Stuff my dog says; on second thought I'll just build a race car

Not to send you off chasing squirrels.... but what's your plans on the body... Steel, aluminum.... composite?

My plan is to do sort of a semi-unibody construction where we'll make files and cut sections on the plasma table to weld together as frame rails, have a second tier above that of tubular steel to make bars for the dash and rear structure, have a sub floor of square tubing that will double as driveshaft loops. Then weld a steel skin to all of that that goes up to the rear edge of the hood with a removable cowl section to be able to access brake and clutch reservoirs. The section from the cowl to the radiator surround will be a four panel job with two piano hinged top pieces that go down to header height, and removable panels from there down to frame rails.

My thinking is that I can get away with smaller everything doing it that way but still have good stiffness. Also the engine and transmission output flange will be rigidly bolted in and should add torsional stiffness that way. The first top down drawing I did was based on 2 inch wide lower rails, but having thought about it I'm thinking now the lower section could be 3x1.5 boxed and then have 1 inch secondary tubing with a decent bit of trussing. The car should weigh very little really even with 700 pounds of engine and transmission, and the tires will be skinny so the loads on everything should be pretty light.

I toyed with the idea of just doing steel space frame and riveting aluminum panels to that. One of the other guys has a Willys Wagon project that we're gonna to a tubular space frame cut the floor of the wagon out and hang the body over that frame though, and I kinda want to one up him. I hate fiberglass so someone on Facebook asked about that and I just laughed. There is a quick sort of back of the napkin sketch around somewhere showing a rough idea of the tube sections I drew a couple weeks ago I'll try to take a picture of for posterity. (It was a 20 second doodle when one of the guys at work asked me to explain the idea)

Always happy to answer questions. I enjoy the questions and feedback, it's always helpful to have other perspectives or just comments to get me thinking.
 
20210128_070534_HDR.jpg

Here is that quick sketch. Looking at it I don't even know how much of it is even relevant any more.
 
While this may be "boring technical crap" to some philistines, it's gold to me. Keep it up!
And for one I like reading about others spending their money! Some may say some of your work/purchases sound like putting "The cart before the horse" but then they probably would have said that to Ferdinand Porsche or Don Garlits among others!
 
Guy brought the piece of glass by today so I've got that now. It's probably three times what I'll end up using and it's probably at least a year and a half until I'll need it, but it was free so I'm not being picky. I forgot to include that I got a free electric fan the other day too. All I need now is some of the other expensive stuff to find it's way to me, like a radiator and a fuel cell and stuff like that. Not really expecting those though.
I feel like I'm kinda in limbo right now because of the uncertainty relating to the wheel/brake setup. This morning I was laying in bed thinking about the shifter set up. The transmission is a top mount shifter, the shift forks are mounted in the case and there is just a plate with a shift lever on a pivot that bolts to the top of the transmission case, really simple stuff. The shifter that is in it is some 20 inches too far forward though so I need to decide how to remote mount the shifter back where it will actually fall into hand rather than being up by my ankles. It will probably need to be a handful of inches off to the right of the centerline of the trans too.
A cable set up would be the easiest way to do things, can probably even buy a set up like that for a little bit of nothing. A rod and pivot set up would look cooler though. With everything rigidly mounted it should be relatively easy to accomplish once the other stuff is place. Want to do a gated shifter with things too so that I can do a reverse lock out. It's too easy driving three speeds now to forget that reverse is where first is for every normal car for the last 40 years. So the thinking there is with a gated shifter I can put a hinged piece that blocks the shifter going into reverse and had it hold down with a spring. Something so that it takes a conscious action to engage reverse. It may be overkill on this but it seems like a good idea at the moment. Test driving floor shift three speeds for work I typically put them into reverse at a stop sign at least once every drive. It'd be cool to be able to do something to move reverse away to have it beside first or third but with it sharing a fork in the trans I don't have a way to do that.
 
Time for another long winded and poorly drawn answer to a question nobody has asked. Why do I need the brakes before drawing the front suspension. To answer that I poorly drew an MS Paint front suspension mock up from a terrible handling car. Something like a GM car from the 60s.
You may have to use your imagination for this.......

terrible.png

Showing the zero offset wheels, the downward inclination of the upper control arm with relatively flat lower control arm, and the ball joints being largely one right on top of the other. I didn't bother with the steering arms in my crummy drawing, they're even more terrible if you're curious. What happens with the control arms placed this way is the camber of the tire works against cornering grip, as the suspension compresses under a cornering load the tire cambers out and the contact patch of the tire picking itself up off the road so only the outside shoulder is holding the vehicle. If the suspension is unloading as in the inside tire it does the opposite and cambers in, against limiting the contact patch on the road. It's backwards from the way any modern race car suspension is going to be designed or work.
There are things that are done on those old cars to try and mitigate the problem, extra tall ball joints and extra tall spindles. The idea behind them is to decrease the angle of the upper control arm and cure the backwards camber arc. It's easier to just fix the inside mounting point of that upper arm if you're designing the vehicle yourself.
better suspension with line.png

Here is an equally lousy drawing of a better control arm arrangement. The upper arm goes upward to the spindle, and is shorter than the lower arm. This type of arrangement causes the tire to camber in under load and camber out when unloaded keeping the contact patch more parallel to the pavement. I've also drawn the wheel with a massively positive offset in this because it can show another aspect of things which is keeping the steering axis inclination, which is the imaginary line made by the ball joints, pointed towards the center of the contact patch.
simple arm compare.png

Made an even more crude representation of things to sort of exaggerate the difference between the two control arm angles and how they operate under load.
simple terrible with ground.png
simple better with ground.png

More crappy exaggerated stick drawings showing the idea of keeping the contact patch level with the ground. The trick in designing things from the ground up is that you can put the arms in whatever position you want, but arm lengths and angles dictate this arc of the contact patch. Obviously you want to have the tire close to flat when the vehicle is going straight, not completely flat mind you but mostly. The way things work is that the anchor point of the upper control arm is pretty much fixed laterally but you can changed the angle of it by changing both the height of the frame and the height of the spindle. The anchor point of the lower arm is pretty much wide open too. It will go to a cross member that gets put pretty much where I want it to be for the purposes of locating that arm.

The point where the upper and lower arms are attached to the spindle are pretty much laterally fixed too. Going back to our terrible 60s suspension from earlier compare the steering axis inclination of it to the drawing of the good suspension.
terrible with line.png

With the ball joints one on top of the other the imaginary line is nowhere near the tire. Making this line essentially a vertical line does two things. First is that SAI causes the tire to camber in while turning, this again helps keep the contact patch flat to the ground provided it's not extreme to the point of causing the tire to camber all the way to having just the inside edge of the tire on the ground. The other is that it limits self steering torque on the spindle. Putting that imaginary line to the ground inside the center of the tire is like the caster on a shopping cart being stuck pointing out to the side in a way. There isn't the drag of a sideways tire, but the rolling resistance of the tire and wheel bearing and brake drag and the bumps in the road all slowly ad up and try to cause self steering. One way to counter act this is having a large amount of positive caster to the front suspension, basically make the suspension always try to steer itself straight ahead to overcome any other incidental steering input. The other way is to make the steering so heavy that it takes an enormous load to actually induce any self steering effect. Neither of those things is desirable for my purposes.
Simple rotation.png

A taller tire would make this easier leaving more distance for our imaginary line to get to the center of the contact patch. Taller tires are super expensive though. Positive offset in the wheels make things easier too, we don't have to make the imaginary line go out as far if we just bring the tire in. That's not really an option either, the wheels are what they are, and it ain't much. So in order to clean things up the ball joints, or Heim joints as it will be, have to be as close to the brake rotor as possible and to achieve that I need brake rotors. With brake rotors I can see how far from the chassis rails I can position the ball joints. When I know where the ball joints are I can measure the angle needed of the upper control arm. Once I know that angle I will know how far off the ground the top of the frame rail needs to be. And now a confused dog.
dog on path.JPG


Edit: Disclaimer that I am not nor have I ever been a chassis engineer. The above simply based upon different things I've read and watched over the years.
 
Last edited:
For tonight: How to eliminate bump steer, and why I don't want to do that.

So most people have heard of bump steer and how bad it is. In a sense this is true, bump steer can make a vehicle frighteningly unpredictable to drive. Hitting a bump in the road can cause the vehicle to spear off into the nearest hedge often toward small children and/or puppies. There are some absolute garbage suspension "upgrade" parts out there and it makes my head hurt every time I see a car with them installed. Let's look at the old suspension design. For the purposes of this point assume all drawings are of the front steer variety where the steering arm is in front of the steering axis.
terrible with tie rod.png

I've added a red line to represent the tie rod. The spindles on those old cars were short, the wheels small, and the steering slow. So the tie rod had to be mounted about the lower ball joint in order to not run into the wheels. Conversely the engines sat low in the chassis to allow a lower hood line, so the inner tie rod had to be low enough to allow the tie rods to pass under the oil pans when the steering was turned.
terrible tie rod line.png

With everything simplified down you can see the lower control arm and tie rod make an X viewed from the front. As the suspension compresses the bottom of the spindle moves away from the frame, but the steering arm of the spindle moves toward the frame. This means the tire is turning towards the frame when the suspension compresses, bump steer. Where things really get scary is if your tie rods are different lengths the rate at which this turn happens is different from side to side. If you're driving down the road and even if you hit something the exact same height with the front tires at exactly the same time, you could have one wheel turn in 2 degrees while the other turns in 7 degrees. A lot of the rack and pinion conversions I've seen on old cars position the rack off center making the tie rods different lengths.

Adding a tie rod to our line drawing of better control arm alignment can see better positioning of the tie rod.
good tie rod line.png

The lines obviously don't cross there, so as the suspension compresses the lower ball joint and steering arm move towards the chassis at a matched rate. The lengths of the lower control arm and tie rod need to be matched too, the idea being the lateral movements of those attachment points match as the lines move in an arc around their inner attachment points. Also the two tie rods need to be matched as closely as possible in length, but it isn't as tire wearing a bad toe alignment so tie rod length takes a back seat to toe alignment.
broken line zero bump.png

An interesting way to look at it is that if you extend the upper and lower control arms along an imaginary line through space where those lines meet is also where a third line of the tie rod meets. I didn't do a drawing of the awful suspension like this, but it's obvious it doesn't work with that since the two lower lines cross without the imaginary extensions. So that's how to completely eliminate bump steer.

The funny thing about bump steer though, is it's considered advantageous to handling when a controlled version is designed into the suspension. That's really all passive rear steer is, and it's touted as the main advantage of 5 link or multi link rear suspensions. You can toe the tires out under braking to aid rear stability and toe them in under acceleration to make the car feel more nimble. This can work a little on the front end as well, tire scrub radius is more interesting though.

When you turn the car the front wheels turn, but the inside tire is making a smaller circle than the outside tire. That's why differentials exist, the outside tire goes farther than the inside and if they were locked together the tire would have to slip to make the turn. It does this even with the wheels not linked at all too though, the inside tire is a smaller circle to that tire needs to be turned more than the outside tire. If the tires turn the exact same amount there has to be tire slippage in order for the vehicle to actually turn.

One of the fun ideas behind that then, is that if you turn right you load the left tire compressing the suspension. The left tire wants to be turned less than the right tire, you can use bump steer to make this happen. The lower ball joint moves in as the suspension compresses hopefully, so you want the steering arm to move in just a smidge more. That imaginary line trio needs to get separated so there isn't a single point of intersection, the lower control arm and tie rod should cross slightly outboard of the point the upper and lower arms cross. Less scrub when turning means you have more of the available friction of the tire able to be used for actually turning the vehicle. That's why I don't want zero bump steer.
 
I forgot to add earlier, the deal for the super charger hasn't completely fallen through but it's not longer the great deal. The guy was supposed to go and do the deal with him taking one and me taking the other. He's decided he both doesn't have the time to drive out to get it and thinking now he has no reason for it as he's thinking of going with an old Mercedes Diesel instead of the V6 Honda engine he was talking about before. I could drive down the few hours myself to get one, they're still showing as available, but having to make the drive myself makes the deal not as enticing and at the moment I need brakes more than I need the supercharger.
 
Hi Nash. I don't know about Tennessee, but around here, superchargers don't appear for sale very often. As the guys on "American Pickers" are fond of saying "...the best time to buy something is when you find it." So true. I've missed many deals, never to be seen again, because I thought "I'll get it next time", or "later", or "I'll wait 'til the price comes down...". If you go get it, you'll have it. If you wait, you might not, and wind up paying more for something "close". :shrug::twocents:
 
Hi Nash. I don't know about Tennessee, but around here, superchargers don't appear for sale very often.

I understand the sentiment, and I absolutely felt that way about the engine and rear axle as those things seemed too good to pass up. The supercharger I was looking at was a Series II GM3800 one from the late 90s early 2000s. They actually put them on a bunch of those cars and a quick perusal of Facebook marketplace and Ebay shows several having come up in the previous six months so I'm not TOO worried about being able to find another in the coming year. Can also use what is essentially the same supercharger from the Thunderbird but those are more rare and more expensive. Judging from the prices I'll probably have to pay $100-$200 more for it then but "C'est la vie." In order to move forward right now the needs are brakes for the front and rear axles, Heim joints for the arms, and then shocks. Obviously metal too to glue all the bits together.
 
20210201_200213_HDR.jpg

Got a chassis drawn so to speak, enough to get an estimate for tubing cost and weight. After thinking about things a half dozen different ways I've decided to make the main lower rails out of stacked 1.5 inch round tubing with the middle connection capped on both inside and outside in 14 gauge. Think in cross section a figure eight within an oval. Then all the upper structure will be 1.25 round tubing. Thickness is still a question right now between .120 wall and .095 wall. The two fabricator guys I've talked two gave me different answers, one said .120 the other said .083, so I'm kinda thinking that .095 to get in the middle, or maybe a combination with .095 wall 1.5 inch and .120 wall 1.25.
Dropping from .120 wall to .095 wall with everything drops 50 pounds. The guys I've talked to are both rock crawler/bouncer guys and they both described what I was thinking as pointlessly beefy for something that will weigh less than 2000 lbs. I want this thing stiff though, especially laterally and torsionally. Doing the capped figure eight for the main channels I'm thinking should have the thing extremely resistant to twisting without unnecessary vertical strength. Like I said the other day though I'm no chassis engineer.
Going off of some online calculators, with everything at .120 wall the weight of the tubing comes to about 240lbs, which with the additional plating and cross bars, floor support bars and such will probably grow to just over 300 lbs. The engine is supposed to weigh 575 pounds so I'm assuming that adding the rear axle and trans to that gets drivetrain weight to about 700. Even with steel skin, the super charger and steering components and gas tank and all that I'm thinking 1500lbs dry is a possibility there. As I mentioned doing .095 wall would drop 50 pounds there and get chassis weight to 250lbs or so and make dry vehicle weight under 1500 almost a certainty. Sort of split really, and may just do the .120 because in the grand scheme 50 pounds isn't all that much. My inner Colin Chapman tells me that weight is the enemy though and must be killed at all cost.
I'd love to be able to order metal this coming week and get started building but it'll depend on the estimate. I did order front brake calipers and rotors tonight so onward and upward (with the money spent so far).
 
I like your idea of the stacked tubing (the oval shape) for your frame rails. That, and all the triangular bracing I see, should make your frame rigid as hell. This buggy is gonna handle like a roller coaster! :thumbsup: :popcorn::smoke:
 
20210202_183939_HDR.jpg
This might be my worst picture yet, but driving back to work today at lunch the answer to my shifter relocation question just popped into my head and I wanted to draw something out real quick before I got back to work and forgot. A two piece hinged lever up by the driver is the answer. The part with the shift knob is hinged to the second part which is suspended on a cross rod. The forward backward selector link is heim jointed to this pivot, also mounted to the pivot is the L bracket relay for the side to side select with the bottom of the knobbed piece of the shifter connected to that L bracket.
That way you can transform both the front to back and side to side motions into forward backward motion to get forward in the chassis to the transmission. Mounting that L bracket to the pivoting lever is how you keep the two motions completely separate and that's the problem I was having. How to isolate side to side movement of the shifter from front to back movement.
Up by the transmission there will be a rod and sleeve set up which will be sort of splined together. The rod will have both the second L bracket anchor for the side to side transfer rod and a connection for the front to back transfer rod. The sleeve will connect to the second L bracket which will move it laterally, be splined to the inner rod to cause it to rotate, and have the finger that goes into the transmission to operate the shift forks. Back at the shifter I'll have a gated top plate to limit motion of everything.
I have the forward L bracket drawn backwards up there, the way it's shown first and reverse would be on the right instead of the left. It should have a very cool precision mechanical vibe to it that should fit the look of the car very well. One of the guys at the shop suggested I just use an old VW shifter and I shot that down immediately.
Decided to go with .095 wall tubing for everything. Tried to get a quote on it, but ended up with a quote for 17/24 of the length of tubing I wanted, and it was also for seamless .120 wall, so the quote was worthless and they didn't get back with the correct one yet that I know of. Got a couple of the guys at work excited and wanting to get to started on this. Thinking I'll get in this weekend and tear the transfer case apart to see exactly what all we'll need to sort of bypass it and keep the thing rear drive only. Talking this evening we're hoping to be able to get the thing to a rolling chassis with the driveline installed by the end of May to have it at the big car show here in Nashville. It's probably going to depend on how quickly I can get wheels from Coker though.
 
News of the day on the race car front: Our local outlets don't have any .095 wall ERW tubing. It seems steal tubing of most types is difficult to come by right now for reasons I don't have, though I'm thinking COVID related disruptions getting steel from China?? The options I have for round tubing are apparently super expensive DOM, .120 wall ERW, or .083 wall ERW.
I can't afford DOM so that's out, .083 is worryingly thin, and .120 adds 20% to the weight of everything. Obviously the answer is option D, where I order the 1 1/2 pipe in .120 wall and use square tubing in place of the 1 1/4 round tubing. One inch square is basically the same weight as 1.25 round and has the added bonus of being easier to work with as far as attaching sheet metal for the body panels. The other thing is the fab guys say they aren't sure their tubing bender will bend .083 without wrinkling it but they know for certain that 1 inch square tubing works well.....and is supposed to be even cheaper than the round. The other fun is that the inch and a half .120 wall stuff is bought in 24 foot sticks, and of course I'm calculating 73.7 feet that I need. I wasn't planning on using it for mounts and don't have those calculated in but I guess at this point I may as well, I'll have plenty left after the main frame tubes after all.

I got brake parts today and reality met my expectation in that the brake rotor depth is 1/8 inch more than the height of the hub. I haven't gone through and tried to draw a spindle up yet, but I can maybe make that happen this weekend now. My measurements are fighting me with things I'm afraid. Without having specifically drawn things out it feels like I won't be able to get the steering axis out far enough. It would be nice to have a wheel now to be able to check the clearances there but haven't heard anything on it since I ordered it.
20210204_182544_HDR.jpg
I went through last night calculating angles for the main tubes where they angle up to get above the lower control arms. Also got the angle for the hoop behind the seat. Tonight I drew up a head on picture at the nose of the car, and I'm not sure I like it.
20210204_200747_HDR.jpg
I had widened the nose a bit the other night when I was looking for radiator prices and having thought it would be easier to have parallel frame rails in the front to mount the control arms to. Having it wider isn't something I'm really feeling now. I'm also looking at things thinking the cowl can go lower or maybe do an asymmetric cowl and have a larger radius to the "passenger" side of the cowl than the driver side. I didn't measure the control arms or spindles or anything with too much care on that drawing but I imagine it's fairly close and looking at is the steering axis is going to end up about 2 inches inside the center of the tire.
Radiator selection is being an issue. Straightening the frame rails bought me two or three inches and allows a 24 inch wide radiator because finding one 21 inches wide that isn't a thousand dollars has so far not been a possibility, whereas a 24 inch wide radiator can be had for less than three hundred bucks.
Northern Radiator is located here in Nashville and once upon a time I could've called them up and got a radiator the side I want but they quick taking retail business a handful of years back. I'm gonna keep looking, if I can get a radiator to work I can run the rails straight but pinch the body work in above them. The other option is to simply have a wider border and narrower grill opening.
 
Kept thinking last night after I went upstairs and hopped in the shower. If I drop the cowl height 3 inches I can drop the seat 3 inches and the cowl behind the seat 3 inches. With the way the pipe construction is planned dropping that rear hoop 3 inches saves me a foot of pipe. I can also shorten the overhand in the front as there really isn't any reason to have the frame heavy outside the suspension mounting so same thing. I can shorten that front overhang 3 inches which saves me another foot. Doing both of those gets me under the 72 foot mark, and then some of the rough estimates I did for curves across the rear hoop and rear overhang will get me a little move and should get me comfortably under that 72 foot length which lets me just have to buy 3 sticks of inch and a half.
I'll have to draw that out tonight and see what I think. The order I did a couple weeks ago of the valve cover and exhaust manifold gaskets is supposed to finally get here today too.
 
Back
Top