Vehicle Dynamics starting points and design process

[Luniz]

Z, Menisk, others: Would you mind continuing your argument elsewhere, preferrably offline? It has kind of steered away from the original topic and onto general FSAE philosophy and rules committee bashing, as usual...

Christian: You're right about the uprights! Welded sheet metal has the advantage that you can make them by yourself and a 10% increase of stiffness in the upright probably gives you more performance gain than a 10% decrease in weight...

Concerning the brake cylinders I posted earlier, yes I meant to say 8mm rod ends... In case anyone is interested, they're made by a VERY small company in eastern France called "Beringer", their lead times are sometimes a few months and sourcing them can be quite tricky.

[mech5496]

Lutz if rodends are 8mm, it is not particularly compact. I know for isntance that AP makes one pull-type model that is really small and light (we use them since 2010). Beringer makes some top-notch components for Supermotos and go-karts, so they might worth a look as well.

[ChristianChalliner]

Forbes - Yes I did find that document and presumed it was the one mdavis was talking about, thanks anyhow

Z - It would be good to see those photos! Interested to see the results. The CAD design for the uprights is a good shout and actually something the eco team have already done for some of their folded alloy parts so I'll mention that to the uprights guy. We have a good welder who can weld thicknesses as low as 0.8mm with TIG effectively so there's no issue on thinner material, of course final machining will be done after welding otherwise the bearing housing would not be round We are definitely focusing on an early build finish although it has been put forward that we are over stressing some of the members somewhat but I believe it will be worth it in the long run.

Luniz - Looks like we're going down the right track on that one then, it also has the advantage of being significantly cheaper than CNC'd alloy and can be produced much faster (no need for extensive CAD models, etc). I will look for the master cylinders later.

Mech- That would depend on what you consider compact, I have no arguments with an 8mm end if the cylinder itself is tiny, likewise I have a problem with a 6mm ended component that is enormous!

I did plan on writing a lot more than this but things came up as usual Again, thank you for all the help all of you.

[ChristianChalliner]

Firstly, sorry about the double post!

Secondly, I've drawn up a rough idea for a front beam axle using 4 trailing arms which is similar to how the ECU car is at the rear. The basis of the design (described in RCVD on page 6XX, sorry I don't have my copy right now) is one where the lower links join up to form an A-frame and the upper links are parallel.

The problem I'm having is that it is not possible to have the upper links exactly parallel because then they would be sticking excessively far out from either the beam or the chassis. What this means is that the convergence point for the upper arms is reasonably close to the car so when the roll axis is drawn between the upper and lower convergence points it is excessively steep. If I understand RCVD the roll axis slope is equal to the amount of roll steer, in which case what is shown below would be ridiculous!

I CAN lower the roll axis to around 1 degree of roll oversteer however, this then means that the upper arm is sloped downwards a large amount. I will confess now that I don't properly understand anti-dive and anti-squat so I'm not sure on the effects this will have, if it is similar to a double wishbone then surely sloping the upper arm downwards will promote a pro-dive effect?

Anyway, the pictures can do it more justice than my explanation so here goes:





I appreciate that there are other issues with this design and I'm working on solving them however, inputs are always welcome

The main reason for picking this design was that it places the roll center in the 5-10cm region that Z recommended and that it also makes fabricating the lower arms and beam easier: simpler jig, one less spherical, less welded tangs on the beam, etc.

[Z]

Christian,

Will post Beam-Axle suggestions in a few more days...

My laptop power supply EXPLODED last night!!! Big fireworks. Burnt a hole in the table! So battery now draining lowww....

Z

[DougMilliken]

My laptop power supply EXPLODED last night!!! Big fireworks. Burnt a hole in the table! So battery now draining lowww.... Z

Seems like your life is always dramatic(??) Fwiw, I've run a laptop (power supply label says 16V DC) from the ~14V in a car (with alternator charging). Might not charge the laptop battery, but will run the computer.

[Loz]

Christian

Looks extremely similar to the beam axle on the front of the 2010 ADFA car. Although, the ADFA car used a pair of leading arms converging at the centreline under the vehicle.



Before you get into any kinematics, there are lots of other considerations you need to work through with statics (although from your drawing the design is already heading beyond conceptual design and into detail design - not what I would call "rough").

The design (beam or otherwise) needs to first be able to appropriately transmit the tyre forces into the body while static (i.e. we are talking about action/reaction and stress/strain, although more specifically, managing system compliance).

With a beam axle the steering concept should also be well thought out and derived early in the process so that you have a good understanding of the nuances which will likely point the approach in a different direction compared to that of independent suspensions.

Even though a beam can be a simple system which delivers a range of favourable system attributes and can enable the delivery of a high performing vehicle in a short time frame, there are also just as many opportunities to fudge-up the design as there are with double A-arms etc, if you are not careful.

Fortunately, the good news is that with some logical thought, application of engineering fundamentals (primarily dealing with forces) and purposely keeping it simple, you can avoid crossing paths with many gremlins that may bite.

Loz

[ChristianChalliner]

Loz - You are correct in thinking it's extremely similar, that's because to me it seems it is the best way of fitting a front beam under the car in the style I intended to use, I did try a folded structure type but found it tricky to fit underneath the body whilst mounting the damper to it in a manner that meant I was meeting the minimum travel requirements. By the time I had got it to work I had something which looked suspiciously like a big U shape so decided to try just a bent tube instead which for the most part fits the needs well and cuts down on the manufacturing.

I understand that the steering is tricky, I have put some thought into it already although I've not written/drawn anything out that I can show yet but I was thinking along the lines of a system which went: Steering wheel > Bevel gearbox > telescopic column vertically to floor > steer arm > link out to upright, along with some manner which coupled the column length to the axle vertical movement in order to prevent wheel toe changes, something even as simple as just a point on the axle which pressed on the underside of the steer arm on the column would do it? I was also thinking that there is no rule stating that both wheels have to be steered independently so why not steer only one upright and run a large link between the two behind the axle line? would give good ackermann I reckon...

Here's some pictures of the arrangement with lower roll steer but resulting in a more aggressive VSAL:





I also have some impact attenuator points I've been asked to discuss by the guys who are working on that. We have an Instrom testing machine capable of performing slow crushing of components (so not a true dynamic test) however, every time we've made up a standard impact attenuator and mounted it to a bulkhead the result has been what is shown below:





For this test the absorbed energy was 4100J which is both below the minimum 7XXXJ required and clearly the bulkhead has collapsed/deflected beyond the maximum 25.4mm , the deflection of the system was 150mm. I did not conduct or view the test but I can ask any questions if anyone has any.

[mdavis]

Christian,

Are the tubes in your frame free to move in the lateral direction (the way they splayed out during testing)? If not (hopefully not), you may want to constrain them. What we did was put some sheet steel (I think .065, extra from making bulkheads, but whatever is laying around should work) strips between the 4 stand-off tubes. This constrained them to each other, and fixed our very similar problem. Make sure you constrain both directions, otherwise the bulkhead will bend along a different axis.

-Matt

[ChristianChalliner]

mdavis - Thanks, yeah we did think about doing that for our next test and it was actually something we mentioned before we crushed it so we will try that Also, some clarification about the deflection, I've now been told that the plate only began visually deflecting after the instrom measured 150mm deflection so I assume the attenuator had already deflected around 150mm before this happened.