Vehicle Dynamics starting points and design process

Quote:

---

Originally Posted by ChristianChalliner

A big killer for us is that are engine isn't dry sumped and the sump on the CBR is ridiculously tall, we've talked about cutting the pan and making it wider but shallower and baffling or gating it to prevent starvation but whether we get that done this year is another thing. If you're wondering why not just dry sump it's because there's a distrust of dry sump systems from the academia for some reason so that's a no go.

---

I do understand that partly... When I joined my team in 2007, we just switched to the PC37 engine. The same you are using. And yes, we too found out that the oilpan is extremely tall, and yes, there was a distrust in dry sumps too ("Team XYZ tried it and killed 6 engines, we cannot afford that!"). Dry sump is a lot of work, some thinking, extra cost, extra weight, yes. But if you do it right, you will have your engine (heaviest component of the vehicle besides the driver) at least 10, 12 cm lower. And dead reliable. And because the cylinder head is so much lower, you can put the seat further back. And the engine itself, more "under" the diff while keeping the same minimum chain length. And you have some freedom where to put the extra weight (oil tank).

Guess what, I was the "dry sump guy" these days, who designed the system. ;)
The car in the picture further up this page still used an evolution of it...
Your engine/oil guy should send me a PM. I will certainly not give a detailed "recipe" but I will tell him if he is on the right track, concerning sizing of scavenge pump, lines and so on.

Using the PC37 with stock sump is a bad joke. And cutting the sump, making it wider... CAN work, if you do it right (rotary oil pickup), but is more work and not as "safe" as a good dry sump system.

P.S.: Pictures in this forum, try it with the [ img ] [ / img ] tags (without the spaces) ;)

Quote:

---

Originally Posted by ChristianChalliner

Mitchell- When you're saying you gained some un-sprung weight how much was it that you actually gained? and how much do you think the overall weight changed by due to switching to the beam? A big killer for us is that are engine isn't dry sumped and the sump on the CBR is ridiculously tall, we've talked about cutting the pan and making it wider but shallower and baffling or gating it to prevent starvation but whether we get that done this year is another thing. If you're wondering why not just dry sump it's because there's a distrust of dry sump systems from the academia for some reason so that's a no go.

---

Change in unsprung due to the beam was <2kg. The chassis was reduced to the rules required roll hoop support and roll hoop support structure. There is no rear bulkhead, diff mounts straight to engine.

You can get an idea of how much weight was taken out of the chassis by this picture.
Attachment 400

Unfortunate about the dry sump. To be fair the ones that seem to fail are the ones where people try and push them to the limit. The oil system we have was designed fairly conservatively. Oil supply is something that is not worth trying to squeeze the last 5% out of, we dont have a large budget and engines are expensive.

Christian,

Most important first.

As Jan said, you can display the image on the Forum-screen by taking its URL-link (usually starts with "http" and might end with ".png" or ".jpg", etc.), and typing [eye-emm-gee] directly in front of it, and [/eye-emm-gee] directly behind it, with "eye-emm-gee" = "IMG". BINGO! The picture appears in the middle of your post.

This is an example of one of those things that you do not have to UNDERSTAND (I certainly don't, and will never try to), but it can be useful to KNOW a little of it. It is a not-entirely-standard standard that is used on many forums. Sometimes you MUST use upper-case for the codes inside the square-brackets, other times lower-case works. Usually excess spaces create havoc. Other useful codes are; I (italics), B (bold), URL (link), and QUOTE (fancy quote).

And I really miss the :mad: (mad) and :rolleyes: (rolleyes) emoticons on the older forum...
~o0o~

Back to VD.

I agree with Matt and Mitchell about moving the driver as far back as possible. Also agree with Jan that you should think about dry-sumping to lower the engine, which then lets the driver move even further rearwards. The dry-sumping is only difficult when you have NO IDEA of what to do. But Jan has said he can give your Team some guidance..., which is a good offer!

The IMPORTANT POINT here is that this big-picture issue of the distribution of the major masses of the car (ie. driver + engine, wrt 4 wheels) has a much greater influence on a car's VD than any of the other issues that most students dive straight into. For example, fine details of suspension kinematics, optimised frame torsional stiffness, shaving another n-grams of the brake-discs, etc., really amount to squat, compared with that major masses distribution.

For example, you have suggested starting with track-width = 1.1m. That would be quite good IF you have a lowish CG. But with a high-CG engine, maybe not? At Oz-2013 one of the four-cylinder cars (bad memory... UQ, Mitchell?) had a narrowish track and unusually high "RC"s, and spent much of its time up on two wheels through the corners. This was not conducive to driver confidence, so not the best lap times...

A simple fix to the above problem is to widen the track, perhaps just by fitting wheel spacers. Or lower the "RC"s (ie. lower n-line slopes). Or both. But this needs a car built and tested early, and a preparedness to make major changes just before comp.

Better to get the big issues right first.
~o0o~

The "chalk lines on the floor" approach is great. Well, here "tape lines", and conveniently on a floor with 1' square tiles! Keep doing this and more like it.

I suggest a 8' x 4' sheet of plywood about 15 - 20 mm thick that you can build up on. Also buy some lengths of softwood as used in house construction (eg. 40 x 90 DAR), and perhaps more plywood about 5 - 10 mm thick. Make plywood wheels, engine, seat, radiator, etc. You can do a "first draft" of these using your cardboard and 2" wide masking tape (works better than duct tape).

Get as many of the newbies involved in this as you can. This might be the best "knowledge transfer" they get. Also the above plywood "~ergo-rig" can be reused every year. Here is a link to another of my recent posts on this sort of old-school CAD (Cardboard?-Aided-design). Also useful stuff on the first page of that thread.
~o0o~

"I played around with a direct acting arrangement and ended up with a fair bit more shock travel than I was actually expecting ...
... closer to 30mm which isn't so bad given that the shocks useable travel is only 40mm."

With only 40 mm damper travel, a MR = 1 (the experts' recommendation!) would have you failing Scrutineering! Ouch!!!

Having the DASD at 45 degrees to the wishbone gives MR = ~0.7 (=1/sqrt(2)), and ~56 mm wheel travel, which is only just legal.

DASD at 30 degrees (ie. very "laid down") gives MR = 0.5, and ~80 mm wheel travel. Good for a bumpy test track.

"On second thoughts that means 1/4 of the shock travel isn't being used... hmm."

As I have noted before, FSAE cars have won competitions with NO shock travel at all.
~o0o~

"... comments from the chassis guy:
...
The front roll hoop cannot be narrower otherwise we will fail the template test..."

I suggested narrowing the TOP of the FRH. Keep roughly the same width at upper SIS level, but bevel the corners off the excessively wide and square "shoulders" of the FRH. The current (draft sketched) FRH is simply following the current fashion. It has excessive mass up high, and the wide "shoulders" may (?) interfere with driver sightlines.

"... I don't understand what you are meaning by four bulkheads at most, I count a minimum of 5: ... fore wishbone mounts ..."

At this stage there should be NO ASSUMPTIONS at all regarding the type of suspension. The car may not have any wishbones! And if it does, then the front wishbone arms might attach to the front bulkhead (which would make a lot of sense!).

"... Also, the flat floor isn't possible because of where Christian currently has the hard-points which would result in the driver sitting too high up."

Again, this is "design by unjustified biases, prejudices, and preconceived notions...".

To restress all of this again, AT THIS STAGE YOU SHOULD ONLY BE THINKING ABOUT THE RELATIVE POSITIONS OF THE MAJOR MASSES (ie. driver, engine, wheels). All the other little details come later.

As such, when I said "flat-floor" I considered it obvious that this would lie under the driver and engine, being perhaps 30 mm above ground (to underside of floor). Suspension pick-up points are TRIVIAL LITTLE DETAILS. Sure, you will need them eventually. But you will also need an On/Off-Switch for the engine, and a whole lot of other little things.

Get the big-picture right first.

Z

Jan - Thank you for the offer :) We have a member who can possibly work on the dry sumping so if it's taken up then he will certainly be in contact with you with regards to his ideas. I completely understand where you're coming from in allowing the driver to sit further back, it's something that we've been fighting with! Currently, if we push the seat back the driver has to sit more upright which reduces the frame length and overall weight but moves the CofG higher, but how much higher? we don't know yet... it needs further analysis.

Mitchell - Looks good, one thing I don't fully understand from your pictures so far is how your mono-shock is actually actuated, I see you've got 4 links taking your beam to the main hoop and then the roll axis is where that bearing sits in the center of the chassis but how does the beam rotate upwards to actuate the shock if it's held on its axis by the bearing?

Z - I did try IMG tags initially but they didn't seem to work for some reason, it's a problem specific to this forum, I have no problems embedding photos on other forums.

I can see where you're coming from and it's something that has been mentioned before, if OptimumG's point mass simulation is within 10% and (to some extent) disregards suspension kinematics, assumes 50/50 WD, no wheelbase, etc then it stands true to reason that it's the big picture items which make the difference not the fine tuning of if you're running 1% or 5% of anti-dive. Before someone jumps down my throat I'M NOT SAYING they don't matter, just that other things are more important!

The problem is Z that I actually have to have some deliverables for my project so these issues must be focused on at some point and I also have check-ups at certain points over the year so I must do something in regards to these items early on to prove I'm actually doing something. I know, broken education system... etc, but if I go in and say well actually all I've done is said lets squash the car as much as possible and push the weight all the way to the back then that conversation only lasts so long...

I did also consider that yes the CofG is quite high but as you say by lowering the RC's this removes the lifting wheels problem but how low is sufficient? Again, analysis incomplete. The plan is to have a finished car by the end of February which should give sufficient testing, it also appears to be the case that this car and deadline is really being pushed by just 3 of us but if that's how it has to be then so be it.

It's good that you mention making a wooden rig as that's the plan for next week, it was the chassis guy that pushed for us to do a full 3D mock-up after we taped out the 2D one so we'll do that.

With regards to the DASD location, on my posted images the upper mount is inline with the upper wishbone mounts, as it turns out this actually gives quite good triangulation between all the wishbone mounts and the DASD mount so I think I'll keep with it as long as is feasible.

I wont move to Lancaster Links because I don't fully understand them and how we would integrate them into the chassis but what I am considering is the beam axle at the rear as it shortens the chassis dramatically and resolves the rear hard-points issue.

The chassis step is my fault but the intention was for it to be flat initially (and 30mm above the ground so we agree on something! :p), the decision to step it was driven by my desire to have the lower wishbone parallel with the ground, initially I thought this would give lower arm loads but now I'm not so sure, again, analysis incomplete!

We also had this idea for a pedalbox...

http://i1296.photobucket.com/albums/...psfhwpd73b.jpg

Which removes the excessive length of this current one

http://i1296.photobucket.com/albums/...ps2ikcrkgd.jpg

We also have an idea for the front of the chassis which we are drawing up currently so I'll upload that once it is done.

Quote:

---

Originally Posted by ChristianChalliner

Jan - Thank you for the offer :) We have a member who can possibly work on the dry sumping so if it's taken up then he will certainly be in contact with you with regards to his ideas.

---

No problem, do that! :)

Quote:

---

Originally Posted by ChristianChalliner

We also had this idea for a pedalbox...

---

This is a possibility certainly (and a certain team I happen to know very good did it just like that in the past), but keep the templates in mind... Depending on how long the assembly under the driver's feet is, the template would have to go over it, which means making your frame higher... Also, don't forget the fluid containers, which have to go somewhere ;)
Or think different. Placing the cylinders on both side of the pedal, but vertically, maybe...? Works, with some linkage... Just a thought :)

By the way, looking at your last picture, try to make the whole thing look a bit neater when it's finished. That doesn't make your car faster. But judges will like it. And one day, after you finished your thesis and maybe find some work in a small company that makes race (in my case rally) cars, you will find out that customers like things that look neat and tidy... Our Group R4 and N cars wouldn't need the carbon fiber and anodized alu "bling bling parts" to be fast. But they sell better with them. And you try to "sell" your car to the judges... Just my 2 cents.

Christian,

Ah, good ... now I don't have to log-in just to look at pics!

"Mitchell...
... how does the beam rotate upwards to actuate the shock if it's held on its axis by the bearing?"

Mitchell's Team uses a "peg-in-slot" for the lateral location. The peg is (I guess from M's post) a needle bearing "cam-follower", and it rolls in a vertical slot, which thus lets the beam move in Heave. A similar system is shown on this "Twin Beam-Wing" post on the "Beam-Axles..." thread. And similar systems were also used on the Mercedes-Benz "Silver Arrow" GP cars of the 1930s, and many others...

I am currently preparing a few more sketches for that "Beam-Axles..." thread that you may be interested in, should you decide to use a rear beam (err..., ready maybe next week...). You might also try looking up ECU's current car (ie. West-Oz team supervised by Kevin Hayward). I think they were at FSUK-14 (memory???) and are running a particularly neat bespoke four with rear-beam-axle. They use the head/crank/pistons/etc. of a sportsbike-four, but have made their own engine-cases that integrate a 2-speed(?) gearbox and final-drive/diff into a very compact package.
~o0o~

The main reason that everything should be pushed rearwards has to do with F:R% and "putting the power down" (as Matt described it). With a reasonably powerful four, and around 50%R, the car simply lights up the rear tyres coming out of slowish corners. All that power amounts to nought, because not enough weight on the driving wheels! This also make fast laps harder because it is so easy to spin the car on corner exits.

As I have ranted about for more than a decade now :(, 60%R allows the car to accelerate much faster out of the slower corners. It also makes the car much easier to drive fast. And, as a bonus, you get to win the Acceleration event! :)

But point-mass simulations don't explain this. They simply allow the car to accelerate at whatever rate you have assumed, which is at the edge of whatever "friction ellipse" you have input. And typical bike-engine layouts make 60%R hard to achieve, because big gap between driver's back and forward leaning cylinders, then another gap between output-sprocket and diff.

So most Teams settle for around 50% F:R (the easiest to achieve), and then argue that they have "perfect balance"!
~o0o~

"We also had this idea for a pedalbox..."

Yes, that can work. The main thing is to try to get the front-bulkhead as rearward as possible. So Percy's ~900 mm long legs (check rules?), plus maybe ~100 mm for the pedals, then the front-bulkhead.

As noted by Jan, the MCs-under-floor makes the frame a bit higher, but many cars also run very high footboxes ("Hey, everyone in F1 does it, so it must be good...").

Here is how I would do Better Brakes. Also more suggestions on the rest of that thread, such as Tokyo Denki's on page 3.
~o0o~

Oh, and don't get too stressed now, with all these different ideas floating around your head. THIS IS THE FUN BIT! :)

Z

Quote:

---

Originally Posted by ChristianChalliner

Jan - Thank you for the offer :) We have a member who can possibly work on the dry sumping so if it's taken up then he will certainly be in contact with you with regards to his ideas. I completely understand where you're coming from in allowing the driver to sit further back, it's something that we've been fighting with! Currently, if we push the seat back the driver has to sit more upright which reduces the frame length and overall weight but moves the CofG higher, but how much higher? we don't know yet... it needs further analysis.

The problem is Z that I actually have to have some deliverables for my project so these issues must be focused on at some point and I also have check-ups at certain points over the year so I must do something in regards to these items early on to prove I'm actually doing something. I know, broken education system... etc, but if I go in and say well actually all I've done is said lets squash the car as much as possible and push the weight all the way to the back then that conversation only lasts so long...

---

Christian,

NASA (USA's Space group) put together mass/inertia data for each major component of a human being. They published this information (I do not recall where, one of my team members found it in 2013) somewhere online, and that allowed us to look at seat back angle vs. CG height vs. Yaw Inertia for various seat back angles. This analysis was done with a nice simple parallel axis theorem, and was accurate enough for us to make high level decisions with. We then refined the angle when we had more parts placed in the frame and had a better idea of how much space we had/needed.

As for the deliverables, you may be able to point out that there is nobody in your list that is focusing on the big picture VD for your team, and that as such, you are working on that. If that doesn't work, maybe link whoever you report to to Geoff's "Reasoning" thread and have them read that?

-Matt

Quote:

---

Originally Posted by ChristianChalliner;121709Mitchell - Looks good, one thing I don't fully understand from your pictures so far is how your mono-shock is actually actuated, I see you've got 4 links taking your beam to the main hoop and then the roll axis is where that bearing sits in the center of the chassis but how does the beam rotate upwards to actuate the shock if it's held on its axis by the bearing?

We also had this idea for a pedalbox...

[img

http://i1296.photobucket.com/albums/ag13/challicm/EDB2D855-7CB8-4FAE-905E-0BCFDD61A2CB_zpsfhwpd73b.jpg[/img]

---

As Z has mentioned, the bearing fixed to the chassis, and rolls in vertical machined aluminium slot bolted to the beam. This allows the beam to be unrestrained in vertical movement and roll. The monoshock due to the mode separated suspension and isn't particularly important to the constraints on the beam. The same link system would work with a shock at each side of the chassis. Actually it makes it very easy to get the MR you desire by the position of the shock mount on the beam.

Unless you move to hinge type masters that is a great way of packaging your pedal box. If you mount the whole assembly on rails even better.

I will be very suprised if you have a complete car by Feb. We design freeze end of March each year for a complete car in July. This year it blew out til mid september, mainly due to learning some new manufacuring methods and a complete new vehicle concept. Our competition is in December.

On the subject of sumps: a well baffled custom wet-sump can be around 40-50mm thick and doesn't require you to package scavenge pumps and separate oil tanks/lines etc. I know anecdotes are a bit silly but here's one anyway: UoW (CBR600F4, now CBR600RR) went to a dry sump system for 2008. Unlike some people's experience with a first attempt at a dry sump system it worked very well, but when I took over powertrain in 2009 I reverted back to a wet sump. This was partly because I was a bit bewildered by the amount of work ahead of me and partly because I/we couldn't justify keeping the dry sump (higher part count, heavier, extra load on engine to drive pump, packaging issues (significant given our intercooled turbo setup)) over the wet sump (simple and proven concept over many years, albeit with a slight CG hit). Anyway, I guess I'm just saying that the standard sump must go, but that doesn't mean going all the way to a dry sump system (which will take someone fairly dedicated to implement).

You could always take your mock up and set it up on two scales and have the driver adjust him/her-self to different seating angles. This will allow you to see in real time the mass distribution change of each style of seating. Mostly what you will see is that as you lay your driver further and further down your weight shifts forward because you can't really move back with an engine in the way. Your driver's cg will also lower considerably. I think for 2014 we were running a seatback angle of 28 degrees from the horizontal (Yes, the ground!). We were certainly the smallest car at the two US competitions, but paid dearly in length due to Percy's massive legs. Just a heads up.


Z, on your point about the front hoop design of most FSAE cars, this odd shape is typically done to pass the vertical template test as there typically isn't an intermediate connection between the roll hoops, so to keep the main hoop from being excessively wide, the front is made wider which drives the top being wide due to the tubes required to attach to them for side impact tubes and the exceptionally large horizontal template that must be passed as well. As much as many teams would like to reduce the size of the front hoops, it's mostly driven by the current template rules.

For some other brake stuff, check out some of my older brake posts, I think I dropped some solid basic info in there.

For the weight balance, it's been argued many times, but there are some other considerations other than "I've been arguing this for decades and no one listens" is that the FSAE market for tires and available sizes is noticeably limited compared to either go kart manufacturers selections or much larger tire sizes. On top of that, the manufacturers and compounds don't mix and match very well throughout the series, so you're pretty much stuck with a few options, most of which have too much mass to try to dump energy into them. Want to use my favorite commercially available compound? Well, it comes in one size, and one size only. Want to move up to the next size? Then it'll be like driving on ice cubes as the tires will never heat up. So, it's not quite like the 80's F1 days where anyone really has the power and weight to put 70% of the weight (or more) on the rear axle and still have tire compounds (and suspension) that will both match correctly in cornering stiffness and roll stiffness with that bias. So something near ~45% Front become a nice happy target to shoot for with equally matched tires all around. Just throwing that out there for everyone, too.

With regard to the high roll centers, you'll want higher roll centers for FSAE than you would want for anything else. How often do circuit cars have to turn there wheels more than 30 degrees? FSAE you might get a few successive turns like that. Too low and your car won't react fast enough, throwing away responsiveness, making it slow. mildly high roll centers in FSAE is like having your cake and eating it too. The soft springs help let the chassis lower the sprung mass during cornering by being softer even though jacking forces are trying to push the car up. So, get it right and the car will play predictably in driving as well.