Reasoning your way through the FSAE design process

[Matthew Bell]

Cheers,

Geoff


Geoff Pearson
RMIT FSAE 03-06

>>>Design it. Build it. Break it. <<<

My computer's gone stupid and won't let me post with anything other than the quick reply, so the quote and bold doesn't work. But the point I want to make is that this tag-line is pretty much my entire engineering upbringing. I got lucky and worked in a machine shop with a bunch of old guys that knew more about engineering than most of the engineers I had met up to that point. I think there's a lot of that physical know-how that's gotten lost somewhere between generations.
Geoff, I will most certainly keep you posted on our progress. I made a presentation with another team member on behalf of our team to the university director the other day, and we're waiting for approval for our proposal. I can't thank you enough for the prompt response with the case study, it definitely clears up the few questions I had about specifics. I'm working on putting together a legitimate document with all of this thread's information, but unfortunately I won't be able to post it here.
What I can tell you is that we're stuck in the level 1 to 2 design loop just like many of the FSAE teams. Where we're lucky, though, is that we have 14 people in one room all day long dedicated to the exact same project with little outside distraction. So it's kinda like every FSAE team's wet-dream. On top of the people designing though, we're in an extremely unique situation to be in the endurance racing capital of the world. Therefore, through university contacts, we get to see and do a lot of really really really neat stuff. Again, I can't give you specifics, but let's just say I've been here six weeks and have had my face melted at least twice already just from things that I thought would be "impossible" for me to experience.
So, we're trying to break out of the level 1 design trap, and the main force driving this is the fact that we have more than one adviser. It's more like a committee. We have an academic director, a industrial automotive guy, a fabrication specialist, and a person experienced in customer service and quick action.
Now we just have to all get along...

I will keep you updated as much as I can. We got to attend the French OptimumG seminar a couple of weeks ago, and my brain still hurts from that one. Any questions you might have, I am glad to answer, but probably only through PM.

[EHog]

This really needs to be stickied.

[js10coastr]

Originally posted by Big Bird:
I've always thought endurance racing to be the ultimate goal for a motorsports engineer, it seems to have a lot more scope for ingenuity than the more controlled forms of motorsport.

I grew up watching indy cars, but am now working in the ALMS and I wouldn't have it any other way. The challenges are so much more diverse... you have to take into account much more in terms of the team and how everyone functions. Also, the pit stops and strategy are much more important... so as a whole the engineer is a lot more involved and has more control.

[js10coastr]

Originally posted by Matthew Bell:

Does anybody have any "success stories" they can share? Or even better, "failure stories"?

Background, circa Fall 2007... I had just started the MBA program at Cal Poly SLO; and had been involved in SAE/FSAE during my undergrad(2000-2005)... enough to get an internship at OptimumG (2005-2006). The current formula team didn't interest me (they were quite the counter example actually), and there was a large amount of money available for a formula hybrid team through plug in america.

I started up the Formula Hybrid team in January of 2008 with the intent of making it to the competition in May of 2008 and winning overall in 2009. However this proved to be too ambitious (even when using a donor chassis from FSAE)... the people of the team had yet to be developed, too short of a timeline, and way too many compromises/shortcuts would have to have been undertaken to make it to competition. At the same time though, I started working on a simulation program in MATLAB.

Looking at the team, we had one EE… the rules also stipulated that a first year team can enter a “hybrid in progress”, which is essentially an electric car. We intended to and did take full advantage of that rule. While it wasn’t as easy as “red to red, black to black” we did our best to make a simple electric system with an off the shelf controller and a really good motor that would pass the rules… having one EE on a hybrid team is a limiting factor.

However, having a large number of mechanical engineers would be our forte. Looking at the photos from the previous competitions, a lot of cars were not mechanically sound and we felt that we could build a better car as a whole, because with one EE there was no way we could beat them with electrons.

Early on, the Formula Hybrid team chose to make a steel tube frame chassis, while the Formula team chose to continue with a carbon chassis. The reasoning behind the steel chassis was that we were a new team with limited experience in composites, if you need to build the frame quicker you can just add more people or more shifts, and if you need to make a change in the frame you can just hack off something or weld a bracket on. To do a carbon chassis properly, you really do need to have EVERYTHING designed and in place before you start cutting foam, and if you get behind schedule, you can't really speed up the process because things have to happen in certain order. Had the Formula Hybrid team been more experienced, with a car layout that had been fairly stable over a few years and a fully populated solidmodel, the choice would have been different.

The simulation was used early on to figure out what was the most important…when looking at the competition and simulating their cars we figured that by bringing a significantly lighter car we could get away with a much smaller motor and still be competitive. Our quest for a lightweight car took a minimalist approach. Regen braking? Too heavy and too complex… transmission? Well, can we get away with just one gear ratio? Yes, but let’s use a split sprocket so that we can change gears easily during testing. Torsen diff? We have one left over… what about a spool? It would be lighter… proably easier to integrate with the axles we got from the baja team… is the compromise in handling worth it? It sounds like if we set the car up right/design the car with a spool in the first place we’ll be just fine. And then on to the most obvious part of the car… our wheels. We also made the decision to buy our pedal assembly… would we gain anything from making our own? Maybe a little weight… what would we lose? Cost wasn’t too big of a difference, there really wasn’t much to learn in the manufacturing of the part, maybe a little bit in the integration… but we’d gain man hours that could be spent elsewhere.

Having spent the summer researching other cars, we came across the idea of using bicycle wheels and tires to reduce the amount of unsprung and rotating weight… after some experimenting we decided that those would not provide enough lateral force/cornering stiffness/the wheels couldn’t hold the forces. However… somehow we came across “racing scooters”. The fact that people race them and the fact that there is enough of a market to make racing scooter tires is beyond me, but… here we are. Having spent a year working for Claude I felt that I had a pretty good understanding of suspension kinematics and its importance (cue Claude’s “I will never forget….”). After a few emails with Doug Milliken and while referencing his dad’s biography, I was able to reverse engineer the suspension on the MX-1, the original camber car. The reasoning and justification for the camber car made a lot of sense for our application… a huge reduction in unsprung and rotating weight, which saves us a lot in terms of energy use, while increasing our acceleration potential… secondarily (and of larger importance for the MX-1), the frontal area drops.

Cal Poly SLO has the best machine facilities available for its students in the state (if not the nation), including numerous CNC machines… however, use of CNC was restricted to the spline on our spindles and bearing cups (or “wafers” if you will) on the suspension components. While the end part may end up being slightly heavier, I felt that the learning curve and time required for CNC setup wasn’t justifiable for the majority of our parts.

We set out to have our frame manufactured before we all left for Christmas break… this gave us enough of a start/platform to work with to get up and running to have a few weekends of testing/finding the bugs. Building the car went smoothly with one or two exceptions… we kept track of our competition through their websites, and set our targets on McGill (returning champions), and Texas A&M (newcomers to hybrid but super successful at FSAE West). Amazingly we only had two or three nights where any of us worked past midnight… “Truth in 24” had just come out and we rallied behind that movie, following the Audi philosophy of executing as a whole team, and engineering outside of just the parts on the car. We put the movie up on the projector during one of our late nights, its really amazing what a little inspiration can do.

Speaking of inspiration… I’ve had chats with the great Kevin Hayward along with Matt Giraffa about the car construction. One of the points that had been made was that the outlook along with the performance of the team goes up when you have a rolling chassis. And so, even if its quicker on paper or from a pure engineering managerial point of view to bring the components together later… the human aspect of seeing a rolling chassis does wonders for the team and from this point of view it makes sense to get the car rolling as soon as possible.

With a simple and cheap data acquisition system (about $1000 from AIM) we were able to optimize our gear ratio while monitoring the motor’s health. I felt that having a simple data system was important and because of my experience at OptimumG, I would be able to extract more information than other teams would while combining that with the simulation software. This feedback loop would be one of our strengths.

Competition was tough for us… shipping the car across the continent was not an easy task, neither was getting everyone over there. The move cost us as we weren’t able to compete in the first dynamic events (accelerations and autocross)… and I was pissed because we would have won the electric only acceleration, shocking the “good luck next year” and other nay sayers.

Even with missing the first dynamic events, we ran all of endurance, finishing second to Texas… which eventually put us atop of the “hybrid in progress” category.

If you'd like to see the car in action just search on youtube for Cal Poly SLO Formula Hybrid.

[Big Bird]

Hi all,

Thanks for the above contributions, unfortunately I haven't had much time to contribute to these boards of late. I'll get onto that later, but in the meantime a few more words while I've got a few minutes over the festive season. And season's greetings to all, I trust you're all enjoying the break.

Selection vs collection of objectives
or "OR" vs "AND"
or Understanding priorities

I’ll start with a common scenario I’ve observed as an incoming team approaches the early stages of the design process. The question is asked “What are we going to do this year?”, and some conversation ensues about design objectives. Maybe some team members have been scouring the forums, some have been speaking to alumni, or some have been involved in previous teams or attended last years comp. Maybe the faculty advisor has chimed in about the university’s expectations, or projected budget and resources, or last year’s team effort.

So after maybe a minute’s / day’s / week’s / month’s discussion, the team learns that the following attributes are important in FSAE:

The car should be:
• Cheap
• Light
• Powerful
• Economical
• Simple
• Strong / reliable
• Stiff
• Adjustable
• Comfortable (reasonably…)
• Compact
• Easy to drive
• Predictable
• Easily manufacturable
• Built to tight tolerances
• Well tested and developed
The project needs to be delivered:
• On time
• Within budget
• Within the capabilities and skill set of the team
The drivers need to be well trained
The presenters need to be well prepared and rehearsed for static events
Etc
Etc
Etc
……..
…….
You can probably add in any number of detail requirements passed on by alumni, design judges etc about roll centres, plenum volumes, tyre selection, weight distributions, injector locations, load paths, types of bearings to use, upright materials, etc etc etc.

The new team is now well informed, has “learnt” some of the key FSAE lessons - and is utterly swimming in information.

The crucial next step is what to do with all this info, and this is the point where I have observed a far too common, critical error in the team’s planning.

The new team “ANDs” all the information together

Rather than select which objectives are most important, the team decides that they are just going to do it all. The team’s mission thus becomes this mish-mash of self-contradicting objectives:
“This year, our car is going to be lighter, AND more powerful, AND more economical, AND cheaper, AND simpler, AND stiffer, AND easier to drive, AND easier to build, AND built to tighter tolerances, AND we are going to get the car done earlier, AND we will do more testing…”

This is the “everything is a priority” school of FSAE thought, and I believe the key reason we continue to see falling finish rates at FSAE events. The new team believes they are somehow the competition’s new “golden children”, and that they have some new insight or level of commitment that was so obviously lacking in those who came before them. They will be the ones who are going to do it all. So the first 1-2 months are spent dreaming up the design solution that is all things to all people, and the next 3-6 months overdesigning it. Another 3-6 months is spent trying to manufacture something way beyond the limits of their budget / material resources / people skills.

Around 1-2 months before comp it might be recognized that time is running out and the team is not going to do it all – so critical objectives start becoming casualties almost by default. Human nature dictates we would rather find consolation that admit to fault - so the failed objectives are declared irrelevant:
Timeline? “It is ridiculous to think you can finish these cars three months before comp, and besides, last year’s team were only two weeks ahead of us at this time of year…”
Testing time? “We won’t need testing time because our design/build quality is so good that we can afford to finish the car the week before comp…”.

So comp arrives, team shows up with a car is some state of disassembly and incompleteness, three of the drivers have never sat in the car before, team doesn't make it through endurance / skid pad / brake test / team sign-in. The team goes home with assorted DNF's or poor results, but some great new ideas of how next year’s car can be lighter, stiffer, more powerful...

This isn’t your team? Your team is better than this and would never just pile up all those objectives, especially given the inherent contradictions? Then turn things around and do a simple test. See what happens when you tell your team that you are consciously trading off one of the sacred objectives. “This year’s uprights are going to be stiffer, and easier to manufacture, but a bit heavier”. I reckon in nine teams out of ten, you’d be crucified. At best you will be allowed to hold your mass target to last year’s value, and guaranteed next year’s upright designer will announce your design to be an embarrassment to the team and that his design is so much lighter, and it is also stiffer, easier to manufacture…..

An even deeper malaise, and we see this one at a much wider level in our society, is that of spin. To achieve one objective, we have to trade-off off on another – but we refuse to admit it. “Our upright is so much better because it is stiffer, cheaper and easier to manufacture”. What about weight? “Thank you for asking that question, it is a very good one. Your ignorance is rather endearing, so let me address this issue by explaining to you again how our upright is significantly stiffer, cheaper and easier to manufacture…” Society is obsessed with the “win-win”, but the world doesn’t work like that. So we resort to word games, and designs that increase incrementally in performance and exponentially in cost, as our design process is pretty well limited to expensive detail optimization of existing designs.

FSAE design is full of contradictions, compromises and trade-offs. If you want a perfectly flat contact patch under braking, you have lousy camber angle in roll (and vice versa). The attribute of engine power usually comes at the expense of fuel usage. Light often plays against stiff. Tight tolerances come at the expense of manufacturing time. Etc etc.

Good management requires the ability to use the “OR” function. You need the confidence, knowledge and team support to make decisions as to which objectives are your priorities, and which you are willing to trade off and by how much. These are linked in to my above words about lap sims, understanding your design structure, and some objective analysis of previous results. But that will be the subject of a later post when I get the time. As for now, back to thesis writing…

Cheers all,

[PatClarke]

Geoff, You are a breath of fresh air!

Best wishes to you and yours

Pat

[Big Bird]

Thanks Pat, and I trust you are having a nice break and all as well. I've just spent an hour or so revisiting some of your Pats Corner musings - some interesting stuff there. I'll be recommending that our guys have a good look through those.

Also a quick disclaimer in case any is reading any of my writings and thinks I'm publically bagging out on my own team. I'm not. My opinions are based on just under ten years of observing, competing and then advising in the FSAE / FStudent arena. Certainly I've been involved in the RMIT team, and I also competed for Monash one year. But I also base my opinions on personal communications with a wide variety of FSAE community members: team members, alumni and/or faculty advisors (probably every team in Oz at some time, plus a variety of European, American and Japanese teams), organizing staff (both here and OS), design judges, sponsors and industry observers, spectators and general public. I've also observed nine Australian comps and three OS comps, and been an active member on these boards for just over seven years. And from my observations and discussions I honestly believe what I've written in this thread is applicable to a significant number of teams. These are common issues, and I feel are worthy of bringing to the board's attention. And I repeat and stress these observations are not focussed internally on my own team.

I just thought I'd better clarify that, as I know our guys have had a tough year this year, and might be feeling that I'm pointing criticism at them. I'm not, and sincerely apologize if that was thought to be the case.

Cheers all,

[Big Bird]

Before I start my random rant, a reading recommendation. I've just started on "The Multi-body Systems Approach to Vehicle Dynamics" by Blundell and Harty. The first chapter has some nice stuff about the merits of simple models, and also their own "V" process for simulation design. It is well worth a look.

Isolation vs Integration
This is one of my pet grievances about the way engineering is taught, and apologies if I’ve already made a meal of it in previous posts.

Engineering is primarily taught at university as a science. The principle of the scientific method is that you hold the world constant, isolate one parameter, change it, measure the results, pick most favourable result, repeat. I have no problem with the scientific method as it has done wonderful things for human progress. Just look at health sciences, nanotechnology, material sciences, imaging technologies, etc etc. But it is one tool, and just like a spanner you don’t try to adapt it to all jobs. It can also be a rather slow and expensive tool – how much money do F1 teams spend on wind tunnels (which are after all, big and fancy rooms designed to hold a small part of the world, the immediate surroundings of a vehicle, constant).

Engineering is primarily taught at university by scientists. To get into a lecturing position you need a PhD, which is effectively a 3-10 year apprenticeship in isolationist science. So the majority of our lecturers are brilliant people, but with specific expertise in maybe one or two fields. They will be able to tell you how to get the best downforce through a diffuser, or how to reduce the void count in your composite layup – but will be most likely flummoxed when you discuss how the diffuser is compromising your suspension geometry, or that you are happy to accept some slight imperfections as layup time is a greater priority.

It is worth noting here there is no surer way to get an expert singing his song at full volume than showing him a design where his pet specialty is not “optimized”.

Engineering is not a science. We use science, but it is just one of the tools we use. Engineering is practical problem solving, and in the real world it is often the norm when multiple things are interconnected and changing at once. We deal with it, scientists get all frustrated and apply for grants for even bigger experiments

The problem with studying specific attributes in isolation is that you lose awareness of the relationships between those attributes. Think about the following attributes (significant ones for an FSAE project), and whether they may be complimentary or contradictory:
Engine Power
Braking Power
Tyre friction coefficient
Component stiffness
Weight
Size
Fuel usage
Reliability
Manufacturability
Cost
Project completion time
Vehicle Acceleration capability
- Cornering
- Forward
- Braking

If you have been taking notice you’ll realize I’ve mixed level 1, 2 and 4 objectives there – no problem at this stage. I’ve probably missed a few attributes too, but just note that there is not one attribute in the list that has a complimentary relationship with every other attribute. Nor one that has a contradictory relationship with every other.

Isolation in FSAE
The classic conflict in Formula SAE is the power vs weight argument. Separately and in isolation of each other we may look at these two attributes and agree on two independent guiding principles.
1. Less weight is better
2. More power is better
So the team sets off to make the car lighter and the engine more powerful. Simple…

The trouble is, we have neglected the important relationship between power and mass. More power usually means more mass, as the greater forces require more material to resist those forces. So now we have a conflict – we want both, but they are working against each other. How do we choose?

So someone up the back of the room puts his hand up and says “we design for the best power to weight ratio”. We now have a more sophisticated argument, everyone cheers, and problem solved. In isolation, we decide on “best power:weight ratio” as our objective.

The team comes up with two designs. One is a 400kg car with 210hp, the other is a 100kg car with 50hp. Our “best power:weight ratio” criteria tells us the 400kg car must be best, right? Of course, the first time you try to turn this thing around a corner you’ll be thinking otherwise. If the course is mainly straights, the 400kg car might triumph – but in many cases the smaller car will be better overall.

So the guiding principles start becoming convoluted:
“We want a good power:weight ratio, but we don’t want weight going too high dependant on the percentage of corners to straights on the track”

Enter Fuel Economy, where fuel usage has a contradictory relationship with power, a complimentary relationship with low weight, and is also linked to track layout depending on the top speed and number of acceleration instances:
“We want a good power:weight ratio, but we don’t want weight going too high dependant on the percentage of corners to straights on the track, and we don’t want too much power as we start losing fuel points, which is sort of dependant on track layout too. Losing weight seems to have the most positives”

Enter durability / robustness:
“We want a good power:weight ratio, but we don’t want weight going too high dependant on the percentage of corners to straights on the track, and we don’t want too much power as we start losing fuel points, which is sort of dependant on track layout too. Losing weight seems to have the most positives,, but that puts us at risk of losing reliability”

Enter cost, driveability, driver comfort, vehicle size etc, and it all starts looking like a nightmare. So it starts looking all too hard and isolationism starts looking easier to deal with. “This is all bullship. We just need more power”.

Technical Leaders
Think of two Chief Engineers making their election speeches. CE1 is an isolationist. CE2 an integrationist.

CE1 stands in front of the team and says:
“We want most power. We want less weight. We will have the stiffest chassis and the most tyre grip. We will have the smallest and most agile car at the track. We will win fuel economy.
CE2:
“We sort of want good power, but not too much as it will drive up weight and probably fuel usage as well. Weight reduction is important, but we don’t want to go too far because it can effect reliability. I reckon weight reduction is around 3-4 times more important that power increase. Also too much focus on high power and low weight will absorb too much project time, and we need to cut back on that to get more testing time before comp. Sticky tyres are mostly a good thing, as long as they don’t absorb too much energy and therefore fuel. Small cars are great, but they need to be big enough for the driver to be comfortable. We’ll have to put limits on all our goals as we need to meet budget and timeline”….

CE1’s are good at winning confidence and support of the team. Their messages are simple, concise, easy to understand. They sound like they are under control. They rarely deliver a functional car.

CE2’s look like hand-wringing neurotic indecisive crackpots. Often, their discussions seem messy and poorly formed. People smile at them politely and try to ignore them at barbeques. A well studied CE2 can deliver a cracker of a car, with a nice balance of power, weight, reliability, driveability, etc etc.

I’ve said it elsewhere here but the trouble with balance is that when you finally find it, some expert will complain how you could have had more of something. The isolationists love this sort of argument. They see their pet attribute lacking in some way and take great pride in announcing how they would have done it better. Unfortunately such arguments win a lot of support because they sound so simple and convincing, especially to the ongoing stream of novice engineers coming into this project.

To be honest, I’ve been struggling with this for a number of years. Our original design was reasonably well balanced – the cars had enough power, enough “lightness”, enough stiffness etc, and enough simplicity that we could complete the car on time and on budget. We deliberately set a low engine power target, knowing that the points lost in straight line speed were linked to gains in fuel economy and cornering speed. By building a quite simple car we finished early, tested well and cashed in on the “incompleteness” of most of our competitors. We found a nice balance point where the relationships between quite a large number of parameters (e.g power, weight, fuel use, cost, completion time, reliability etc etc) were working well together.

Of course, with each new project the incoming isolationists would have a field day. “If we can win with that fat / oversize / slow / insert-isolated-attribute-here car, imagine what we’ll do when I give it more power / less weight / insert-isolated-attribute-improvement-here……..” Trying to convince a newbie to settle for a simple low power engine as it offers better fuel usage combined with lower mass and size leading to better cornering, combined with simpler manufacturing linking to more testing time, combined with improved parts availability due to high volume sales of the donor bike etc, will nine times out of ten invite a blank stare and the same response – “but more power will make us faster”

The critical attributes with large points impacts for this project – completion time, reliability, testing time etc – are rarely linked to the vehicle based attributes, which are often the less points-effective attributes. The argument goes along the lines of:
“We can lose weight by making the ***** out of carbon fibre”.
“How many points is it worth?”
“Every little bit counts mate, Stuttgart is 500grams lighter than us.”
“Reliability?”
“Nah, we’ll make sure we design it right.”
“Completion time?”
“Don’t worry about it, we’ll work harder if we have to.”
“Robustness?”
“That’s nonsense, this is a race car – it is meant to fall apart when you cross the finish line.”

The trick is knowing your design problem. What are the attributes that will bring your team the greatest improvement? Is it really more power? Or is it getting the car finished earlier? Can you trade one to buy the other? Are there other attributes that are more significant again? Only you can answer that, but you need to be honest with yourself.

You need to do two things to lead your design process:
1. Pick out the attributes most important to you
2. Rank and weight them in order of importance.

This can only be achieved by analysing the competition itself (level 3) and your own team resources and history (level 4). I've already spruiked the value of a simple lapsim for the competition analysis side of things.

A bottom up, level 1 driven design process invites oversimplified generalizations like “more power is better” and “we need this to be as stiff and light as possible”. It will deliver blown out budgets and timelines and an unbalanced vehicle. You will never fully understand why you are doing what you are doing.

A top down level 4 driven process will help you understand how to rank and prioritize competing objectives, as it will give you a detailed understanding of the design problem itself.

Enough for now, I’ve got a few more words coming about opportunities for lateral thinking but that can wait. Happy designing.

[ed_pratt]

Thanks again for these insights Geoff, I think a few of us here are hoping you'll collect your thoughts together into a book at some point!

I have been thinking about this for a little while now.

Originally posted by Big Bird:

The trouble is, we have neglected the important relationship between power and mass. More power usually means more mass, as the greater forces require more material to resist those forces. So now we have a conflict – we want both, but they are working against each other. How do we choose?

I suppose it's not directly related and I don't want to hijack this thread but I can not figure out how to go about including this as an argument in a LapSim.
At the earliest stage of design - level 4 I suppose - how can we work out the consequences of choosing a 4 cylinder over anything else i.e. how can you quantify the extra weight gained in the chassis/drivetrain/anywhere else by making that choice? Is that even possible without halfway designing both scenarios to see what you come up with?!

Sorry for the total brain dump - I'll try to collect my thoughts a little more next time but it's a friday morning : )

Ed

[Big Bird]

Hi Ed,

Thanks for the ongoing encouragement - and yes I have been thinking about a book. Or maybe a PhD thesis if someone out there thinks this could be extended into one (please contact me if you think so). Anyway it is Friday night here, so I'm hoping I can gather my thoughts given my brain has signed off for the week.

Good question, and the answer does require a bit of pre-assumed knowledge.

Firstly though, the question of power vs mass isn't really a level 4 question. L4 is all about establishing the overall framework in which you are working - how much money do we have, how much time, who is on the team, what skills do they have, what resources do we have for completing the project, who are our supporters, what are the overall team goals.

Level 3 is about understanding the competition, and a lapsim starts becoming useful here. A simple constant acceleration lapsim will give some good information about how sensitive the lap time is to changes in forward accel, braking decel, cornering accel. Change forward accel by 1%, see lap time change. Return forward accel to original value, change lateral by 1%, see lap time change, compare to change from forward accel. Repeat for braking. You should be able to establish relative sensitivities to different accels - "forward accel change gives 1.8 times the points return of braking", or whatever numbers you come up with.

If you are clever, you know velocities at every point, could make guesstimates of drag and frontal area, convert all this into energy required - then calculate fuel used via a thermo efficiency.

Now this level 3 stuff helps you understand the competition idiosyncracies. You should have a rough idea of acceleration goals (level 2 goals), and which accelerations you might like to focus upon. "Optimize", eh Pat?

The mass vs power argument sort of comes in just below the level 2 goals. You are wanting to set overall mass and power targets to get near the benchmark accelerations you have set.

I think it is rather easy to get a rough idea of what a good car weighs. Generally a good 600/4 is somewhere around the 200kg mark, and might be putting oout somewhere around 80hp. A 450 single motor weighs around 30kg less than a 600/4, puts out around 50hp, and you can make a rough guesstimate of what mass variance there is in driveshafts, tripod joints, engine mounts and structure, exhaust and intake etc.

Now take your earlier lapsim and try calculating your forward accel as a function of mass and power, (Sigma P = mav). You might also like to calculate your estimated lateral and longitudinal tyre grip (and therefore lateral and braking accels) as a function of vertical load, to capture how the mass might influence cornering speed, (you need TTC data for this, get a rough value for % change in friction coefficient for each Newton of vertical load).

Now, put your 600/4 values of mass and power into the sim - see lap times. Then punch in the mass and power of the 450 single, or whatever other option you are looking at, check times. Compare, observe interesting results. Just be honest with yourself about your input values - don't go just changing power values without modifying mass to suit.

I'd say that at this early stage your sim can be pretty low fidelity. The point of the exercise is that you are just wanting to get an idea whether one design is close or not to the other. 5-10 points difference? Negligible, either can win, decide on which is easier for you to build. 20-50 points difference? Becoming significant, but decision still dependant on what you can deliver easiest. Over 50 points? Getting uncompetitive. Over 200 points? Something very wrong with either your sim or your understanding of motor vehicle design, take up origami.

I hope all that makes sense. I'm getting weary, so if I haven't answered the question properly just say so and I'll have another go tomorrow.

Cheers,