A case study in tinkering run amok… by TD (Troy) Graham
Part 3: Fun with Mock-ups.
Misery loves company:
At this point, a little introduction is in order… where “me” becomes “we”; the other half of “we” being my buddy Tom Wilhelm. He is the President of General Crafts in Goshen, Indiana, which is a fully-equipped fabrication shop that specializes in lower-volume, intriguing design/build projects. A past “Jeeper”, I got to know him during my original rotary engine install, as I was in his shop every couple of days looking for this piece, that part, or “could you form me up something like this (hokey) cardboard template?”.
I think he found he my little project rather amusing at the time.
Anyway… I eventually coaxed him into joining our little motley band of Zooks for a day run at the Badlands, arranging for him to ride shotgun with the legendary Bill “AirZuk” Maulding. Tom wore the biggest stupid grin all day, punctuated only by moments of sheer terror as a helpless passenger at the mercy of a true lunatic. He was hooked, bought a Zook, and enlisted my help to build it up.
From that point, it was all screeching downhill. Tom and I spent the entire summer exploring the merits and pitfalls of every imaginable part of a four wheel drive vehicle, and now that fall was here, it was time to put all our silly ideas to the test.
Although I have access and expertise to model the suspension system on a computer, it is extremely time consuming (and BORING) to do so. Therefore, I decided to just use the computer to work out the basics of the suspension design, and build a mock-up to refine it. Having a spare (junk) vehicle to hack and chop on is handy too… thanks to my buddy Claude Reeder.
Actually deciding on a suspension type took an agonizing amount of brainstorming time, due to a syndrome known as “Option Paralysis”: 3, 4, or 5-link? Wishbone, “W”-link, or Panhard Rod? Locating links above or below the axle? Rod ends throughout, poly throughout, Johnny Joints throughout or a mix of all three? ad infinitum. In addition, the offset diff design of the Samurai axles presents some unique problems all by itself.
I am now convinced that the reason most homebuilt, super-flexy rock monsters are NEVER used on-road is that designing a limber/flexy off-road linkage suspension is relatively easy when compared to the complexities of designing that same suspension to also work well on-road. Which is a shame, as I feel that most of the qualities that make a suspension geometry work well on-road translate directly to off-road benefits as well.
…to locate the links on the axle and frame.
Get it wrong, and the axle will actually lift under torque, the vehicle will have excessive dive and instability under braking, will squat under hard acceleration, will try to spin at the first corner, and just generally become an ill-handling, very dangerous pig. Suffice it to say, the geometry is complex enough with just a little bit of suspension travel, but with a lot it’s daunting.
…to use rod ends (a.k.a.: “Heim Joints”).
Sized properly for the loads, and used per their design specifications, rod ends are incredibly strong and reliable. Their ill-deserved reputation for failure is 100% due to improper use in a severe-duty application. There are extremely specific guidelines that the rod end manufacturers publish that should be followed to the letter, and in two years of observation, I’ve seen more improper than proper use on various homebuilt (and production!) rigs. Can’t imagine why they fail?! One of these days, I’ll write a brief about their usage.
…to use springs with a linkage system.
Springs should have but one function, and only one. The count of the functions shall be one. Two is right out. (Bonus points if you recognize the paraphrase.)
That function is: to support the weight of the vehicle. No more, no less. They are not supposed to be used to limit axle torque reaction, brake dive, or acceleration squat of the vehicle, all of which are the job of the suspension linkages. Placing the springs “inboard” (i.e.: inward a good bit from the stock location) may do wonders for your RTI numbers, but will destroy the on-road handling of the vehicle while making it far tippier in off-camber, sidehill-traversing situations.
…to limit compression and extension of springs.
Whether you’re talking coils or leaves, making them work outside their design parameters is a recipe for sagging, weak, dangerous springs. They can let go at any time, possibly injuring bystanders or causing loss of vehicle control.
Coil springs are designed and wound to either work in compression or extension, but NOT both. All vehicle coils are the compression variety and forcing them to extend beyond their unloaded (“free”) length will significantly shorten their life. In addition, compressing a spring by more than half its free length will similarly damage the spring, causing it to sag very quickly. For example, a coil spring that is 14″ long with no weight on it should only be bumpstop-limited to 7″ of compression, and limit-strap-limited to 14″ of extension.
Multi-leaf springs are also designed to work in compression only, with the same ill effects as their coil brethren if used in severe extension (“droop” or “negative arch”). There is no hard rule as far as limiting compression is concerned, as the specific spring design can effect the rating significantly. A “Rule Of Thumb” is to not allow the spring to compress more than 20%, nor extend more than 10% of its overall length. For example, a 46″ long Wrangler spring should only be compressed 9.2″ and extended 4.6″ from its static, unloaded position.
So what’s it gonna be?
We originally started out with the most popular setup: A 3-link system, using straight lower control arms and an upper “wishbone”, but after a fair amount of testing we discarded the wishbone part for a few reasons:
1. We couldn’t find a rod end or spherical joint that would give us the required 70 degrees of motion we needed at the apex of the wishbone. Most folks simply don’t use a jam nut on the wishbone’s rod end, leaving it loose to rotate on the threads. Which will give you 360 degrees of rotation, but is a HUGE no-no and will only result in eventual failure of the rod end. Plus, this application means you are using the rod end in a plane it was not designed for.
2. Using a tightly-mounted bushing of rubber or poly in its place produced bind when the axle was articulated, or would likely introduce axle hop if run loose enough to minimize the bind. The best answer we could come up with was to use a custom, two-axis joint, not unlike a U-joint.
3. The points where the wishbone mounts to the frame cannot be spread very far apart, or they start to interfere with the diff, driveshaft, and exhaust routing. Which left the mounting points about 11″ on center and meant we needed to add a bulky crossmember to mount them to. Knowing that we are planning four wheel steering, we felt it wouldn’t give us the side-to-side axle locating stability we would need.
Four-link “W”-type rear it is:
After about a month’s worth of intense design, test, re-design, re-test, re-re-design, etc., we binned the wishbone and put two separate upper links in its place. With properly-bushed, standard Aurora 3/4″ rod ends on each end of each control arm able to go to 35 degrees, that now meant our range of motion was now in the 70 degree neighborhood… right on target.
Amazing when the light bulb goes on, ain’t it?
Better yet, now all the rod ends were working in their proper planes, we were able to mount the upper links to the inside of the frame rails which spread them apart for significantly better stability, we could can the bulky crossmember, and we had zero, zilch, nein, nada bind OR play anywhere in the system.
We still had to do some R&D to get the rod lengths right to meet our objectives, which seemed to take FOREVER. But in the end, we got the geometry where it needed to be, quite in spite of ourselves.
Four-link “W”-type front too:
The rear setup gave us so many challenges, that we pretty much dreaded doing the front. Our ace in the hole here was the fact that the rotary sump is at the front of the motor, not the back like a Zook engine, which gave us gobs (technical term for “oodles”) of room to mount the links up on top of the axle.
We briefly discussed using a four-link panhard rod setup up front, but threw the idea out as the track bar would eventually crowd our hydraulic steering components, and the “W”-type setup gave us better side-to-side axle movement when moving through its articulating motion. It’s tough to explain, just trust me on this one.
After all our fretting about, turns out the front pretty much fell into place. We made some minor adjustments to the control arm mounting points to accommodate the driveshaft while maintaining proper geometry, and we were done.
Viola! ….Oh yeah, springs….. Next Installment…..