Torque Arm vs 4-Link

[Vimes]

nope that's what you said... what I said is



rear suspension has 3 basic functions 1) weight control (hold the car up) 2) Lateral location 3) axel location

and to answer that now asked question.. they are all a compromise .......

In that case, I'm glad we agree, and that you got a laugh from it to boot. The 4-link compromise is it gives up handling performance in favor of straight-line performance. A torque arm gives up excelling in any one area to be a decent street suspension that is competent in multiple areas, which is why I still feel the OP would be better off with one. He's not drag racing, and he wants something comfortable for long drives that will keep his wife happy.

[dontlifttoshift]

4) drives the car forward. That one is pretty important.

Vimes, have you driven on any or all of these rear suspensions? Not looking to validate or invalidate your opinion, just curious about your background and the foundation that you are forming your opinions on.

[dhutton]

I’m no expert but don’t we have to differentiate between parallel four link, converging four link and canted/triangulated four link suspensions in this discussion?

Thanks,
Don

[Vimes]

4) drives the car forward. That one is pretty important.

Vimes, have you driven on any or all of these rear suspensions? Not looking to validate or invalidate your opinion, just curious about your background and the foundation that you are forming your opinions on.

No worries. I currently own two vehicles with 4-links, one is a 2015 Tahoe and the other is a 1970 Chevelle. I've owned about 40 different cars over the last 35 years with just about every suspension commonly used in the US. Oddly enough though, I've never owned a torque arm car. However, that's about all my brother has owned since the 4th gen Camaro has come out, so I've driven one often enough to know how they handle.

Disclaimer, I've never competed in any organized events before. I don't doubt that I could do well, but I'm not willing to invest the time or money into it.

[Sleeper68]

I’m no expert but don’t we have to differentiate between parallel four link, converging four link and canted/triangulated four link suspensions in this discussion?

Thanks,
Don

Absolutely. Lumping all under "4 link" is really a gross generalization

[Mark@lateral-dynamics]

So, we decided to throw an empty housing under the shop car to show the amount of available suspension travel before bind. We have, from the very beginning emphasized "zero bind" with our setup so here are a couple pic's to show what we mean by that. This housing is empty so we didn't have to deal with the admittedly very heavy center section, axles, and brakes. The suspension articulates like butter. Silk. Gorilla snot, or any other super slippery medium you choose. Driver's side is 6" off of the deck, while the passenger's side is 24 inches. Inches, not millimeters. This is what "we" call zero bind.

Who cares? We do. And here's why. Performance vehicles rely upon precise tuning in order to operate at their very peak. With (for all intents) zero friction, the setup relies completely upon the springs, shocks, and sway bar to manage both the steady state and dynamic inputs. OEM type rubber joints, for instance, have an increasing "stiffness" in roll/bump by their very nature. At some point, they go full stiff, and that typically results in an abrupt reaction. Ever drive a fox body Mustang in anger? Once the bushings compress to their limit, the setup effectively locks up, and you get snap oversteer. I have personally had several of those cars and LOVE them, but that's no bueno. Especially in the wet.

I challenge anyone that has a five link type setup (i.e. parallel four link with PHB or Watt's) to do this same experiment and show the same outcome.

Yes, we are very passionate about our approach, but again, it's not marketing hype - it's based on physics and engineering. That's the same thing that race teams do at all levels. And it's the same thing the OEM's do too, all with exceptional success. It's just the target requirements that are different.

For the record, this is with all Heim/rod ends. I guarantee the same results with the same rubber (NHV isolating) bushings we have chosen to go with, I drive on them every single day.

Take care, be well.

[Mark@lateral-dynamics]

The single shear lower mounts can be problematic when the shocks bottom.

Please calculate the shear strength of a 1/2", 150 kPSI Grade 8 bolt, and explain with data, how this can be problematic. And I truly mean this respectfully.

[CSG]

I agree on the snap oversteer. My first experience with it was in a fox as well. I was tasked with driving a car I had zero experience in with no practice lap. The particular car handled very well until it didn't. When the rear went full stiff it did exactly what you would expect, luckily I had enough track to chase it. I have not had that same result though with a "proper" designed 4 link & bar. As stated above we can't lump all 4-links together. Since you are/were into foxes as well have you driven one with the original (not second generation) Steeda 5 link? For those unaware this is a redesigned 4 link with panhard bar for fox mustangs. It was a bolt-on deal with no floor cutting or anything so it was still a compromise setup but it worked really well.

[stab6902]

Please calculate the shear strength of a 1/2", 150 kPSI Grade 8 bolt, and explain with data, how this can be problematic. And I truly mean this respectfully.

I was curious, so I consulted google and did some quick and dirty math.

According to the internet, potholes can cause 30g+ vertical acceleration: "Pothole events that cause jounce bumper impacts can impart over 30 g's vertical acceleration (measured on a control arm near the ball joint). Those same events will produce over 3000 lbs longitudinally (at the tire patch)." Source: https://www.eng-tips.com/viewthread.cfm?qid=149427

So if you hit a gnarly pothole that resulted in metal to metal contact on the coilover assembly, (i.e. bottoming the shock without a jounce bumper or coil bind), that 30g's would be transmitted through the shock bolts. We can't really rule out metal to metal contact since it's up to the customer to pick suitable springs, shocks, and jounce bumpers, and I don't trust the average hot rodder to get that right 100% of the time.

Assuming a rear corner weight of 800 lbs, 800lbs*30g's=24,000 lbs shear force on that shock bolt. The 800 lb corner weight assumption is actually on the low side, especially if you want to consider load transfer.

According to Fastenal, 1/2"-20 grade 8 bolts have a shear strength rating of 13,376 lbs. Source: https://www.fastenal.com/en/84/load-calculator

Since 24,000>13,376 lbs, it appears Donny's concern (probably rooted in experience) is warranted for a street car.

[Mark@lateral-dynamics]

Since 24,000>13,376 lbs, it appears Donny's concern (probably rooted in experience) is warranted for a street car.

Except there are TWO bolts holding the car up, on each side. Also, the spring and damper consume the overwhelming majority of that impact force (even if 30 G's is stupid conservative), BEFORE the shock goes into bind, hits the bump stops, etc. The tire also compresses. And so on.

But appreciate the effort! At the same time, this is a good example of (even though well intentioned) misinformation that folks need to be wary of.

Since you are/were into foxes as well have you driven one with the original (not second generation) Steeda 5 link? For those unaware this is a redesigned 4 link with panhard bar for fox mustangs. It was a bolt-on deal with no floor cutting or anything so it was still a compromise setup but it worked really well.

I had a good amount of experience with those cars, and have a small Army of buddies who raced them extensively in NASA American Iron. I haven't driven the Steeda stuff, but did install a PHB to otherwise stock rear on my '89, it was a lesser brand but I didn't know any better. It tightened up the car a bit, but didn't really make it "better," until it broke. The battle of roll center location was won by the factory convergent 4 link. When I returned it, they weren't surprised at all, pretty much all of them were breaking. For the life of me I can't recall who they were, located in Orange County, CA.

[stab6902]

Except there are TWO bolts holding the car up, on each side.

I considered that in my calculation - I assumed a (low) corner weight of 800 lbs. If you mean there's one bolt on each side of the coilover, they're in series, and the top one is in double shear, so the bottom one would "feel" the whole load.

Also, the spring and damper consume the overwhelming majority of that impact force

My point was that since you leave it to the customer to choose the proper shock/spring/bump stop combination, you have to consider the risk that they get it wrong. There's plenty of posts on here where people have issues because they're running shocks with too little travel, coil bind, no bump stop at all, etc. People like running their cars low and not everyone is a suspension expert.

(even if 30 G's is stupid conservative)

What have you measured? I don't have any test data and had to resort to the internet. Obviously the pothole situation is much more of a concern for street cars, so typical racecar numbers don't really apply.

BEFORE the shock goes into bind, hits the bump stops, etc.

Agreed, as long as the proper spring/shock/bump stop is chosen, there will be no issues at all.

The tire also compresses.

From what I gather, the 30+g was measured on a real vehicle hitting a real pothole (so the tire was accounted for).

But appreciate the effort! At the same time, this is a good example of (even though well intentioned) misinformation that folks need to be wary of.

I think my point is still valid - if someone specs the wrong coilover and it bottoms out hard, that lower bolt will break. I've heard of it happening on much lighter weight street rods. Really, if there's going to be a "fuse," that's one of the better spots to have one. I still think your design looks fine - you just need to be careful to coach your customers to buy the right coilovers or you might have warranty issues.

[Mark@lateral-dynamics]

I follow your logic, but fortunately, (and respectfully) some of it isn't correct.

If you mean there's one bolt on each side of the coilover, they're in series, and the top one is in double shear, so the bottom one would "feel" the whole load.

This is incorrect. The forces are indiscriminate, both sides of the shock see exactly the same force (neglecting the mass of the shock, which is insignificant when considering the magnitude of the vertical force). IF, and a very big IF, the forces are high enough to shear off one of the bolts, you are correct that one in single shear will go first as both are the same bolts.

Hitting a curb/pothole straight on produces a force vector, your assumption is that ALL of the impact is going only vertical, but there is (just about as large) a longitudinal force trying to push the wheel back as well. As such, the vertical force is approximately only half of the total impact force. "Half" of the force is reacted through the combination of LCA's and to far lesser extent, the UCA. I say approximately because it depends upon how deep the pothole is, the deeper it is, the higher it will contact the rim, and less vertical load relative to the longitudinal.

Also, your assumption that the lower BOLT will see 30 G's is way over-estimated. If you see that type of a load on anything on a car, you no longer have a car, and are also probably in seriously bad condition, or dead. The 30 G factor is an impact load imparted on the TIRE. It travels through the tire, into the rim (which in this case is now destroyed), into the axle, through the wheel bearing, lifting half a 250 lb axle assembly, which directs the load into the spring/shock assembly. 30 G's is the type of impact a 200 MPH race car sees when hitting the wall. They usually don't survive that.

By the way, we are not the only ones who single shear a shock mount, I don't think too many people would bash on ridetech, for instance?

We don't let the shock selection completely up to the customer. We leave the SUPPLIER of the shock up to them. The dimensions for the actual shock depend upon the static ride height, and we include a set of guidelines to consider when selecting the proper shock and spring length. The last thing we would ever, ever, ever do is to leave a customer, who trusts us enough to buy our system, hanging with a critical decision such as this.

Hope that helps.

[stab6902]

The forces are indiscriminate, both sides of the shock see exactly the same force (neglecting the mass of the shock, which is insignificant when considering the magnitude of the vertical force). IF, and a very big IF, the forces are high enough to shear off one of the bolts, you are correct that one in single shear will go first as both are the same bolts.

We're in total agreement. In my example, each shock bolt would see the 24,000 lb load, just like each link in a chain sees the same load (neglecting the mass of the shock or links respectively). The load doesn't get "split up" if you add bolts or links in series.

Hitting a curb/pothole straight on produces a force vector, your assumption is that ALL of the impact is going only vertical, but there is (just about as large) a longitudinal force trying to push the wheel back as well. As such, the vertical force is approximately only half of the total impact force. "Half" of the force is reacted through the combination of LCA's and to far lesser extent, the UCA. I say approximately because it depends upon how deep the pothole is, the deeper it is, the higher it will contact the rim, and less vertical load relative to the longitudinal.

The guy on the other forum explicitly stated the z-component of the acceleration (vertical only) was measured at 30 g.

Also, your assumption that the lower BOLT will see 30 G's is way over-estimated. If you see that type of a load on anything on a car, you no longer have a car, and are also probably in seriously bad condition, or dead. The 30 G factor is an impact load imparted on the TIRE. It travels through the tire, into the rim (which in this case is now destroyed), into the axle, through the wheel bearing, lifting half a 250 lb axle assembly, which directs the load into the spring/shock assembly. 30 G's is the type of impact a 200 MPH race car sees when hitting the wall. They usually don't survive that.

I figured there would be some pushback on the 30 g figure. Honestly my gut reaction was the same, but not having any data of my own, I'm an innocent until proven guilty kind of guy. He said the 30 g's was measured near the ball joint (1:1 ish motion ratio). My understanding from reading the thread was that this was measured on a real car hitting a real pothole, and I have to assume that car had tires. If you read the spec sheet of the wheel force transducers he references, the "passenger car" sized model measures Fz up to 11,000+ lb, and that's at the wheel, not the tire.

According to wikipedia, Indy car crashes are in the 200 g range, and somebody survived a 214 g crash. Source: https://en.wikipedia.org/wiki/G-force

By the way, we are not the only ones who single shear a shock mount, I don't think too many people would bash on ridetech, for instance?

Nobody is bashing on anybody. I know your product means a lot to you and the last thing I want to do is discourage innovation in our hobby. ridetech didn't ask anyone on here to run any calculations, and they use a 5/8" stud on their lower single shear shock mount. 5/8" bolts have over 60% more shear strength than 1/2" bolts. BMR also uses 5/8" bolts on their lower coilover mount. Maybe they're onto something.

We don't let the shock selection completely up to the customer. We leave the SUPPLIER of the shock up to them. The dimensions for the actual shock depend upon the static ride height, and we include a set of guidelines to consider when selecting the proper shock and spring length. The last thing we would ever, ever, ever do is to leave a customer, who trusts us enough to buy our system, hanging with a critical decision such as this.

Glad to hear it! Judging by your responsiveness on this forum, I'm sure your customer service is great.

[Rod]

4) drives the car forward. That one is pretty important.

ok ok ok yes forward motion... dang all picky

40 different cars over the last 35 years

Car dealer?

[Vimes]

Car dealer?

No, I've just been around for a long time.

[raustinss]

Im glad that Ryan brought up the topic of the indy car crash which ,is funny because immediatelythought about that crash i believe Emerson (not going to butcher rhe last name lol ) was the driver or had one similar. Regardless as to what the forces are or arent on a bolt, its the amount of time the bolt is stressed with those forces that determines if the bolt breaks or not . I would say clearly there's a mistake in calculation somewhere or we would be hearing about suspension failure due to broken bolts all the time yes ???

[stab6902]

To be clear, my back of the napkin math represented the extreme worst case. I'm guessing the engineer who quoted the 30 g figure worked at an OEM, and the test was probably done on a proving ground. I've ridden along on a few durability loops (as a passenger during training), and it's amazing how abusive they are, both to the cars and the driver. We're talking impacts that would put big dents in the rims of the average pro-touring car.

In addition to an extremely hard hit, the coilovers would have to be in mechanical bind (coil bind or bottomed out with no bumpstop) to see that kind of load. If you choose your coilovers correctly and are running some kind of bumpstop, you'll never have that problem. Like I said, extreme worst case.

A lot of people can and do get away with 1/2" coilover bolts in single shear, but it wouldn't be my first preference. You don't hear about it often, but they do fail in the real world. Some discussion on it here: https://www.jalopyjournal.com/forum/...estion.991025/

If I were building a coilover car, I'd go with 5/8" bolts if I had to do single shear, which is what most aftermarket companies do anyway. The shear rating of a 5/8" grade 8 bolt is very close to the "worst case" I calculated. The theory (crude as it is) seems to line up with real world experience and best practices.

[dontlifttoshift]

Please calculate the shear strength of a 1/2", 150 kPSI Grade 8 bolt, and explain with data, how this can be problematic. And I truly mean this respectfully.

I know what the math says and I really don't care what the math says.

In the past 20 years I have a handful of real world instances on street cars, not race cars, where the bolt in a single shear lower shock mount failed. When those bolts bent or broke they did not care that mathematically they weren't supposed to. Most of those were 5/8" in diameter as well in a street rod application. We can point to a number of factors for the failure but the fact remains that they failed.


My personal ridetech equipped 71 Camaro still has the factory bump stops on the rear frame rails above the axle.

[dhutton]

I know what the math says and I really don't care what the math says.

In the past 20 years I have a handful of real world instances on street cars, not race cars, where the bolt in a single shear lower shock mount failed. When those bolts bent or broke they did not care that mathematically they weren't supposed to. Most of those were 5/8" in diameter as well in a street rod application. We can point to a number of factors for the failure but the fact remains that they failed.


My personal ridetech equipped 71 Camaro still has the factory bump stops on the rear frame rails above the axle.

So what is the design solution for these failures? It seems they are all done this way.

Don

[stab6902]

So what is the design solution for these failures? It seems they are all done this way.

Don

I totally believe what Donny says regarding even 5/8" bolts failing. We didn't even talk about fatigue yet... lots of smaller hits can do the same damage over time.

The best solution is to run the factory style bump stops between the axle and frame rail like Donny said. Cycle the suspension without springs, and make sure the bump stop is really what limits travel.