Suspension Design

What Has Been Established So Far.
=============================

1. Anyone who believes the world is a globe circling the sun is a lunatic, a risk to society, and should be burnt at the stake! As everyone knows, the world has hills and dales, which makes it truly 3-D. But other than that it is a flat 2-D plane. Err..., supported at its corners by four huge elephants... that stand on a giant tortoise... and below that it is turtles all the way down! http://fsae.com/groupee_common/emoticons/icon_smile.gif

2. If the Bible of RCVD says "In true three-dimensional space, instant centers are replaced by instant axes..." (p612), then it MUST be true! And, of course, the 250+ year old concept of a "screw axis" is the work of the devil!!! If a computer Suspension Program claims on the packaging that it is "fully 3-D" (like the one Z trialled a few years ago), then that MUST also be true. Even if the results it gives for "anti-pitch/roll" are calculated according to RCVD type "instant axis theory", so often have errors of 10++%. http://fsae.com/groupee_common/emoticons/icon_frown.gif

3. A US manufacturer of pick-up trucks makes the goddamn best handl'n vee-hikels on the whole gosh-darn planet! Apparently, they do this by spending a lot of time in casinos rolling dice... Hey, that's gotta be more fun than doing geometry, especially during Happy Hour! http://fsae.com/groupee_common/emoticons/icon_smile.gif Oh, and Porkers handle like... well..., porkers... http://fsae.com/groupee_common/emoticons/icon_frown.gif

4. Z will never, ever, ever be welcome in a Product Development Design Review Committee Meeting. Err... ever! ("Yahoo!", thinks Z, "They're sooo boring..."). And the game of Tennis, like so many other fields of endeavour, really does benefit from KISS. Winning Tennis players do, indeed, just keep hitting the ball over the net. By analogy, winning FSAE cars just keep going fast around corners. Which they can do without suspension... http://fsae.com/groupee_common/emoticons/icon_smile.gif

5. Finally, for those slackers who are too lazy to do their homework (try Find!), Z is a nobody from nowhere, who has done nothing, no-wise, ever... Err..., other than mindlessly ramble on about some long lost bunkum called "Classical Mechanics". And, err..., endlessly ask for some rational discussion on suspension design. But, geez... "reason" ain't gonna educate engineers, is it? http://fsae.com/groupee_common/emoti...n_confused.gif

~~~o0o~~~

Moving on..., (or back?) to FSAE suspension design.

So far nobody has argued against the notion that "suspensionless cars" can go fast around smooth circuits. Likewise, no compelling arguments that the most complicated suspension types are necessary for a FSAE car to be fast.

Is this agreed? Any "Nays"???

So, on the assumption that FSAE suspensions could be much simpler, I will, in later posts, go through the pros and cons of the top five on this list.

Quote:

---

... the major suspension types are,

1. Beam-axles (eg. on all vehicles ever, can be live or De-Dion at rear),
2. Sliding-pillar (eg. Morgan and Lancia, and most motorbikes and aeroplanes at front),
3. Lateral swing-axles (eg. Tatra (still on their trucks), can be high or low-pivot),
4. Leading and trailing-arms (eg. Citroen 2CV front and rear, and most motorbikes at rear),
5. Semi-leading/trailing-arms (eg. F100 at front, and many mid to late 1900s cars at rear),
6. Strut-wishbone (eg. McPherson at front, strut at rear),
7. Double-wishbone (eg. originally mostly at front to allow shorter wheelbase with front engine),
8. 5-link (eg. recent fashion, mainly for NVH reasons).

Of these, any of the first five are more than adequate for a winning FSAE car.

---

But would anyone else like to give their thoughts first?

For example, Claude says,

Quote:

---

While I am less convinced about the efficiency of semi-trailing arms and even less about live axles (actually I call them dead axles, I always have been convinced that you can make a good car with a McPherson.... but if you if you miss the window of good mix between of tire exploitation and kinematics design it will be more difficult to make a change. Providing some suspension pick up points adjustability, with a 5 links or a double wishbone you can get and/or adapt the heave and roll camber variations, roll centers, pitch centers position and movement to what the tires (and good drivers) "need" much better than with a McPherson. And if for whatever reason you are forced to switch to another brand of tire, your design could not be adaptable...

As a student I will most probably go for a double wishbone or a 5 links.

---

And Tony says,

Quote:

---

Double wishbones are an obvious natural choice at the front, but at the back, I am not so sure.

There can be a lot of problems with the potential for bump and compliance steer at the back with a double wishbone design, which seems to be a regularly occurring design problem area on many FSAE cars.

Good solid toe control at the back would have to be be pretty high on my wish list.

A single stiff rear trailing arm, with transverse upper and lower camber control links would give excellent toe and bump steer control while being simple.

---

Now I don't like any of the above options, and have rational critcisms of them which I will give later (and, of course, which is why I'm not welcome in any meetings! http://fsae.com/groupee_common/emoticons/icon_smile.gif).

But in the meantime, Claude and Tony, would you like to "defend your decisions" any further, in a Design Tent sort of way?

It might be helpful to the "education of young engineers".

Z

Z,

there is the distinct possibility that some of these "young engineers" will be lucky enough to end up in the automotive industry, from which they will be promptly expelled should they suggest beam axles on a sports/race car. My current car has a rear beam, and yes it is fantastic on a nice smooth road (in Australia? Where?) but off that it can give some frightening and unexpected tail movement.

Some input from one of the ADFA guys would be interesting here, as would someone from Monash. Independent to beams on the former, direct actuation to push/pull rod actuation on the latter.

Ah !

All working professional engineers have to function under constraints, sometimes very severe and unreasonable constraints, be they political, managerial, or financial.
And so do student teams.
And sometimes, just sometimes, a really clever SIMPLE design can really triumph in cost, performance, and reliability.
And that can take you far.

One day your boss will demand you and your team come up with a superior design of mouse trap. And do it in two weeks.......
And then JUSTIFY all your decisions.

That is what FSAE is really all about.
Boot camp for real practical engineering in the real world.
And pie in the sky ain't really a part of that.

And neither is noxious aggressive hubris.
Be kind, be patient, be humble, and help others whenever you can, or remain silent.

Have fun, and creative engineering can truly be a most fulfilling vocation.

The ADFA beam axle car was designed using a clean sheet approach and a fresh look at what the FSAE rules were all about. After learning FSAE the monkey see monkey do way over 4 years as an undergraduate I was lucky enough to have another go as a masters student after a year to reflect on my undergraduate experiences. I should caveat that Z was consulted during the conceptual design of the car. He might come across a bit gruff and blunt at times on this forum but I cold called him from the telephone book and went on to chat to him for about 3 or 4 hours discussing different concepts. Z was instrumental in helping me to muster up the courage to do something very different to the accepted norm.

The first question I asked myself was what is the job of the suspension. I came up with the following answers:
- To control wheel attitude (camber, toe, steering angle, wheel base and track)
- To allow adjustment to roll stiffness distribution (elastic and geometric)
- To try and minimise contact patch load variation over undulating and rough ground
- Allow for +- 25mm suspension travel to be compliant with the rules

The team I was designing the car for had next to no construction or design experience due to a break in the program at our uni so it was basically made up of 1st and 2nd year students who had never built an FSAE car. I very quickly ruled out double A arm suspension because of the number of jig points required on the chassis, number of components required for fabrication and the jigging required to make the A arms.

The beam axles gave me the ability to do all the things I listed above and only required 10 accurate points on the chassis. I think I talked about this more in the beam axle thread I started. Now that the car has been running for 4 years and competed in 4 competitions (FSAE-A x 3 and FSUK x 1) I think we have a good feel for the weaknesses of the design. The first main weakness I discovered was excessive camber compliance. I designed a camber compliance rig and quantified it, we then welded in some gussets which made a significant improvement and was very noticeable when viewing the car on the track and improved tyre wear. The next one is toe compliance in the steering mechanism. To avoid or minimise bump steer I came up with an R&P mounted to the beam, this combined with trying to fit steering linkages inside a 10in rim meant I had to use a bell crank to transfer the rack motion to the steering arm. The steering arm design, number of joints in the linkage and large linkage angles (close to straight) on the inside wheel at full lock all contribute to make the mechanism more flexible than I would like.

The car has a large amount of steady state understeer and requires much more roll stiffness on the rear to try and tune it out. I think this is partly due to the roll axis being too close to level and the toe compliance in the front. We also have very short trailing arms on the rear axle which seems to promote brake hop. It’s hard to say if it’s all the suspension as the engine tune isn’t great, especially on the over run and not many drivers have mastered down shifting properly.

I have no regrets in choosing that design and would love the opportunity to develop the concept and take it further through better understanding and detailed mechanical design. One of the biggest issues with choosing the design is that as new people have come to the team it has been very hard to get them onboard with the concept and explain it to them. They can’t pick up any of the popular suspension or vehicle dynamics books and learn much about it. They should read the Olley book put together by the Millikens and a few good FSAE papers on twist axles but they are not what I would call Tune to Win level references. The car was received very sceptically at its’ first competition and design event. We took the feedback on board and then managed to make it into design finals at FSAE-A and FSUK turning quite a few heads along the way.

I have recently been talking to the current ADFA team as they approach the design of their next car and the most common thread of conversation is ‘we don’t know enough’. Any first year team and most undergraduates (even post graduates) are in the same boat. FSAE is an opportunity to learn and learn through mistakes in a relatively benign environment (compared to real racing, mining or aerospace). So my advice to new or inexperienced teams is take a reasoned stab if you don’t know about a parameter and design in some adjustment. If you stop worrying about design decisions and make some you could even have the time to redesign and iterate to improve towards the end of the project. I think it takes years to learn enough about FSAE and vehicle dynamics to be able to properly design a race car system with high level design targets and seek to predict the performance through modelling and simulation and then verify through testing. I am learning more and more in my career in industry that coming up with a solution you are confident will work through reason and experience with a little analysis to prove it is a highly sought after skill. Analysis paralysis will not take you far in a career. A good engineer will use a variety of design methods to achieve results i.e. it looks right from experience method, big and strong can’t go wrong and finely optimised and refined. They all have their time and place and it’s about knowing when to use them.

Cheers

wah wah wah wah Look at me I'm the smartest man in the room! wah wah wah wah

http://www.saabrally.com/forums/down...1217271771.gif

Quote:

---

Originally posted by BillCobb:
You know what 'Saab' stands for, right? Still Ain't A BMW

You know what Lotus stands for, right? Lots of Technically Useless Shit

You know what Pontiac stands for, right? Penniless Old Nigerians Think Its A Continental

You know what Porsche stands for, right? Proof Only Rich Suckers Can Have Everything).

Historically Orientals Never Drove Automobiles

---

What does Volkswagen stand for? http://fsae.com/groupee_common/emoticons/icon_razz.gif

Thanks for the response Olly

Interesting to note that even a much simpler design had its share of problems. I always saw the ADFA car as an interesting concept that could have gotten some much better results with some gun drivers

I dont know if I should be happy because you complimented our car, or sad because you insulted my driving (though I admit it is quite average).

One of the major issues that affects ADFA in that sense is the university only has a total population of around 1000 people, you dont have much of a pool to draw on so it can be difficult to find a team let alone drivers of any skill.

When you take into account the time burden as well, ADFA does a lot of military training instead of uni holidays and also seems to really enjoy scheduling "compulsory fun" activities in fairly regularly, there isnt much time left to do driver training.

The above points are some of the other reasons why we went down the design path that we did. I manufactured the suspension for the last double A-arm car in 2007 and was also involved in the manufacture of the beams. The beams were signifigantly easier to make, they required less skill from the people working on them (the majority of the sheet metal work was done by first years, if they screwed it up they just did it again as sheet steel isnt that expensive). The majority of the components on the car were designed to be done this way, granted some items on it are rather complicated (rear bulkhead, sump plate etc) but these were manufactured as part of the team sponsorship deals.

Designing with the ease of manufacture in mind allowed a small team to build a car around the constraints and still get testing and driver training done so that we didnt look too bad on track.

Having said that, I would love to see that car go round the SAE track with some good and experienced drivers in it, just to see what it is actually capable of.

Quote:

---

Originally posted by Z:
As everyone knows, the world has hills and dales, which makes it truly 3-D. But other than that it is a flat 2-D plane. Err..., supported at its corners by four huge elephants... that stand on a giant tortoise... and below that it is turtles all the way down! http://fsae.com/groupee_common/emoticons/icon_smile.gif

---

When I want to pull up a map and figure out the distance to get from my home to the local deli, to be precise I should be considering the curvature of the earth and its effect on the trajectory I shall take.

Or, for the sake of time and my own sanity, I can approximate it as a 2D planar problem and get an answer quite sufficient to my needs.

I would make the argument that in professional motorsport engineering you are much more up against the clock and needing to provide someone with a quick and simple answer than one which is of 99.99% accuracy.

Point I've been trying to make all along is that there is a level of precision you need, above which is a waste of time. Or as Mr Cobb points out, if your process already involves a K&C rig or ADAMS model with elastokinematics, the whole process of working out all this stuff in your head is a waste of time. If you have any sort of multibody analysis then you get straight to the answer. Or any kinematic solver really. I've never, ever heard any race engineers sitting and debating the pros and cons of 2d vs 3d analytical solutions because it's all a waste of time and already wrapped into your analysis package.

Leave the programming to the programmers and the physical modelling to MSC. The team engineers have a list of 1000 items to do before a race, of which they might accomplish 10. No time to waste.

Quote:

---

2. If the Bible of RCVD

---

Not at all a knock on RCVD, but it is not the end-all-be-all of racecar engineering.