While NVH isn't really something I normally concern myself with, it has been brought to my attention that Multi-Link can help with ride harshness by isolating the loads through the bushings better, and this is probably a contributing factor to its selection given Mercedes' road-friendly approach to the car. Enjoy the video!
From kinematics point of view, there is no difference between the basic operational principles of double wishbone and multilink suspensions: Both of those suspensions are virtual swing arm suspensions, and both of them are formed from 5 links. With double-wishbones, the outboard ends of the 4 links are connected to each other (2 per wishbones) and the steering axis goes through the balljoints. With multilinks, the steering axis is virtual and can be located with little more freedom, for example to archieve centreplane steering, negative sides being the need for more balljoints, and the difficulty of doing any suspension adjustment. With both systems, the fifth link is there to control the steering of the hub. Double wishbones can be built from individual components, one example is the front suspension of 1971 Tyrrell F1 car. Individual links can more easily be built accurately, and if using any type of spherical bearings or ball joints, manufacturing errors have much smaller impact in the assembly or working of the mechanism. Multi-link can be little bit more easily be built in a way that when hitting bumps, the wheel is allowed to move backwards a little bit in relation to the car body, and this can increase the ride comfort (when hitting bump, reduced horizontal speed of the hub + vertical acceleration = the wheel hub does not need to accelerate when travelling up the bump). If built in this way, the multi-link behaves like a virtual semi-trailing arm suspension.
Im fabricating my race car to inboard antisway system & Im trying to calculate the spring rate i need for the torsional rod .in the rocker assembly & all the antisway bar charts i can find don't account for mount location on the control arm so i don't have a accurate starting point or a way to convert the antisway bar #s over to the requirement for a 6" rod Any help would be appreciated
Kyle, please please please make a video where you explain the different types of suspension setups and the ups/downs of each setup. Kindly explain using similar point of views so it would be easier to understand the differences between each setup!
I find multi-link suspension fascinating. the kinematics get a lot more complex but some of the kinematic results and practical implications are really interesting. Each arm being a pure tension-compression link allows them to be really lightweight if they are straight links as seen on this car (an example to the contrary would be the thrust arm on most modern BMWs). The two arms also create a virtual pivot point that extends beyond the physical reach of the arms, so you can have your virtual steering axis in the same place as, say, the physical brakes - not possible with a double wishbone system if your brakes are too big. if you want to limit scrub radius while also controlling kingpin, for instance, this would be advantageous. The steering axis also moves with steering input, this has a minor effect at lower steering angles but can still lead to interesting dynamic camber behavior as the instantaneous caster and kingpin change with steering angle. At the rear, the steering alignment considerations go away, but you can still lower forces through the toe link (and suspension deflection as a result) by minimizing your scrub radius and the moment that the center of mass of the corner assembly has about the virtual "steering" axis by moving the steering axis as desired - again, this steering axis, being virtual with the multilink setup, can be almost anywhere at the rear with steering geometry considerations largely out the window. I would absolutely love to have a long conversation with the suspension engineers that worked on this car, it's a really unique implementation even six years later. We're starting to see more and more complex suspensions (Mustang GTD has dual-lower-link "double wishbone" front, the highly-regarded Lucid Air has a true 5-link front, etc.) but still nothing quite like this to my knowledge. The rabbit hole goes way way deeper than what i mentioned above - bushing compliance, accounting for how the front pushrod outer pickup point on the control arm should technically move with steering, Ackermann considerations, etc. It's crazy how a production (well, "production") car can have such a wild decoupled spring/damper arrangement and it's still not necessarily the most interesting aspect of the suspension system as a whole
Something that hasn't been touched is that under one wheel bump no vibrations are sent over to the opposite wheel like on an anti-roll bar installation. This is as close as a suspension can come to perfection without being active.
Not only that, but antiroll bars increase responsiveness at a cost of overall grip because they pull against the inner tire and reduce the load on it. Springs and dampers have no effect on load transfer but roll bars do. You can run really stiff springs and get rid of the roll bars to have more even grip between all four wheels and higher total grip but that results in a really harsh ride and with cars that are traction limited you would spin out much easier since you wouldn't be able to load the rear tires
You guys are actually both wrong. As the dampers act on two different modes, they will have different spring and damper rates. So the forces of a single wheel bump is not canceled out and the other wheel will feel an effect of that. Second Anti roll bars do not decrease grip by reducing the load on the inner tire. They do not change the load of the tires at all, they just change the position of the chassis relatively to the wheels. Load on the tires is only determined by cornering radius and speed.
Talking about multilink suspension - it effectively puts your pivot points further inside the wheels (less scrub radius) , makers use this for more than 20 years ( VW/ audi, Merc, BMW, Nissan ...) mainly- because cars where getting faster and heavier which forced to use bigger wheels , tyres and brakes , not allowing to position suspension pick up point where they supposed to be. On damping - also it's an rather old concept made possible with damper technology coming where it is now , the only thing that is pretty hard to wrap the head about is - bump or drop on one side should effect another ... Unless damper takes those load's so well it becomes irrelevant. I'd be happy to be corrected . Good job on those videos , been following you for a while now.
The double-jointed outboard links are known as virtual ball joints and are designed to move the effective pivot point further outboard to reduce kingpin inclination, kingpin offset and scrub radius.
The beauty of this solution is its simplicity. I wonder why no one else thought of this. In recent videos of the Project One, I can see that there's no body roll and it doesn't feel bone cracking hard either, seems to glide over the rough bumpy tarmac of the Nurburgring.
can you do one on the aston martin one-77 since the shocks are in a very weird position in front of the engine on that pushrod like setup, one on the aventador would be sick too, great video
By separating the A-arms, they will be getting a 'virtual pivot point' for the steering. This brings the ball joints inboard, which makes room for a better brakes package. I.e. Brings it inboard too which allows for better cooling. Look how inboard the calipers are!
If you haven't already you should check out greenteam stuttgart's solution to the independent roll and heave spring/dampers. Pretty neat solution that doesn't require a through shock or two springs. On the topic of FSAE Suspension I'd love to see you do a video on AMZ's active damping hydraulically mode separated suspension, a bit of a mind bender.
Hadn't seen it, but pretty neat solution! Hadn't seen the AMZ one either (I haven't been keeping up well with FS lately!) but it looks like a slightly more elegant version of what Monash was running a few years back. I really love seeing these types of suspension out on FS cars.
KYLE.ENGINEERS yeah similar to monash's solution...leave it to the European FS teams to take it to the extreme. I'm on the UQ team and two of our previous cars (2014 & 2016) employed purely mechanical mode separation by using the under tray tunnels as longitudinal beams transmit roll load to a single ARB. Soft warp hydraulic mode separation would probably be ideal for off-road racecars like yours, lots of compliance for all the bumps.
AMZ's suspension system isn't an active one, it's all passive ;-) You also have to mention Rennstall Esslingen's roll heave system in their last two cars I think, but they have one heave damper and two roll dampers and also High-Octane Motorsports roll-heave system in their 2015 car. They only had one roll damper without an spring, but included an anti roll bar as roll spring.
Currently I work on my master's thesis which is about design new type of suspension for lmp1 car based on this idea. It is a lot o fun So far. Let's See what willl be the outcome.
Shouldn't damping be more soft on single wheel bumps, for more traction on the other wheels ???? Otherwise the car would tilt and I think you would lose traction on two or more of the other wheels.....
Kyle! there is something interesthing abou the rear arms as well! There is an extra rear suspension arm located between the rear arms, a 5th connection. It looks interesting as it seems, if you mentalize the suspension compression motion, as it has a shorter length when compared to the inferior arms, when the suspension gets compressed during heavy cornering centre roll, that arm tends to modify the toe angles accordilgly to the chassis rolling, opening it when compressed, closing it when lifted... Check this video at 7:30 and onwards! please! ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-EzV_1hF4pxM.html Also, it seems the engine doesn't have the same exhaust manifold the F1 car has. The F1 uses a shorter ram lenght exhaust duct and it makes sense for a road car since spooling time won't be a problem to it given it's electric motor in the compressor and bigger batteries. this buffs the lower rev area. I bet it doesn't have the same cam profile as well. And i see a stressed member chassis which means this car is going to be stiff and light. Only a few car manufacturers make use of the gearbox case as stressed member. Usualy it only suits racing cars.
So.... compared to a conventional setup with an anti-roll bar, more stiff on one wheel bumps(two springs and dampers active) and softer on roll, pitch and weight(one spring and one damper active). Hm.... Need furthur clarification on how much spring tension and damping is desirable on each modes.
At 6:04, wouldn't one have two separate shafts ? Not doing so would effectively relocate/create a "second" rocker pivot point that would, in effect, "lock the rocker on heave and compression? The horizontal damper would then be rendered useless if that was a single shaft! Additionally, looking at the gold mounting base, there are two sets of damper adjustments. High/low speed adjusters- I'm thinking separate left/right adjusters.
The prognostication, from 5 years ago, that 'grip is likely going to be obscenely good', appears to have been verified as True, given that the real, road-going (well, if you have USD 2.7m to spend..) AMG ONE, utterly smashed the road legal Nurburgring lap record, last December (ie 2022)
Couldn't they hypothetically use a traditional roll bar as the roll springing mechanism and then only require a typical damper attached to the roll bell cranks?
5:15 ...Upper roll spring cant strech... not retained...so used 2... FEEDBACK FROM AUDIENCE REQUESTED If a retainer was used could this pice be a single spring? Hope the new (years old) job is happy. Cheers. Wish you could work out some kind of deal for a "limited disclosure agreement " of non Co. data or analytics . If nothing else, that you may inspire your replacement who is only 7 years old now.
Thanks! It was very good! but why in this car we don't have an inerter or complicated suspension like on the current Mercedes f1 car? Maybe this suspension is good for this car but personally, I don't see this so complicated! Maybe this is all rubbish, but, it's my opinion!
This type of heave spring combined with laterally connected antiroll-mechanism is relatively common in road cars. The Citroen hydropneumatic suspension operates like that. Those hydraulic suspension spheres are laterally interconnected making them work like a heave spring. The hydraulic pipe offers some amount of roll damping. And there are lateral antiroll-bars front and rear to prevent the car from tipping over because of the lateral interconnection. About that separation of roll motion and single wheel bump motion. Aren't those two motions actually same motion: Wheels moving opposite directions in relation to each other. No matter how I look at that suspension mechanism, every time the left and right side wheels move in opposite directions in relation to each other, that motion is resisted by the roll spring and heave spring working in parallel. When looking at the rear suspension 3:05 if the left side rocker is rotated clockwice, that motion is resisted by both heave and roll spring because of how those springs are connected to the rocker at the right side. It would be interesting to see modern vehicle where the antiroll-bar is not connected between lateral pair of wheels, but is instead connected between longitudal side-pair of wheels. One for left side and one for right side. Citroen 2CV had this style roll spring (those coil springs inside suspension tubes, during roll or heave motion. During pitch they behave in different way), those spring were just very soft and because of that 2CV leans so much. The British Hydragas/hydrolastic was simple hydraulic version of that. LLTD could be simplified to simple geometrical leverage calculations, and ground shape would have no effect on lltd. I think with those front suspension arms, mercedes has tried to create centrepoint steering geometry using virtual geometries to remove KPI and offset. At least from pictures, it looks like the n-lines of the suspension linkages intersect somewhere around the center plane of the wheels, so the instantaneuous screw axis of the hub should be somewhere around the center of the wheel.
Hi there! Great explanation 👏🏽. Do you know how could potenciometers be applied on the roll damper? Do they need to have two linear potenciometers one for each spring or can it be just one since the springs will have the same symmetric movement?
So as some that might want to make this at home with parts you can buy what if you make a x with the shocks and leave the top parallel shock off. Or if a do that would I still need a top shock alone with the x patter on the bell cranks to get the same effects. I'm also guessing I'm not going to need springs with high tensile strength like I would in other set ups. Kinda would like to know how I could go about calculating what springs I need for a weight of the vihicle in a set up like this or the one I described.
What other cars have used this type of suspension? Is there a name for it that I can google? Ive seen a similar thing on Formula-Vee but its for heave only and its to encourage roll. Not minimize it.
The connected bell crank set up is genius. It is ant roll without a roll bar. It takes the preload off the wheel that isn't under pressure from the roll
Can you take a look at "Formula X" from project cars 2? Also why is compliance from the suspension such a problem in F1 but compliance from the tires fine?
Compliance from the tires is not fine, there is just nothing they can do about it since it's heavily regulated, the same for everyone and supplied to them so the suspension has to be designed around it. Thus you get things like mass dampers and inerters.
Comparing this car's aerodynamics to that of the P1 doesn't seem to be giving this car enough credit, one look at the interior will tell you that this car is a lot lower than just about anything but the Valkyrie and maybe the probe 16. It seems like it's aerodynamically in between the Valkyrie and the more conventional hyper cars. I have to wonder if the fairly recumbent driving position increases the risk of dozing off on long drives. Also I've never been this interested in fuel economy figures for a hypercar before.
The comment above mine is wrong. Of course it produces downforce given it's shaped like a giant wing. It also produces drag which is a byproduct of downforce.
the statement above mine is also wrong, since a windshield is not shaped like a wing at all. But a windshield will partially act like a spoiler as it directs the airflow upwards. And therefore the air will press down on the windshield
Actually, I believe it will provide some amount of lift. As air flows over it and back down over the rear of the car, the movement of the air will cause a low pressure area above the car. If this isn't compensated for by providing a similar affect under the car, it will have a net upwards force on the vehicle. Airplane wings provide lift using this concept, having a curved top that causes the air to move without providing to much drag, causing a low pressure area on top, and a flat bottom that leaves the air as undisturbed as possible to keep the pressure as high as possible. Wings on racecars are similar, but turned upside down, and are far more aggressively shaped because drag is a much smaller issue when downforce is so beneficial. I believe the tops of them are curved because it can provide a ram air effect, raising the air pressure on top at the expense of very high drag. The most important part of them is the bottom though, where the air is made to be at the lowest pressure possible over the largest surface area possible.
Note that the ONE was a prototype for a road car, not a race car. If you can call a car than, now it may be bought, costs almost USD 3 million, can fairly be called a 'road car'!
can one say that this is independent suspension then? I would think by lookin at this that a bump felt by one wheel could disturb the other due to the higher damper.
the problem is with that roll system. corner one way and the damper compress then rebound on return yet corner the other direction the damper droops then compresses on return..since there is one damper on roll...how would they set that up? mechanically speaking if compression and rebound are set at different loads...having one damper handling two opposite characters would be really complicated
No double a-arms in the back? How is that a revelation? My 1996 C220 has a 5-link rear suspension. Mercedes only recently downgraded to double a-arms on their rear suspensions.
believe or not but toyota has design and manufactured similar suspension on the front suspension,its called superstrut suspension and it can changes caster during steering change and i believe reduces torque steer,it was use in corolla ae111 levin bzr
They did this superstrut on the Celica GT 4. This is not at all similar to the mercedes suspension... But it does indeed reduce torque steer by giving a lower scrub radius for example.
The thing I’ve never understood about these types of suspension is wouldn’t the assymetric setup be more effective to one side than the other? or is that counterbalanced with weight distribution?
Simple, adjust the preload on the heave spring! Or alternatively you could increase the pressure inside the heave damper, which will increase the shaft force.
If you had an active cartoon of how it worked in motion it would be a lot easier to understand but I know thats expecting a lot and not that easy to produce but I didn't understand anything that was said as interesting as it is. I guess I'm just a dummy. Thanks anyway
Here is not a direction or device action, only where your finger is pointing. I found it quit annoying after the 7th time you employed this term. Try compression, extension, stationary, relaxed or immobile as the the likes of a non-car subject would mindlessly watch. DK, ASE master tech since 78, retired.
