Anybody know any facts about reversed four links

I don't think his brother is building his '53 hardtop so he can load his girls up and do laps with the frame an inch off the ground. What works on the track does not necessarily translate to the street, especially in suspension design. I'd say he got his answer about backassward 4 links and a very good suggestion on how to locate the axle too. Mark me down as another vote for Chevy truck trailing arms.

 

Ah good, the semantics cops have finally arrived. This signals the nearing end of civil discourse. It's very easy to pick apart the words used to try and make someone look "foolish", but it doesn't prove anything, other than you may be an asshole.

Around where I race, it was called a swingarm system, but the proper term is Z-link, and you will note that I used both in my previous post. The Z-Link system uses one lower link running forward and one upper link running rearward attached to a birdcage with either a bearing in it or a plastic bushing. The spring rides on the lower link about 3/4's of the way along it's length (biased to the rear), while the shock (or damper, since the semantics cops are here now) attaches to the birdcage to maintain a good motion ratio.

BTW, I've owned and raced cars with the Z-Link rear suspension, and they were damned fast. No "backyard engineering" (as though such a thing were somehow shameful or bad), straight from a fairly big name chassis manufacturer.

You quote the litany of anti-dive, instant center, and wheel hop like it's some sort of mantra, but do you even know what those terms mean, or how you go about calculating them?

I just get really tired of hearing the same "If it worked, more people would do it." or "It's obviously dangerous." answer, 'cause it looks (to me at least) like most all the folks saying that really have no idea what they're talking about.

I am by no means a suspension god, but I am an engineer, so if you're going to say something like that (especially if your aim is to "prove me wrong"), you'd better bring some tangible proof with you.

If I'm full of shit and no rearward facing linkages of any kind could ever be made to work correctly, prove it. Do the math, make a drawing, and post it up. If your math is right and your reasoning sound, I'll apologize and concede the point.

But I'm guessing that's not gonna be the outcome.

And I still agree, that truck arms are the best and simplest solution to the OP's conundrum.

 

Hey, I'm an engineer, too! Saying you're a engineer doesn't really hold much water when you're sitting behind a computer and posting on an internet message board. Do civil engineers take dynamics courses? Did I mention I can be a bit of a jerk?

Honestly, I don't care to run any numbers as there are dozens of books on the subject with examples that the OP can take a look at if he wishes; SAE's text on Race Car Vehicle Dynamics might be one worth checking out at the local college library. I'd much rather pound on my keyboard and be an ass on the HAMB during my lunch hour than whip out my calculator and do some free-body diagrams with imaginary linkage geometries.

Instant center: On a linkage with 4 reversed bars, the IC is located behind the car. Not an ideal position, as it should be located in front of the car's center of gravity. A linkage with 2 forward and 2 reversed doesn't have an "instant center", as the suspension doesn't rotate around a single theoretical location. This magic little point in space is an important one when it comes to the suspension's dynamics.

Anti-squat (so I used the wrong term earlier): This little deal has a role in planting the rear wheels under power, and is determined by the location of the IC in relation to the COG and the ground. A linkage with forward bars will have this imaginary intersection point above the ground and below the COG, at least ideally. A linkage with reversed bars will put this intersection point below the ground. Since this does deal with weight transfer of the car, it does interact with the anti-dive properties of the front suspension.

Pinion Angle: On a forward linkage the pinion will basically rotate towards the output of the transmission. On a reversed it will do the opposite, and on a "Z" it will rotate to an extreme degree. We all know what the effects of an improper angle feel like and the problems it can cause on driveline seals, bearings, and components.

Will reversed linkages "work"? Yeah. Will they do everything a suspension is supposed to do in a correct fashion? Physics, geometry, and the utter lack of use of reversed linkages in the production car and professional race-car worlds would lead one to say, "no." Unless, of course, you've got the diagrams and math to post up and prove that it does indeed work like you say. You are, after all, the one with experience racing on such a suspension design; surely you proved out all the geometry for them.

I will post up a diagram though, since I get paid to piddle around with CAD software all day and like to take a break every once in a while.

 

Reply in bold red.

Diagram to follow.

 

OK, here's a quick-n-dirty diagram to hopefully help illustrate what I've been trying to say for several posts now.



Here you see that in both instances we have a level lower link, and the top link at 6 degs down toward the front (which may or may not be a little too much for a street car, depends on the CG height, but using the larger rake makes the angles easier to see).

See where the instant center lands in both set-ups? Exactly the same place. Nothing else was changed except the mounting location of the frame side of all the links. Length and angle remained the same, as did their attachment points on the rear end housing.

Lets take a closer look at the two to see how pinion angle and wheelbase change between the two systems.



Ok, here's the forward facing system shown at ride height and 3" of bump travel. Note the change in pinion angle and the length of the wheelbase.



Now, here's the rear facing system, again shown at ride height and 3" of bump travel. Note the change in pinion angle and the length of the wheelbase.

The rearward facing system actually has less change in wheelbase than the forward facing system, so the slip yoke will move in and out of the tailshaft housing less than it would with the forward version. In fact, the difference is greater than an order of magnitude (although in this case, even the larger number is pretty small).

The rearward facing system also has slightly less change in pinion angle than the forward facing system as well.

I'm not seeing the doom and gloom here man.

 

Here is a link to the 4-bar calculator that I have used often. http://www.pirate4x4.com/forum/showthread.php?t=204893&highlight=link+calculator

there are a few quirks like the bump and droop on the main page do not update the travel page, so best to bump and droop from the travel page, other than that, you can see the changes in your pinion, anti-squat, etc. The anti-dive is not calculated, nore are the forces of braking or acceleration. However the forces due to breaking and acceleration need to be calculated. You will note that the rear is unloaded during acceleration due to rear facing links. Two good resources are Tune to Win and Engineer to Win by Carroll Smith.

 

The rear will unload only if you use rear facing links mounted solidly to the housing like truck arms. With the ends of the links mounted to heims or bushings, the only forces they can carry are tension and compression loads, so their geometry (angles relative to the chassis and one another) determines how the suspension is loaded under acceleration and deceleration.

And yes, I agree, Carroll and Steve Smith's books on suspension design are some of the best written on the subject.

 

draw the moment and force diagram and look at the forces

 

OK, I will:



See, with pinned connections, the links can only bear tension or compression loads, no bending.

So, the horizontal link contributes only to forward thrust, while the angled upper link contributes to both forward AND downward thrust.

This means that any time the wheels apply torque to the ground to move the vehicle forward, that upper link is applying some positive force downward to aid traction. No matter how small the force, the vertical component will always be down, due to the link angle.

Now, that same arrangement will produce a negative (or upward) thrust on the tires under braking. Fortunately, the brakes have a hard time applying as much torque as the motor does, so the lifting force will always be lower than the planting force. Even if it weren't though, the upward force would need to exceed the downward gravity wheel load from the mass of the car before it could go into wheel hop.

After all this arguing is said and done, I still think that truck arms are the best option for the OP's brother.

 

Well I think I will have to follow the herd here and agree that a reverse four link is a bad idea.
There are tons of reasons but I am not going to try to explain them.
Instead I am going to toss out another not so traditional option.

This kind of thing is taboo in some circles but it's hard to argue with the technology.

I am speaking of nothing more than IRS.

Go yank the IRS system out from under a late model T bird or jag and replace the springs with some bags.

A friend of mine did it with his F100 and it rides and handles very well compared to the stock ride.

Here are some pics



It does have a considerable amount of frame work but it does sit plenty low and is something that I would think could be done with the brother's car.

 

Nice fab work there, I dig that frame design and execution.

I think you're wrong about the four link business, for the myriad reasons I've already posted, but I'm not gonna beat the horse anymore.

If you think the idea of using a reverse four link caught some flak, wait 'till you see the comments on an IRS.



BTW, are you guys not running shocks on that truck at all, or are they just not mounted yet in those pictures?

 

Yeah I thought IRS would be a bad thing to bring up here but it is an option.
I would think it would be more of an option under something where you could not really see it.

And, how can people dog on IRS when we are talking about air bags here.
That was not exactly commonplace back during the hayday of hot rodding.

And to answer your question, Yes it did have shocks on it in the final product.

However I have seen some morons bring some shit into the shop some time ago and think it was perfectly ok to run without shocks.

I was amazed by their stupidity to say the least.

 

Yeah, I know. Personally, I don't understand why folks around here have coniption fits when certain kinds of suspenions are discussed, but they do it none the less. I was just sending a tongue-in-cheek warning to be on the lookout now that we've gone there.

That's why I asked, 'cause I have seen the same thing.

For some reason folks think that because it's not a regular spring, it's immune from the need to be damped. I dunno how they get there, but it happens all the time.

 

+1 on the truck arm suspension.

Here's an example of truck arm suspension that my friend did on his '51 Chevy, worked out real nice.

 

CoolHand, I really like your illustrations for instant center and pinion angle. You are on the ball, except for one point. The instant center will not be in the same place because the thrust angle is different. It's actually hard to predict how the four link in the rear will behave.
Here's why: The angle of the bars in a traditional four link control IC. The thrust of the axle is always trying to pass the front pivot of the lower bars. It can't because of the solid link between the two. By lowering the front of the lower bars, the thrust of the axle is encouraged to go over this pivot. This causes the car to squat, rotating it's mass close to the intersect point you outlined in your drawing. Lowering the front of the top links moves this point toward the rear, raising it moves it to the front.
In the case of the the rear four link, these forces are diminished as the axle is dragging the load rather than pushing it. The tendencies are similar, but consider this. On a traditional four link, the top links are pulling back on the car as a result of the rotational force in the axle housing. The lower links are pushing. Lift and squat is primarily a function of the lower link as it tries to climb under or over the front pivot of the lower link. This is heavily influenced by the top bar pulling on the car. In the rear four bar the lower is pulling the car with minimum tendency to lift or squat only being driven more by inertia than in the previous case. The top bar is now pushing on a rearward part of the chassis. Push a little and it is no consequence, push a lot and it will encourage squat.
I believe this setup is biased heavily toward suspension squat and would be hard to encourage lift. In a cruiser, no big deal. In a drag race, the harder you try to launch, the less traction it will offer.
Did I lose anyone? It makes my head spin.

 

I think you lost yourself.

You can't fight a free body diagram. When you run the numbers, the answer is the same, regardless of whether the bars go forward or rearward.

I've done all that I can to explain it without setting everyone down in a dynamics classroom for a couple of weeks.

It can be made to work, but not by just mirroring a standard four link. I've shown this graphically, and with calculations. If that ain't enough proof, then I'm out. I know enough to know that you can't fight superstition unless you build one and prove them all wrong, and even then most folks won't believe it.

Since I don't particularly care to build a backwards four link just to prove my point, so I'm gonna have to throw in the towel on this one.

IMO, the reverse four link is, although technically viable, "An answer to a question nobody asked."

There are easier (simpler) and less controversial ways to accomplish what the OP's brother wants to do (IE preserve the rear seating area of his auto while still being able to use airsprings), and being an engineer, I find elegance in simplicity.

 

So, you ever tuned a 4-link at the track?

 

The forward facing kind, yes, though on circle track cars, not drag cars.

Four links in the roundy round world usually run on birdcages with a fifth (and sometimes sixth) link to control axle wrap, but under certain track conditions, some geometries will have the left side birdcage locked down so it rotates with the axle tube.

I've never track tuned a drag racing four bar, but I'd imagine that the principals are the same, you just apply them to a different set of circumstances and geometry.

We use bar thrust in the circle track cars too, so it's not like I don't get what you're talking about. I just think you're mistaken about what impact it has on the viability of the design.

When you turn it around backwards, you still have one bar in tension, and one bar in compression, they're just different bars. You don't just assume that because the bottom link is now in tension that the system just won't function correctly anymore, you work your geometry around to take this into account. The problems don't solve themselves, but that doesn't mean they can't be solved.

Like I said, I'm not gonna argue this anymore, 'cause we're just going in circles now (I know, 'cause that is familiar territory for me ).

Suffice to say that I think it is entirely possible to build a reverse four link that will have the same anti-squat properties as a forward four link.

Obviously you and several others do not. It's a free country (at the moment), that's your call.

The only way to really settle this aside from CAD drawings and calculations (which you all seem to have already decided aren't good enough) is to build one and test it, but I'm not that concerned with being right.

 

You sure use an awful lot of words for being unconcerned.

I'm not sure either way. I just suspect there are forces in play that your fancy drawings won't account for. (I design all my suspensions in auto-cad, so I like your drawings)

The real world is different than any calculator. I think the biggest mark against such a design is that it's never been proven. This is a world where there are very few new ideas.

 

I didn't say I was unconcerned, I said I wasn't concerned enough to build a car just to prove my point.

 

Well, you'd have to totally separate the coilovers to get it to over center, a malfunction, I might add, which is possible on front suspensions equipped with a four bar as well.

No one seems to fear the rearward facing four bar when applied to the steering axle. I fail to see how a system which sees service on the front of a great many hot rods the world around is suddenly super dangerous when applied to the rear (in nearly exactly the same configuration).

You can mock me with your pipe-fitter inspired venom if you like, but my father, and I as well up until just a couple of years ago got dirty for a living just like you, so you might stow a bit your "you're educated and thusly totally unable to do anything with your hands" condescension.

I run into guys with your disposition all the time, and they all change their tune when we've worked together for a little while. I don't look down my nose at guys who actually work for a living because not only have I done it myself, but I realize that those guys usually know a lot more than I do about their little slice of expertise, and I'd be a fool to dismiss it.

In that same vein, you'd be a fool to entirely dismiss me just because I've got some "book lernin" under my belt.

BTW, I'm not nearly so well spoken in person. I speak with a pretty twangy southern accent that even folks in Alabama have told me is "an adorable accent", which is really saying something.

Honestly, I'm probably a bigger redneck than you are, I just find it time consuming and somewhat cumbersome spelling wise to try and reproduce the way I speak in text on the internet. Plus, sometimes folks wouldn't have any idea what I'm talking about, as I've found that local colloquialisms do not translate well to other parts of the country (or the world as sometimes is the case).

BTW, having the chassis get up on top of the bars a little bit under heavy braking is actually a desirable trait, as this will help keep the rear end loaded (which will help keep the auto pointed in the right direction throughout the panic stop).

All of that aside though, as I've said four or five times already, I agree, truck arms are what I would suggest as well.

 

Probably the best set up for your situation, would be the mid 90's Camaro/Firebird type with two lower link bars, like in a four link, with the torque arm running along the driveshaft. We have done a few of these in early coupes and they work great. You will need a panard rod with the bags or coil overs. This set up won't affect the floor or back seat, and they handle quite well.

 

I really mistook you then, and for that I apologize.

As you can tell, I get that one a lot, so it struck a nerve, but that's no excuse to be rude.

Also, there's a secret that a lot of engineers will never tell you. We're forced by law to stand behind our designs with more than just our reputations, so we kinda get this "All knowing and In Charge" persona put on us, but once you get to see behind the curtain, it's a different story altogether.

It was best summed up by a fellow on the Practical Machinist forum who once said:

That's pretty much it. We're professional guessers.

Now, we're good, 'cause if we make a bad guess, we go to jail, but we're still making our best guess based on the data at hand most of the time.

There you go. I've just outed the entire profession. I'm sure the gestapo will be around to collect me momentarily. It's been nice conversing with you fellows.

 

OH GOD, did smart dude nail it with that!!

As another engineer who 'left the trenches' then sat behind a desk for 20 years (pulp and paper industry) and then 'went into hiding' as a car nut, this has been a most enjoyable read. Thanks to all of you for the entertainment.
...and since I am not a big fan of bags I'll defer to you guys for design...


.