Unca Frank's: Some Important Chassis Relationships & Dimensions

2-Dimensional Relationships and Dimensions. We can start with a top or "plan" view of a simple chassis. Understand that we're not talking about unachievable absolutes here, either in relationships or dimensions. Rather, we're looking for achievable minimums, based on whatever reasonably accurate measurement methods we have on hand. Some measurements, as you'll see, have to be approximated (or interpolated), and some important conditions are affected by a combination of chassis relationships. So let's take a look at the elements of Figure 1.0, below, and see what they mean.
		90° - In the 2D plane, there are two important component relationships where the elements should be at true right angles to the center line of the chassis. These elements are the drive axle (B to C at left) and the wheelie bar axle (D at left). They are, as you already know, actually 3-dimensional components as well, in that they (must be or) can be at, above, or below the plane of
the chassis. For purposes of discussion, however, it's convenient to ignore those facts for a while and concentrate on the 2D for a bit. So, you're wondering, what's the point of insuring that this relationship is accurate? Basic straight-line tracking. Slot drag cars have only one meaningful external directional force acting on them: the guide in a (hopefully) straight slot. The less side load on the slot the chassis creates or imparts, the better and more consistent the car will perform given the power and traction available. Beyond simple (actually, not all that simple) guide misalignment and slot/track imperfections, if any, the most common cause of slot "loading" is drive axle misalignment ( for additional guide misalignment discussion, see "Where a Lot of Other Stuff...", below.

	Figure 1.1, left, shows an exaggerated example of what tracking variations might occur if poor design or assembly were left uncorrected. This isn't normally a problem with most commercial chassis of a "1-piece" nature, where the entire layout, from guide hole to wheelie bar axle plane, is cut with a computer program. With one lone exception, regardless of basic design, I've never seen or measured one where the slots and corresponding tabs for axle location were off-plane enough to show up on the devices I use to measure such things. That one lone exception also featured main chassis and wheelie bar structures that varied a great deal in width, side-to-side (and not, counter-intuitively, where you might think some additional width might help), so I figure they just had a bad day at the old CAD station and never
bothered to measure the finished product. Besides, lots of people successfully race them, so what do I know, right? Where most people run into a problem in this area is in building their own wire or tube chassis, or in the process of assembling a kit that requires a bend, rather than a straight run, from the motor box to the nose piece. While kits that have equal-width nose pieces and motor boxes are pretty much straightforward - they're basically self-aligning unless you completely miss the point - a little more care must be used on the tapered variety. In the absence of an accurate chassis jig (ideal, but occasionally expensive and sometimes not nearly adaptable enough), there are still a few ways the average builder can try to establish reasonably proper alignment. Leaving the wheelie bar alignment for a bit later, let's look at a few other 2-dimensional chassis relationships using the references from Figure 1.0, above.

top

A-B/A-C - It helps to think of the chassis true centerline as also being the ideal traction or "thrust" centerline. If you have some difficulty in visualizing the effects of offset traction, recall what happens when you launch a car with one drive wheel completely loose. Exciting, huh? Well, with a bit of work, you can almost duplicate this entertaining reaction by ignoring where the drive wheels are located.

	There are two primary ways to mess this up; one is correctable most of the time, and one not easily so. The least correctable way is to locate the rear uprights, which carry the chassis drive axle bushings or bearings, at different dimensions from the chassis centerline. Figure 1.2, left, shows an exaggerated example of what you end up with if this occurs. With a little work, you can also incorporate at least one of the 90° alignment problems as well. Again, this isn't usually much of a problem with either commercial one-piece chassis or kits, but has the potential to become one when scratch-building or modifying a variety of manufactured components.

Let's assume, for purposes of discussion, that we're either scratch-building a chassis or attempting to determine the centerline of a kit component for checking or modification purposes. We need to find out whether what we're working with is symmetrical, so what we do or add will also be that way. So let's look at a few ways we can accomplish that, working with the tools and knowledge we have.

	How to determine a center line. One of the most important things you can do is to establish the true center line of all the major kit or chassis components before you begin assembly. To do this, you have a few options. The first option is to use simple geometric division (remember that stuff?) by using a little Dychem and a compass or divider with scribing points. The other method is to use a caliper for total width at any symmetrical point on the piece and divide by two for a center point. In both cases, two (or more) center points reference one center line. This is actually one thing in chassis construction that takes longer to describe than it does to do. A
caution, however; that center line is only as accurate as the basic symmetry of the part you're measuring, as well as your ability to use exactly the same opposing points when doing so. The good news: this being slotcar drag racing and not space shuttle construction, both ways generally get you well within the ballpark of acceptable tolerances. Figure 1.3, above, illustrates both methods. Use the one you're most comfortable with. G/H - And *the other* way to mess it up?** Lots simpler, but almost instantly correctable. If you look again at Figure 1.2, above, it should occur to you that you can accomplish this very same condition by the simple use of axle spacers. Given that the axle is perpendicular to the centerline of the chassis and its supports are properly located and equally offset from that centerline, it follows that any spacers used for whatever reason should equal the same dimension side-to-side to maintain that relationship. This is usually not much of a problem during construction, but haste during that at-track thrash can cause some axle spacers to head for "Spacer Heaven" (that being the place where I presume all those axle and/or armature spacers I swore were on the bench or track pit table went when I wasn't looking). One also has to be careful when assembling a spacer "package" when dealing with an outside gear chassis. I occasionally convert an innocent inline chassis to this configuration when faced with a space or clearance problem. To make up for the width of the gear and any other clearance spacers inside or outside that gear, I measure the entire package as it is installed on an axle, less the width of the spacer nearest the bearing/bushing, and make a spacer from aluminum tube exactly equal to that dimension. It's not hard to cut and file this to within .001" or so of the exact number, and use it with exactly the same inner spacer to come up with the same wheel/tire offset dimensions side-to-side. Or, at least, close enough for this application. Speaking of close enough, two additional thoughts. First, after going through this drill, it doesn't hurt to actually measure the offset, tread sidewall-to-chassis rail, after assembly. Not all tires of all kinds are precisely located in relationship to the wheels they're mounted on. Second, if any of the above makes sense to you, at least in regard to trying to build a car that runs dependably and predictably, then reversing one wheel on the axle to try and obtain more clearance to the body should make almost no sense at all. At best, it means the chassis isn't mounted symmetrically within the body. At worst, it means the chassis isn't actually symmetrical, period. Either way, finding another solution to the problem is always a better idea (see Figure 1.4, below, as well).

top

D, E/F - So what about wheelie bar alignment? I have some pretty strange opinions and theories about the actions, reactions, sequences, and duration of events during a slot car drag chassis' brief action phase (translation: the stuff a chassis really does during a pass, as opposed to, say, the Oral History handed down by people who never stopped to think about it in the first place). One that isn't so strange is the effect that wheelie bar alignment has on chassis tracking. No matter how hard they may be loaded during a launch, nor how long they may actually contact the track, if the wheelie bars impart any side load or directional "steering" while they are in contact with the track, they're affecting (or at least trying to) the directional stability and tracking of the chassis. The technical term for this is "no ******* good." Granted, the friction coefficient of the average sleazy O-ring used as a wheelie bar tire isn't all that high, not to mention not having all that much surface area in contact with the track. So let's ignore what I think here, and ask, instead, this question: if you think they do something (or anything) important beyond keeping the guide in the slot at rest, doesn't it make sense to insure that if and when they do, it doesn't counteract what the tires and chassis are (hopefully) trying to do? In the two-dimensional plane, at least, this is fairly easy to both measure and correct. Unless you're aware of a chassis trick I'm not, where steering with the wheelie bar axle and its wheels is desirable, make sure that axle is at least parallel to the drive axle. This doesn't eliminate the possibly of preload in a three-dimensional sense, but it gets you half way there. See the wheelie bar segment within "3-Dimensional Relationships," below, for additional thoughts on this area.

top


Some 3-Dimensional Relationships. Now let's take a look at the other component relationships our chassis have, those above, if you will, the two-dimensional plane

		First, let's take another look at the uprights (sorry, but I refuse to call them pillow blocks. If the function they perform is called an "upright" on a real car, I can call it an upright on my stuff, so bear with me here). While, as you might imagine, the use of a solid axle - as opposed to an articulated one, not a hollow one - makes sure the wheels are presumably concentric to the axle, the alignment of the uprights in relationship to the chassis is what determines the relationship of the chassis
bearings/bushings to the axle. Translation: one answer to the question "Why can't I get the ***** axle through the **** bearings!?" Making sure this relationship is a true and accurate 90º one can be a little harder that it first appears. Or, sometimes, a lot harder, depending. Figure 1.4, above, shows you what we're talking about. Unc's "Best Guess" Method** On chassis kits, assuming the uprights mount flush with the very edge of the kit motor box, I build the motor box/uprights assembly first, using blocks on my building surface - a piece of tempered glass - to get me close. After each inner upright-to-motor box joint cools, I check the parts with a machinist's square to make sure I'm as close as I can get. After both uprights are soldered in place, I tie them together with whatever manner of transverse bracing I plan to use, and test the bearing/bushing holes with both a diameter "fit blank" and whatever type of axle and bearings I plan to use. Once I'm through checking, by, for example, measuring the top-of-axle to bottom-of-motor box dimensions on both sides, I proceed with the rest of the chassis construction. On those chassis where the uprights are located in from the edge of or within the chassis rail, I start about the same way, soldering the inside joint only, and measure the distance from the top of the upright to the edge of the machinist's square blade. This is not a situation where "eyeballing" is usually close enough; I sometimes start that way, only to discover that a) the caliper hardly ever lies, and b) it sure looks different when viewed from the back or the other side. As you may have already discovered, properly installing two simple uprights can occasionally get to be a trial of patience. Bear with it. Your bearings and/or bushings will thank you for the time you spend making sure they don't die a premature death. But Wait! There's More! (sorry - always wanted to say that on this Site) There are certain conditions and circumstances where all the measurement in the world won't get you the bearing alignment you're looking for. Most particularly, this frequently occurs when dealing with rather large and complicated uprights done in thin material. Some of these pieces, no matter how they're manufactured, have a tendency to curl or bow slightly in some rather odd planes. One solution to the problem is to identify any such condition before you start soldering stuff together. Thereafter, you have a few options. You can slightly alter or "reflatten" the parts (you wondered when I was going to get around to hammering on these things, didn't you?), either by hand or via tools, until it's accurate enough. Alternately, you can determine where and how much the part varies from dead flat, and install it so that its most critical feature, the bearing carrier bore, is within acceptable range. My approach is usually a little of both. Thankfully, this isn't all that common a problem since most manufacturers became aware of it and resigned accordingly. It is, however, something to keep in mind when you're building.

top

			There are a few more variations of misalignment that you should be aware of. With the variety of chassis kits currently available, not to mention the similar wide variety of chassis materials you might use with them, you can maintain some accurate geometric relationships while altering others at the vary same time. Figure 1.5, left, shows one such variation during kit building, wherein you alter the relationship of
the motor box to the chassis rails by attaching the box to the top of one rail and the bottom of the other. Yes, I know that usually requires one to flop the chassis over and solder it on the other side, but trust me, it happens. Don't. Your chassis won't like it. A lot. See the guide and chassis plane alignment sections, below. There are some other simple material-difference problems you can avoid by simply remembering what you're doing, measuring, or both. Common wire sizes involved in building slot drag chassis include .046", .055", .062", and .078", while current tubing sizes are limited to .064" and .072". If your chassis utilizes more than one piece for its rails, unless you're the Jerry Bickel/Warren Johnson of slot car drag racing and know more than most of us do, make sure that both (or, in some circumstances, all) pieces that are supposed to be symmetrical to the chassis centerline are the same diameter. Don't laugh at this one either; it happens, and it's normally accompanied by the standard "Why can't my car make just one decent pass?" question. One other area where a similar displacement problem can occur is the relationship of the nose piece to the motor box. This is usually discovered when the wheelie bar axle is about to be soldered on and one notes that that end of the chassis is either way up in the air or almost below the track surface. If you look at Figure 1.5, above, one more time, and envision the nose piece flat at the top of the rails and the motor box similarly located at the bottom, it doesn't take much alteration to guide-to-axle length or wheelie bar length with the right combination of chassis and rail thickness to put the wheelie bar end of the chassis well outside conveniently "solderable" tolerances. Alternately, if you think about it, this can actually be a solution to anticipated problems in this area. Either way, make sure what you end up with is what you want to end up with.

top

			Other Kit Material Diameter & Thickness Situations. It's easy to forget that all kit pieces, despite, say, having exactly the same measured thickness, do not necessarily maintain the relationship you want when used with certain chassis materials. Let's look at Figure 1.6, left, as an example. The first condition shows an assembly where the thickness of the chassis kit material is greater than the radius, or ½ the diameter of the
tube or wire it's being used with. This a) puts its "contact point" at the outside edge of the cylindrical cross-section, and b) means that any other structure or assembly at that very edge, e.g., an upright or pillow block, can still maintain a perpendicular or 90º relationship. This condition also doesn't affect either the inside-to-inside dimension or the nature of the solder joint. When the material of the kit, however, is less than the radius of the tube or wire, there are a few subtle changes that you should be aware of if you're not already. As illustrated in the example at right in the Figure above, you'll note that the chassis kit piece now functionally slips ever-so-slightly under the horizontal radius of the tube or wire. While the actual physical distance involved is relatively small (on the order of .002" to .004", and if some of us hadn't been reading car magazines instead of paying attention during Plane Geometry, we'd be able to precisely calculate and diagram if for you), it can have a meaningful affect on any 90º assembly at that juncture, the inside distance between the two chassis rails, and the nature of the solder joints that link them all together. Practically, even though we're talking some thousandths of an inch here, this difference can occasionally explain why two chassis kit pieces of the same width can generate a noticeable widening of frame rails down the length of the chassis. Solutions? If you look at the illustration, you'll note that the objects that are practically involved are the chassis uprights at the edge of this assembly, trying to maintain their location and both a perpendicular and parallel relationship. Since the actual, functional solder joint strength difference of a kit piece, such as a motor box, at the edge of or slightly under a tube or wire is negligible, one might be tempted to do, uh, a little grinding on the suckers. Possible, but perhaps not a great idea, in that you're talking about reducing the thickness of a piece that's already perhaps only .025" thick in the first place. The remaining options are to make up the difference by slightly reducing the cross-section of the tube/wire where it meets the upright, or to simply ignore it. Yeah, I like that last one, too. An old Unca Frank motto: "A good-looking solder joint can cover a multitude of sins." Most of the time, anyhow. Since I usually work with stainless steel tube rather than wire when building kits, whether or not I carefully diamond-file a .004" deep flat on the side of a .012" thick tube wall depends on how the rest of the solder joint(s) on the motor box seem to be fitting. If it looks like there won't be much "bridging" of gaps with solder, it goes together as is. If there are some visible gaps, the file comes out until the gaps go away. And no, I have no clue why I can build the same kit with exactly the same dimensions three times and have three different conditions. Uh, how about "craftsmanship?" "Good Old American Know-how?" What I do know is that gently notching the frame rail has no meaningful effect on either the rigidity or deflection rate (flex, or, in this case, the absence thereof) of the assembly involved, whether it involves tube or wire. The components involved significantly reinforce one another, creating a more rigid "beam" structure than either of the two elements taken individually.

top

Where a Lot of Other Stuff Beyond Rubber Meets The Road In this final Section, let's take a look at some of the practical effects the situations we covered above may have. Some are bad, some are not bad, and some are... who knows? To start with, how about the "who knows?" part?

			Guide "Preload" This area gets us back to slot car drag racing "oral history." There are those who will tell you the guide should be dead flat in relationship to the plane of the chassis, those who say it should be preloaded up, and those who believe it should be preloaded down. On the cars I have built and run, where possible (and sometimes where not possible), I have tried every variation to varying degrees, and have noted absolutely nothing that would make
me change the way I currently build. Which is, simply, flat, mostly because it's the easiest way to both accomplish and check for accuracy. This range includes both Top Gun and AA/FC, Classes notorious for either braid destruction (TG) or acute sensitivity to braid alignment and condition (AA/FC). I eagerly await those with far more knowledge than I possess to correct the error of my ways, and when they share both the why and the how (e.g., taking a happily flat piece and making it less flat in a manner that one can measure in all planes for accuracy), I'll be more than happy to share it with you. Until then, my understanding of friction, electrical contact, and planar "measurability" being what it is, flat, or zero longitudinal preload, is close enough for me. My personal experience tells me that although I may not be gaining anything by it, I'm not losing anything, either. Like a drag racer I met a gajillion years ago when I first started messing with real cars said, "If you see it on the clock, it's real. Otherwise, it's just a theory you can't prove." Your call.

	Other Guide Conditions to Avoid A fact of slot car drag racing life: all tracks aren't dead flat and all slots aren't perfectly straight and smooth. That means that every imperfection in these conditions your car encounters is seen by the guide first. It's up there, trying to guide the car, pick up the current, and deal with the downforce, if any, the body is generating and the primary and secondary forces the chassis is encountering and the motor is producing. Busy little piece, the guide. Which is one reason, at least, you should pay some attention to how it meets the slot, at least from a "vertical" perspective. As you might imagine, doing all of this is a lot easier if the forces it encounters are roughly even and symmetrical. Figure 1.8, left, gives you a
view of the guide with a somewhat exaggerated side preload. When you recall that a guide does not magically and effortlessly zip down a slot without contacting the sides, several things may occur to you. First, the guide alignment, relative to the traction center line, as discussed above, is really important in avoiding any "hunt" in the guide other than that generated by the track. Second, while the little hummer is going down the track, imperfections within the slot, e.g., at the section joints, will be better dealt with if the guide encounters them in a "flat" condition. If you examine the illustration, you'll note that when installed at an (or much of any) angle, even identical physical joint displacements in the track slot, however minor, would actually contact the guide blade at different height locations side-to-side (since I'm writing this after I did the drawing, it would have been nice if I'd made that a little clearer in the picture, huh? Oh well). The application of force with levers being what it is, and the guide, at least in this condition, imagining it has won a trip to Lever Fantasy Camp, the chassis will see these two identical displacements as two widely disparate forces. O.k., So The Guide is a Little Off. So...? How, when you can't actually see these displacements, can this possibly be meaningful when applied to the average slot drag car pass? Consider: on an average 90 gram slot drag car, there's usually something on the order of 22-25 grams of static weight resting on each rear tire (and more or sometimes less when the car is in motion, depending, but let's not confuse this with the facts, huh?). We have narrow drive tires set at a (relatively) narrow track width that are trying to generate lots of forward motion, with a friction coefficient (probably, as far as I can figure) less than 1. To this mix we add the chassis, courtesy of a misaligned guide, asking the tires to make instantaneous yet random corrections to the theoretical 0º tire slip angle. This of course, doesn't include having to deal with any traction differential imparted when the chassis tries to deflect in reaction to uneven forces the guide is transmitting. Translation: If you're serious about this stuff, on some level, everything regarding slot drag cars is meaningful. It simply depends on how much you choose to think and/or worry about it. Of course, since both you and I always TQ and get .400 lights, spending time worrying about a few thousandths of a second here and there is a waste of time, right?

top

		The Worst Case Scenario Maybe I shouldn't call it that, because, at least in my experience, said scenario is far more common than you'd like to believe. Beyond simple construction problems, storage and handling of some cars will impart various tweaks, twists, and bends that give you the equivalent of Figure 1.9 at left. Perhaps not to that exaggerated degree, but frequently to a degree all but certain to meaningfully affect performance. What this means, among other things, is that no matter what care and skill you use when building your cars, some level of
inspection and preventative maintenance is required throughout their competitive life, at least with Class racing cars, if not all cars, to insure that they stay competitive. At some later date (and at this rate, probably just slightly before the sun cools), we'll take a look at the things you can do while building a chassis that make maintaining it a bit easier in the long run. Until then, pay attention to the little stuff when you can, because whether or not you know it, there'll be a quiz at the next Race you attend. Your little friends won't be grading on a curve, either.

top

All Contents © 2001 Frank M. Eubel. Please Note: permission is granted for not-for-profit (only) printed or Web-based reproduction of this information or any portion thereof in this editorial form provided it is credited "Courtesy of Unca Frank's Home Page" and a link to this site's index or Main page is printed on the document or available within the applicable Web site. f_eubel@juno.com

return to home page