air pockets. When you press on a marshmallow, the first thing that happens is you press all of the air out of the air pockets and it’s pretty easy to do that. Only after you’ve squished the air out do you start to compress the sugar that forms the rest of the marshmallow. A shock has to be filled with an incompressible fluid for it to work. Foam isn’t incompressible.

You have to pressurize a shock. One reason is because oil sloshing around the inside of the shock won’t provide much resistance for the piston. If we fill the top part of the shock with an overpressure of air, the pressure above the oil will prevent the oil from sloshing. Air is about 21% oxygen, with the rest primarily nitrogen. Oxygen is more soluble in oil than nitrogen, meaning that it is easier to dissove oxygen in the oil than it is nitrogen. Depending on the type of oil, the difference can be a factor of two. This is why nitrogen is used to pressurize shocks. Nitrogen gas is less likely to create foam than air. In fact, if you press down hard enough on the oil, you can actually decrease how much gas is dissolved in the oil. You literally press the dissolved gas out of the oil.

NASCAR allows the rear shocks to have nitrogen pressures between 25 psi and 75 psi. Apparently, the rear shocks on Montoya’s car had a pressure of 85 psi. There was an interesting discussion on NASCAR Now with John Darby in which he pointed out that after the initial overpressure was discovered, they allowed the shocks to cool to ambient temperature before re-measuring the pressure. Why? The ideal gas law in action. When gas gets warm, the gas molecules in the tires move more rapidly, and that increases the volume and the pressure. The same thing happens in your tires.

NASCAR wanted to give the team the benefit of the doubt: perhaps the shocks had gotten so warm that the pressure had increased beyond the allowed value. I made a quick calculation (remembering that the 75 psi/85 psi are gauge pressures, so you have to add 15 psi of atmosphere to those numbers, converting degrees Fahrenheit into kelvin, and making the assuming that the change in volume is negligible) and I estimate that the temperature of the shock would have to rise about 60 degrees Fahrenheit to create a 10 psi change in pressure. It’s not at all unreasonable for the shocks to reach that temperature during two laps of qualifying, which is why NASCAR waited until the temperature of the shocks had come down before making the measurement.

What difference would an overpressure make? When teams realized the importance of the car’s attitude in terms of aerodynamics, the primary job of the shocks became keeping the rear end of the car up in the air to get maximum downforce. Higher pressure in the rear shocks could be used to keep the tail end of the car in the air longer. The upper pressure limit used to be 175 psi. That was changed after the 2005 fall Dover race, where two Hendrick cars finished 1-2. The Hendrick cars were set up so that the rear of the car didn’t come back down very quickly after a bump. Both the cars failed the post-race max height inspection after half an hour of waiting. The language in the rule book now requires the shocks to return to their normal position after compression within "a reasonable time", and a maximum value for the nitrogen pressure. A really high nitrogen pressure prevents the shaft from returning quickly.

There’s another reason for a maximum pressure limit. A closed container with a very high interior pressure is also commonly known as a bomb. If there were a failure in the threads or (more likely), the seals on the ends of the shocks, you could have shock parts flying out without warning.