+ Reply to Thread
Page 5 of 11 FirstFirst ... 3 4 5 6 7 ... LastLast
Results 41 to 50 of 105

Thread: Roll rates in RCVD

  1. #41
    Quote Originally Posted by nowhere fast View Post
    Dynatune,

    You are telling us that soft warp suspensions did not prove to be beneficial in your experience, but you aren’t providing an engineering interpretation of why that was the case. I’m not saying this as a criticism; but because I am genuinely interested to hear your reasoning on why it was not beneficial. People like me that have no experience with soft warp cars* are left to reason our way through the issue in our own minds, and this is why merely telling me that your experience suggests a conventional suspension is better than soft warp suspension is an unconvincing argument.

    * I have driven several soft warp vehicles including a forklift, wheel loaders, and a grader. The grader in particular was very smooth over bumpy terrain; how it would handle around a racetrack is another matter though.



    Z,

    I understand how LLTD can be adjusted on a zero twist stiffness car and have posted about it before:
    http://www.fsae.com/forums/showthrea...1157#post21157

    But I have, perhaps incorrectly, been using the terms warp and twist interchangeably. I don’t recall a formal definition of these terms, and have been using the term ‘warp mode’ to refer to any mode where the front and rear axles roll relative to the chassis by different amounts (opposite directions implies a negative amount); bringing the wheel prints out of a flat plane. Others might also be using their own slightly different definitions of ‘warp’.


    Nathan,

    The main problem as I said was the fact that the car due to it's low warp stiffness did not sufficiently quickly build up (diagonal) load transfer which lead to a rather slow build up of tire slip angles, which was generally interpreted by the drivers as a "lazy understeering pig" setup. Imagine a road car with soft roll bars and then putting some really stiff bars in it ... the car changes from day to night . ... Beyond that there were issues with roll center heights (that do also affect the LLTD) which made finding a good setup a pain. If one considers that for instance on flat surface circuits like Magny Course we put into the car the stiffest springs and rollbars in the car that we could find (to keep the tires as hot as possible ....) one can see that there are many other factors that can come into play than "just" having a good idea .....In those days I had the pleasure of working together with some of the brightest vehicle dynamics engineers and we did look at all the options known to man-kind in those days. And with us there were many other teams looking into the matter. The 90s were famous for all the developments on suspension technology. Apart from the fact that some people here in this Forum do tend to "overestimate" their seniority, I have always worked on the principle that "if an idea of mine has not won the hearts of the clients and especially if the clients are very capable of judging whether the idea works or not, something might not be as good about the idea as I initially thought". And especially having tried it myself, together with many other clever people, especially having seen what "alternative" technology can do, and especially since neither of us all ha been able to make a faster car than a "conventional" car made this particular idea no "winner". These are just some simple facts, experienced in a world where one was prepared to spend 1.000.000 £ for each 0.1sec of lap time, without any politics or "presumed" so called expert knowledge.

    If anybody believes he can do better (since in his visionary wisdom obviously all the "other ones" in the past have been complete & utter idiots, or if active now, are at the least ignorant peers) this person is either the long awaited genius or one of those poor souls that will wander around the globe forever waisting his time in trying to baptize the non-believers ....

    Cheers,
    dynatune, www.dynatune-xl.com

  2. #42
    Dynatune,

    Can only agree 100 % with you on the vehicle dynamics experience and explanation of warp stiffness as well as the understanding of both usefulness and appreciation of different engineering methodologies. But I am only one of the codswallop idiot of this forum.
    Claude Rouelle
    OptimumG president
    Vehicle Dynamics & Race Car Engineering
    Training / Consulting / Simulation Software
    FS & FSAE design judge USA / Canada / UK / Germany / Spain / Italy / China / Brazil / Australia
    [url]www.optimumg.com[/u

  3. #43
    Senior Member
    Join Date
    Mar 2008
    Location
    Brighton, MI
    Posts
    686

    Warp with a twist of Limey.

    Actually, there is an important difference between the technical aspects of twisting. Warping is the displacement of the elements of a beam along the twist axis. It is caused by the non-colinearity of the shear center axis and the centroidal axis of a beam.

    So,,,, a solid tube does not warp but twists. A slit tube will warp as it is twisted. The edges of the tube extend in opposite directions.

    For example, in a twist axle type of (rear) suspension, the U or V beam is usually oriented in such a manner that the edges welded to the control arms are steered when the vehicle rolls during cornering. The orientation of the shear (warp) center defines whether the steer by roll characteristics are understeering or oversteering (or neutral steering). Anyone care to speculate why a U or a V section shape is chosen ??

    When it comes to body/frame/ chassis warping, the same observation is made. (The body/frame is not a tube, that's a mistaken assumption, called a tube steak). When the structure is twisted from induced roll or single bump inputs, the fore/aft displacement (warping) of the suspension attachments is considered and can be measured and quantized.

    One difficulty in the understanding of all this is that the torsional stiffness (moment stiffness distribution) is certainly NOT distributed evenly down the length of the body/frame. The powertrain contributes vastly to the body stiffness because of the engine containment strength requirements. And, the powertrain is the most likely cause of twisting moment reactions from bumps and cornering. Its inertance and inertia change the tuning needs for roll moment and roll damping tuning. Something as simple as a pencil brace or an underslung cross cable can significantly alter the TLLTD percentages (front to rear). Modal analysis tools are often used to identify the mode-shapes active in each type of event and structural and other parts are chosen or altered to reinforce or accentuate 'modal participation' or modal 'isolation'.

    That's warping (with a twist) from a structural mechanics perspective. (I wrote an SAE paper on this that's a lot older than most of you).

  4. #44
    Dynatune,

    I see your point that in some circumstances a setup that is ‘worse’ in terms of additional tyre scrub or tyre vertical load variation over bumps can improve lap times by bringing the tyre temperature up.

    If the issue with soft warp suspensions was the rate of diagonal load transfer and not the amount; then I don’t understand why bumping up the stiffness of the longitudinal Z-bars (the springs which resist roll in this case) would not speed up the transfer in a similar way to increasing the roll stiffness of a conventional car by reducing the final roll angle. Of course other factors including damping and kinematic load transfer become more prominent when we are talking about transients.
    A soft warp suspension introduces a new set of compromises; roll stiffness may be linked to heave or pitch stiffness depending on the implementation instead of being linked to warp stiffness as is the case with ARB's, maybe there are some issues here. It seems that stiff roll and heave modes would be desirable for a race car though, considering conventional race cars are often fitted with ARB's + 3rd springs.

    I have no horse in this race, and am open to hearing both sides of the debate. I just find this topic interesting.

    Quote Originally Posted by BillCobb View Post
    Anyone care to speculate why a U or a V section shape is chosen ??
    I’ll bite. Is it because the shear centre lies a long way from the centroid in U and V sections compared to other shapes? (Shear centre some distance below the U, while the centroid is somewhere inside the U).
    Nathan

    UNSW FSAE 07-09

  5. #45
    Quote Originally Posted by BillCobb
    For example, in a twist axle type of (rear) suspension, the U or V beam is usually oriented in such a manner that the edges welded to the control arms are steered when the vehicle rolls during cornering. The orientation of the shear (warp) center defines whether the steer by roll characteristics are understeering or oversteering (or neutral steering). Anyone care to speculate why a U or a V section shape is chosen ??
    Quote Originally Posted by nowhere fast View Post
    I’ll bite. Is it because the shear centre lies a long way from the centroid in U and V sections compared to other shapes? (Shear centre some distance below the U, while the centroid is somewhere inside the U).
    The orientation about the axis of symmetry of the axle dictates the orientation planes of shear induced warp at the axle ends. The shear deformation (warping) either contributes to roll steer (gain or loss) for a beam orientated like up or down (looking like an n or U ) or even camber gain/loss if the axle is orientated like [ or ]. This is for a very basic prismatic axles and is dependent on how they are attached to the control arms. Feasibly the axle could be used to correct or enhance roll steer induced by the angle of the control arms relative to the body.

    Another particular reason for having an open section axle though is that it significantly reduces the torsional stiffness of the axle hence preventing it from acting like a large and very stiff ARB (i.e. decoupling the roll and pitch/heave modes). As Nathan says the shear centre is significantly offset from the centroid and as the centroid is likely to lie very close to the axis of the couple exerted on the axle produced by the control arm movement, the torsional stiffness is reduced as much as possible as is the transverse shear stiffness of the axle section. This results in more warp at the axle ends and potentially a significant amount of slip angle loss/gain at the wheel with subsequent ramifications for changes in yaw contribution from the wheel pair (understeer/oversteer effects). The roll stiffness contribution from the axle then also changes the LLTD during roll hence undesired changes in handling can creep altering dynamic behaviour.

    For relatively simple axles (mechanically), twist axles still have a lot going on in terms of how they behave structurally and kinematically and how they can alter the forces at the contact patch.

  6. #46
    Senior Member
    Join Date
    Mar 2008
    Location
    Brighton, MI
    Posts
    686
    Very Good replies to both of you. Comforting to know that some people still know the WHYS of car designing instead of the common alternative.

    Now about that inclined roll axis stuff....

  7. #47
    Senior Member
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    1,690
    I often wonder if this Forum isn't hosted by the "Society for the Appreciation of the Dramatic Arts, and Other Trivial but Delightful Indulgences...". You know the sort of place, where NO well-reasoned thoughts, or numbers, or facts, or any sort of rational thinking AT ALL is required to support your argument. Nope, you just call everyone "Daahlink...", and then try to scratch their eyes out!

    Here is an example.

    Originally posted by Dynatune:
    ... since there was a flagrant lack of load transfer to make the car react sufficiently on corner entry.
    Question asked by Nathan:
    You are telling us that soft warp suspensions did not prove to be beneficial in your experience, but you aren’t providing an engineering interpretation of why that was the case.
    Reply to Nathan given by Dynatune:
    ...the car due to it's low warp stiffness did not sufficiently quickly build up (diagonal) load transfer which lead to a rather slow build up of tire slip angles, which was generally interpreted by the drivers as a "lazy understeering pig" setup. [* See Note 1.]
    ...
    Beyond that there were issues with roll center heights (that do also affect the LLTD) which made finding a good setup a pain. [* See Note 2.]
    ...
    In those days I had the pleasure of working together with some of the brightest vehicle dynamics engineers and we did look at all the options known to man-kind in those days. And with us there were many other teams looking into the matter. The 90s were famous for all the developments on suspension technology. [* See Note 3.]
    ...
    [followed by lots of hissing, and snarling, and scratching of eyes...]
    * Note 1. If we are allowed to get technical for a moment, then it is my interpretation that a "slow build up of tire slip angles" means that the tyres are rolling in the direction in which they are pointed (ie. they have a SMALL SLIP-ANGLE!). So if the driver has actually steered the front-wheels into the corner, then the front of the car will also being "turning in to the corner" just as quickly. So, quick turn-in, and NO UNDERSTEER.

    So, Dynatune, can you please give a technical explanation for why a car with low front-wheel slip-angles can be described as a "lazy understeering pig"?
    ~o0o~

    * Note 2. Roll Centre heights have no direct relationship with springing, in that they can be adjusted entirely independently of each other. Or put another way, if your RC heights are inappropriate for the car, then that is a problem related to RC heights, and not to the type of springing used.

    So, Dynatune, can you please give us some technical insights (with as much detail as possible, maybe even some numbers) as to why you, and the other "brightest vehicle dynamic engineers", had difficulty separating the effects of RC-heights and modal spring-rates, and their F/R LLTDs?
    ~o0o~

    * Note 3. I lived through the 1990s. I cannot remember a single significant development in suspension design from that time. At least not one that had not already been thoroughly investigated at least 30+ years earlier. And then, perhaps, forgotten by the "brightest vehicle dynamic engineers"...

    Anyway, I might be asking too much here (geez, 3 x tech questions, all at once!!!), but...

    Dynatune, do you understand how a "mechanical" suspension can be completely Twist-soft, and still be able to give any LLTD from 100%F:0%R to 0%F:100%R? (And, BTW, also any ratio outside that range, though not recommended.)
    ~o0o~

    Well, I would love to talk more "engineering" than that, but perhaps too much here already (hiiiissss, snaarrrlll!!!).

    Z

    (PS. What the heck! Bill's "warping" of flexible structures (= twist-beams) is a bit different to a suspension's movement of its four wheelprints out of a flat plane, but interesting nonetheless. I think Nathan and Loz got it right...
    * Shear-centre lower = roll->toe-out-of-corner = +OS.
    * Shear-centre higher = roll->toe-in-to-corner = +US.
    * Shear-centre forward = roll->less-camber-gain = +OS.
    * Shear-centre backward = roll->more-camber-gain = +US.
    * And, most importantly, twist-beam-near-chassis-pivots = easier-packaging-of-junk-between-the-wheels, hence inverted "U/C-shape".
    Also, lots of interesting historical "twist-beams" from the 1930s, but best not flood the Forum with too much tech at once! )
    Last edited by Z; 04-03-2014 at 09:13 PM.

  8. #48
    Senior Member
    Join Date
    Mar 2005
    Location
    Modena, Italy
    Posts
    363
    Quote Originally Posted by Z View Post
    * Note 1. If we are allowed to get technical for a moment, then it is my interpretation that a "slow build up of tire slip angles" means that the tyres are rolling in the direction in which they are pointed (ie. they have a SMALL SLIP-ANGLE!). So if the driver has actually steered the front-wheels into the corner, then the front of the car will also being "turning in to the corner" just as quickly. So, quick turn-in, and NO UNDERSTEER.

    So, Dynatune, can you please give a technical explanation for why a car with low front-wheel slip-angles can be described as a "lazy understeering pig"?
    I'd interpret "lazy" as being a fast build up of slip angle on the front tyres, but without the corresponding lateral force. So something is delaying the build up of lateral force. Possibly some load dependancy on the relaxation lengths maybe? I know they are supposed to be constant, but given that response times to steering are in the range of milliseconds, an increase of relaxation length of 100mm will add 3.6ms to the response time.

    Also, I guess that with very different normal loads on the front tyres during turn in, a soft warp suspension might need a different steering setup w.r.t. KPI, caster and ackermann to get the same response as before.

    By the way, in the 90's, in F1 there were a lot of attempts at hydraulics and other interlinked systems to try to mimic the behaviour of the active systems...
    Last edited by Tim.Wright; 03-31-2014 at 07:17 AM.

  9. #49
    Tim is correct with the interpretation of "lazy" but it was not the front causing this behavior, it was actually the retarded delay of building up of slip angle on the rear tires. Compare it to a step steer maneuver where on measured data one can see the "braking" free of the back end at the "famous" initial negative slip angle turning into positive and the small pike followed by drop off in the lateral acceleration followed again by the increase towards peak values.

    Yes indeed in the 90s there a lot of attempts to mimic the behavior of the active systems .... I was trying to explain that 3 pages ago, not with much success though - that all the interlinked systems were not all as good....

    Cheers,
    dynatune, www.dynatune-xl.com

  10. #50
    Senior Member
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    1,690
    Quote Originally Posted by Z View Post
    Dynatune, do you understand how a "mechanical" suspension can be completely Twist-soft, and still be able to give any LLTD from 100%F:0%R to 0%F:100%R?
    Dynatune,

    ?????

    Z

+ Reply to Thread
Page 5 of 11 FirstFirst ... 3 4 5 6 7 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts