I understand what you're saying about it being more rigid, the guys I've talked to that run that setup say the same thing, even though they've run it for yrs with no problems. but I am re-doing my upper links, just so you know.
Sorry Kerry, but that is just not true under HARD acceleration. If it were, we would have no anti-squat lift reaction through the upper links, which is the KEY to all drag racing 4 link designs... The upper links act as the anchor or "anti-rotation device", restraining the pinion tendency to climb the ring gear. There is exactly the same force present there as the torque applied to the wheels.
Anti Squat is not Torque reaction, Anti Squat is the theoretical path of acceleration against the theoretical path of resistance.[ this is similar to how a pole vault creates lift ] Anti-dive is the same but during deceleration . Torque reaction is at the point where the pinion tries to climb the crown wheel.[ brake torque is similar but opposite ] On a Grand prix car with a transaxle and IRS, they can dial in anti-squat by tilting the A-arms so the path of suspension movement travels similar to a 4 link, Yet the torque reaction from the Crown & Pinion is totally independent of the suspension movement. If a Grand prix car snaps an A-Arm the wheels can violently toe-in [ because of the accelerating forces of the wheel against the resistance of the chassis ] Ive seen rear-end housings on drag cars bend also ,because of the accelerating forces from the wheels. These accelerating forces are transferred forward through outer suspension links to the chassis [ the nearest mounting point to the point of traction ] The accelerating forces start from the footprint of the tyre [torque] and convert to thrust [acceleration] at the centreline of the axle. The linkage that is furthest from the centreline of the axle takes the highest percentage of axle torque reaction. Lotus placed the outer links close to the centreline of the axle for 2 reasons. To keep the linkages parallel while using the factory front hangers, and to minimise torque reaction so the linkage takes mainly compression forces during acceleration and tension forces during braking. The majority torque reaction is handled by the lower A-Frame on a Lotus Cortina [ also compression during acceleration and tension during braking ] On a car like a 3rd Gen Camaro with outer links and a torque arm, you can change the Anti-squat by raising the outer linkages at the front [ the torque reaction is totally controlled by an independent torque arm ] If you raised the outer links too far [ at the front ] the accelerating forces of the rear end would try to Drive-under the resistance of the chassis creating lift at the rear end of the car. Yet the torque arm tries to lift the front of the car through torque reaction [ causing the rear to squat due to weight transfer ] The forces of squat due to acceleration are usually greater than torque reaction [ you can see this on dirt cars a lot with short links pointing too high at the front ] I hope this helps [ It took me a while to get my head around this also ]
Alternatively you could see the LH upper and the diagonal as an A-arm pivoting on the frame, defining an instant centre at the pivot. In other words we have two mutually-inverted asymmetrical A-arms with the diagonal in common, and a pair of instant centres at the respective ends of the diagonal. That suggests a roll centre at the mid-point of the diagonal - which is what one would expect if one looks at it the other way, i.e. as a 4-link plus short, diagonal Panhard bar. Either way it would suffer from the usual 4-link bind issues in simultaneous roll and bump.
Didn't Magoo use torque arms to the center of the car about the back of the transwmission and used a single Heim end as the pivot? Similar to ladder bars with single pivot? That would eleminate, or greatly reduce bind and end up similar to Ford's original design. Jack
I understand your point, and you are correct, especially when describing an IRS system. On this continent (and referring to drag racing in particular) "anti-squat" IS a torque reaction created by the (attempted) reverse rotation of a SOLID axle housing as the ring gear transfers torque to the axle shafts. If you agree that the housing attempts to rotate to the rear under acceleration, then imagine a winch cable wrapped around the axle tube and connected forward to the chassis. As the housing rotates, the winch cable will "lift" the chassis up relative to the housing. This counteracts the natural tendency of the rear to squat, hence "anti-squat". If we replace the "winch cable" with a solid upper link, any down angle in the link will result in the same effect. Drag cars typically run with the lower or "thrust" links level (under load), to eliminate the additional thrust anti-squat effect you described.
Bushings....yeah. We've all seen bushings on front suspensions of newer cars that are mounted with the bolts vertical! Everything works fine and quiet...because side clearance for the bushing flex is considered in the design. This kind of flex clearance is especially important to consider in a 4 link with all the bushings in line with the rearend. (OP style) As the car leans in a turn with that mounting configuration the bushing must flex as opposed to just rotate around the mounting bolt torsionally. Bushings take quite a beating in any position but are well up to the task. Its what they were designed for. All we need to do is select the proper style and give them the room to do their work without undue stress. With this flex clearance taken into account...there is no reason ALL the bushings couldn't be mounted with the bolts vertical...and any suspension movement allowed by the bushing flexibility alone. Weird and no real need for such mounting...but totally possible. This whole thread seems to suggest that the bushings are a weak link that must be pampered or something. I don't believe thats the case at all. Just let them do the job they were designed for and use the proper bushing for the required job!
Good point. Unfortunately, big flexy bushings like the OEM use aren't very popular in aftermarket or DIY suspension builds. When you go to something like the popular urethane material (which doesn't really work well in EITHER tri-4 link design...), the end links must be mounted in such a way that minimizes the required deflection. Again, the "right" answer to the OP question depends on the primary motion anticipated for the individual suspension system.
I still don't get how the rear end swings in an arc described by the torque arm and the outer links, at some other length, are able to follow this arc. Help!
Look at the leading-end of all of those torque links. There is a link that functions as shackle there. It is not a fixed point. The torque arm swings AND moves fore-aft during suspension cycling. The arc that the axle moves in is produced by a combination of the movements of all of the links.
That Winch cable theory ? It doesn’t quite happen that way. You need to realize that torque is what drives the vehicle forward but torque reaction is a parasitic loss, and yes there will always be torque reaction at the pinion causing it to want to climb the crown-wheel. Earlier I wrote this comment <TABLE style="COLOR: #000000" cellSpacing=0 cellPadding=6 width="100%" border=0><TBODY><TR><TD class=alt2 style="BORDER-TOP: 1px inset; BORDER-RIGHT: 1px inset; BORDER-BOTTOM: 1px inset; BORDER-LEFT: 1px inset">Originally Posted by Kerrynzl During acceleration both the upper and lower linkages are under compression , even though we think the outers are under tension . [Acceleration of the wheels against the resistance of mass] </TD></TR></TBODY></TABLE> This is correct , in a 4 link during acceleration BOTH the uppers and lowers are under compression [ the only time the uppers can be under tension is during braking or if the axle is fixed [so the car cannot move forward] see below if this helps understand The torque multiplication ,because of the rear-end ratio means there is a greater forward [or compression] load on the links than there is lift at the pinion. The car will ALWAYS try to move forward. If there is excessive torque reaction at the pinion this tries to rotate the rear-end upwards [ we already established this ]. This is “rotation” is transferred to the linkages by having a higher PERCENTAGE of compression on the lower links and a lower percentage of compression on the upper links. The difference in these loads is the amount of lift at the instant centre [ this can be calculated using simple trigonometry ] The instant centre is also the theoretical point of acceleration , if this is in front of the CG the rear will squat, behind the CG and the front will lift [ note: this is a basic explanation and not 100% accurate ] It does get a little more complex than that though because there is an anti-squat line and a CG height. Whenever mass is accelerated there is also a reaction called weight transfer. If the instant centre is low and the CG is high the weight transfer becomes an “overturning moment” [ in layman's terms “it tries to wheelstand” ] which is considered a parasitic loss in drag racing.
I love these discussions. I learn stuff I never knew. I also learn that I know more than some people, and a lot less than some other people. And no, I have nothing constructive to add to the discussion other than I used heim joints on my triangulated four bar.
Linkage geometry can enhance traction, but the effect is momentary. It is a common misconception that with the right suspension and/or the right geometry the axle's torque reaction(pinion climbing the ring gear) can lift the front of the car. Just as you can't lift yourself by pulling on your own shoe laces, that is essentially incorrect. While angling the links to improve traction is fine on a drag strip, it causes other problems on a "real" car. The resulting roll steer will cause the rear of the car to steer over bumps, steer to the outside of a turn when cornering, and accentuate the effect of crosswinds. All of those things make the car less stable and less pleasant to drive.
The general consensus I'm getting from those that are opposed to the setup I have, is that somehow by driving over a bump, or turning a curve, the bushings will be in such a great bind that everything will break, and fail. Like I said before, I know several people running the same setup on bagged vehicles, even some with side to side, that haven't had a single issue. Both setups work, but my point is that one binds more than the other. So, I went to the shop this weekend, and set my car up on jack stands at ride hight, and opened the temporary valve so I could fully compress the bag, and I had plenty of side to side travel. I was able to completely bottom out one bag. and as you can see, there is no binding in the bushings. (you'll have to excuse the bolts being too short, they were just some I had laying around the shop)
Look at the left upper rear mount, you can see it is no longer parallel to the axle tube (assuming it normally is ). So your bushings seem to be absorbing the flex nicely.
In your first picture the upper left bushing is not only deflecting, but so is the mount and the bolt head is not seated to the mount either.
You're right, the tab is flexing slightly, but keep in mind that this is an extreme example, I have one bag completely bottomed out. So, with the bags filled, and with eight valves so I don't have air transfer, even in a hard corner it will never see this much side to side travel. I just wanted to push it as far as it would go, to show that nothing is going to break. I want to get past this notion that somehow just by going over a slight bump, my setup is going to break, and fail. People run both setups, and they both work, thats not what I'm arguing. my point in this thread, is that one setup has more flexing than the other, and I happen to think that having the pivot points parallel to the rearend has less flexing.
Absolutely Correct! Almost all of the lift in a drag car is due to weight transfer over the instant centre. Drag racers harness this parasitic loss by using 90/10 front shocks and soft springs to aid traction. They also try to tune the suspension to give the tyres a good Hit [ to momentary stage where the sidewalls absorb torque ] Race car suspension on the street! What can I say except it isnt just drag suspension that is unpleasant. When you drive a Road Race car slowly [down the pit lane] these things are unbearable, They dart left and right because of camber thrust and crossply slicks, locked rears etc BUT when you start driving them aggressively and leaning them on the sidewalls the normal defence mechanisms programmed into our human behaviour suddenly turn into an Insane Satisfaction [ racers will understand this psyche ] On a road race track there is no such thing as a straight [ just longer corners ]
apparently he has not read anything people have posted on this thread, just because it "works" doesn't mean its correct, i have stayed off this thread as long as i could. its just wrong man, buck up and admit your mistake and move on. we've all done things that "work" and most of us have found out that even if it works for a little while it is going to break eventually. jacking one side up slowly on a shop is a lot different than a panic lane change or one of a thousand things that can happen on the street in an instant. weight transfer, grip off camber turns put so many different stresses on parts that you just can't test for with a jack. they really need to close this thread!
You're right, just because it works, doesn't mean its right. the same thing was said in the e-mail response i posted earlier in the thread. "The way guys build triangulated 4 links 90% of the time is incorrect. Excuse my crude examples but this is how most setups look. Now imagine how these bars are going to travel up and down. They are going to want to close in on each other, due to the geometry of the pivot points. As the rear end travels up and down from theoretical ride height with the pivot points set up the way they are here, the triangle will close up. Putting the bushings in a bind." "The way pivot points should be setup for ZERO binding, is as follows. All the pivot points in the 4 link should be parallel with each other. An even better way of saying it would be that the points need to all run on the same axis in order for nothing to bind. " mindliss metalfab (thats how they spell it) has also been building the same setup for years. whether you agree with it or not, both setups work. you have completely missed the point of this thread. and if you don't like it, stay the fuck off of it. it's that simple. this is a discussion, some people here have agreed with me, and some haven't, you getting upset, and thinking that the thread should be closed over something thats not even the point of the thread, is ignorant.
Mr. "mindliss metalfab" may be a 'pro', but he's not an 'expert'. We could have a discussion on whether the earth is round or flat, and hear some widely differing opinions. Some basic facts would still apply, regardless of opinions. Same applies here. As others have stated, the laws of physics and engineering apply regardless of opinion. You cop a big attitude when someone disagrees with you. You ignore the facts at your own peril. Good luck.
OK, the mounting tab flexes. How many times will it flex before it fails? In your picture it sure looks like the bushing has a tremendous amount of flex. There's so much stress that the mount is flexed. I understand your reluctance to not take backyard builders' word for it, but post #82 contained an opinion from an engineer from a car company. A key excerpt from that post...."The version with the link bushings parallel to each other on the frame and axle exhibited zero bind in a straight bump situation with no articulation of the suspension. As we started to articulate the suspension, the bind increased at an exponential rate putting unbelievable amounts of force on the bushings initially, then the bolts and eventually the mounts. Since bumps in the road are rarely equal on both wheels, this is not a good design and will create large amounts of rear roll stiffness while inducing undue stress. We believe the mounts will fail in this design if used in real world driving conditions as they are most likely the weakest link." Some very knowledgeable people have explained why the parallel mounts are not good for a street car. Engineers from a car company agree. If that weren't enough your pictures confirm what they said. You seem unconvinced. So be it. I hope it works for you. There's some good info here for those that reference this thread in the future.
Like I have said before, I've seen both setups work, but if anyone has read my response in post #124, you'd know that I am re-doing the upper links, so everyone can calm down.
So by your first diagram above, to increase anti-squat, the top and bottom links should be angled more uphill to the frame, correct? Then, PLEASE explain to me why every, EVERY 4 link equipped drag car has the upper links angled downhill to the frame... I can provide thousands of photos if you like. Also, if the reverse rotation of the housing is not a factor, how does a torque arm work? What magical force is lifting this car almost 6 inches above ride height against the force of those wings? Oh, and the C/G height of that car is about 11", and the instant center is above that, so it has NO leverage... I patiently await your answers.
If you go back and read everything I have written, I have never stated torque reaction is not a factor. You will always get torque reaction as a parasitic loss. What I originally said was the upper linkages are under compression [ NOT TENSION ] during acceleration, even with torque reaction added. The diagram I posted was merely to demonstrated compression or thrust as they wrote [ I stole this off google images ] You dont need to provide 1000s of photos of pro-stock suspension to support your argument when the answer is actually already been written. Most drag racing 4 links have the lower links horizontal [ they try to get them as low as practical ] But for the sake of your question we will assume the 4 links are pointing down. This will lower the instant centre.[ IC ] The instant centre is the theoretical point of acceleration. It is also the pivot point that weight transfer will rotate around. If the CG is higher than the IC the front will lift and the rear will squat during acceleration. By lowering the IC helps weight transfer via overturning moment onto the rear suspension. Note :you will always get weight transfer during acceleration even with no suspension [ jump in a go-kart and feel the G-forces ], the idea of setting up the suspension with a low IC is to help transfer weight from the front onto the rear wheels. The formula for weight transfer is [ weight x CGH x Gs divided by wheelbase ] ,The formula for weight transfer via overturning moment is [ weight x (CGH ICH) x Gs divided by wheelbase ] CGH = Centre of Gravity Height, ICH = Instant Centre Height, Gs = Acceleration. On a sprint-car the IC is at the Bell / torque tube which is non adjustable, but they have a 60%+ rear weight bias, sprint car suspension is usually dialed in with spring stiffness A short wheelbase with a massive rear weight bias and soft springs, it is easy to see how they can carry the front during acceleration. But dont let looks deceive you, MOST of the time a sprint-car, midget or super-modified only carry the left front wheel. On oval cars they usually stagger the rear wheels to aid cornering [ smaller diameter inside ], to get this to work correctly there need to be 50/50 weight on each rear wheel at the point of maximum lateral acceleration. So they try to load more static weight onto the inside rear. One method is to offset more left side weight bias, but they also pre-load the inside rear wheel [ weight jacking or wedging ] When the inside rear is Jacked, the majority of the static weight is supported by the inside rear and outside front [ hence the term wedge ] , the left front only holds the car off the ground in the pits. When these cars accelerate the weight gets transferred diagonally from the LF to the RR usually carrying the LF off the ground.
The Jalopy Journal
