The ever popular diet

[BStoney]

How much does your electrical system weigh without DAQ? If it is not less than 10 or so lbs with ECU and relay/fuse box and even battery...then you're probably haven't engineered it and it's just moderately pieced together. I see all too often teams with wires everywhere and using the wrong wire for the application/wrong components. (And yes, I've been there too...)

Use the Keep It Simple Stupid philosophy on most stuff, and you should be able to keep weight down. Also, set goals for each of your sub-systems for overall weight of that system and make sure you stick to them, and do what's necessary to meet them.

[VinceL]

Quote:

"Barring that... the easiest way to lose weight is go to 10" wheels. "

Remember that the reason for loosing weight in the first place is to maximize the value a=F/m. Going to 10" wheels reduces m, but also reduces F (take a look at the TTC data). So the performance gain from losing the weight will be far less than 1:1. Obviously 10" wheel cars can go fast (RMIT, ETS, Delft). What I'm saying is that I think you need to have a pretty light car to begin with to make 10" wheels work well. Just simply getting your car and switching to 10" wheels might not be the best idea. But then again I haven't done any sims of that scenario, so maybe you know better than me.

[murpia]

A lot of the above advice advocates a 'bottom up' approach to weight loss, going round all your car components and trying to save a gram of two with refinement. I often hear this kind of thing from Formula Student teams when I ask them about their design targets, typically something like 'last years car was too heavy so we told all the subsystem designers to save 10% of weight'. I also often hear 'we set a design target of 200kg but actually it weighs XXXkg' (insert embarassing amount, typically 10->20% over).

I'm here to tell you that's the wrong approach. Imagine if Boeing or Airbus offered to produce a new aircraft simply by slashing 10% off all the bits - would anyone buy it? No. No-one would know for sure how it would perform, or if it would even fly... What customers want is product performance, defined at the start of the design process then carried through by analysis to the detail stage and then by component, subsystem and system testing back up to verification of the entire product. This is often referred to as the 'System Engineering V'. It's not the only approach, but it's a good one.

Don't think that saying 'as light as possible, as stiff as possible and as low a CG as possible' when asked about design targets gets you off the hook. Firstly, I think that's worse than unambitious numbers, because it promotes incoherent design and doesn't drive the detailed designers hard enough. Secondly, more importantly, those factors trade off against each other and you need to make a decision as to where you stand. And finally, those aren't product performance targets, they are actually the first stage of analysis.

Following this kind of discipline reaps all kinds of benefits: done properly you'll know up front your weight, parts count, cost, CG and inertia values and you shouldn't have any embarrasing structural failures. Most importantly you'll have an excellent foundation in engineering skills for the real world, which is the point of the competition after all.

Regards,
Ian Murphy
FSUK Design Judge Team Leader

[Kurt Bilinski]

Murpia, I agree to some degree that it's not as simple as saying, "Make it 10% lighter than last year's car." However, just because it's a top-down approach doesn't guarantee success either.

What bugs me is where this ambitious weight loss value comes from - it's an emotional target that isn't based on engineering. Yes, we all need targets, but engineering should drive it, not someone's emotions. A clean-sheet design team leader deciding that since other cars are around 400lbs, his is going to be 360lbs, is bad news. Why *that* number*? Where did it come from? He then makes sure that every component is forced toward that weight target, which is fine from a Project Management standpoint, but where's the engineering? That is, is it practical, reasonable, and affordable to knock off XX% of a car's weight and set that as a goal? You are right that ALL the project's subsystems must work together, but there are limits. If the suspension engineer says he's already using the lightest (affordable) wheels, what would you have him do? If the roll-hoop is already of minimum size, where's this presumed XX% reduction coming from? Subsystems that cannot meet this arbitrary emotional goal will transfer their share of the required weight loss to some other hapless subsystem. I agree that the factors trade off against each other, but often to the detriment of the last guy in the door. Imagine being the last subsystem guy to get to work, discovering that because others couldn't achieve their goal, your required weight reduction is now 60%...

I guess my point is, I agree with you to a point, but once the subsystem engineers are all saying, "I can't reduce weight affordably and safely, and meet the requirement", you're done. Having the Project Engineer saying the car's going to be lighter is not good engineering - unless he knows where the weight's going to come from *before* making it a requirement. Yes, I know people need goals to work towards, but not knowing ahead of time what's achievable means that most of the time, the goal won't be met. It's not a failure; all it indicates is that the design is as good as it can get without changing the entire approach.

Personally, I think round-table discussions before the new design starts, bouncing ideas off one another - that's the only right way. This way, everyone's on board, knows why things are changing (not arbitrarly), knows the target, and is working hard to achieve it.

To me, weight loss is a consequence of good design, not a goal unto itself.

[Mike Cook]

533lb is not all that bad. Our first car was 620, our second was 530, third was 520, and forth was 466.

A few things to consider:
A lighter car will be faster (F=MA) but to a lot of teams, there are more important things than taking off 20 more lbs (consider the endurance finish rate). In my experience, it took a lot of analysis and time to go from 530 to 466. Compromises must be made and it isn't easy. The biggest thing a team can focus on is reliability, and that is certainly at odds with lowering weight. It is not a good idea to say we need to be 40lbs less weight, but rather, we are going to reduce the weight of parts we did analysis on, and our weight will be what it's going to be.

All that said, some suggestions:

1) If have no good justification for making your car larger than 60" WB, don't. A smaller car will be less weight.

2) Use appropriate size bolts.

3) Use a lightweight wheel, (6-7lb for a 13")

4) Don't over build your frame. 1"x028 exists. Our first frame was somewhere like 90lb, our most recent frame was about 55lb.

5) Size your wires correctly for the current they're carrying.

6) Fuel tanks can be made pretty light. .04" aluminum sheet. (You could go thinner, but I won't)

7) Seats can be made less than 5 lb.

8) Thin walled exhaust tubing and intake tubing.

9) Appropriate sized brake rotors.

10) Produce a clean car. Clean cars weigh less. In my experience, rushed cars always weigh 10-15lbs more than they should because people just throw shit on there to git r done.

11) Low weight body. There shouldn't be anything on the car that doesn't need to be there.

etc.etc.etc.
Make small steps each year. The teams that win are low weight, but they DONT win entirely because they have a lower weight. Rather, there low weight is an consequence of good engineering.

[MrSwa]

Reading all of this is giving me a ton of ideas. Keep it coming. Our frame weighs about 55lbs this year, so we're looking to really buckle down on the other subsystems. I feel a lot of computer time coming.... Oh well. There is a lot of weight we can lose. Now to prove it.

[flavorPacket]

the way we think about it is:

If Renault F1 teamed up with McLaren, Ferrari, and Audi Sport, and they built a FSAE car with the same resources and materials as us, how much would that car weigh?

We thought about it for a while, came up with a number, and that's our goal. It'll probably take us 2-3 years to get there, but there are ALWAYS places to take weight out.

[Bill_Murray]

Got to agree with mike (above) but wanted to add one thing:

When you a are searching for weight in parts that have hopefully already been correctly sized for the expected loads (right?), you can't usually cut xx% weight with the same concept. If you are going to cut the weight it requires a system design change. A change in the way parts or systems of parts carry their loads (direct load paths).

Mike's suggestions will take you from 520 to ~460. Then you will need to look at things on a systems level as Murpia is talking about and understand the interactions of these systems to get to 400 or less. This is where good team management comes in. Read some of the other posts on managment.

[murpia]

Originally posted by kb58:
What bugs me is where this ambitious weight loss value comes from - it's an emotional target that isn't based on engineering. Yes, we all need targets, but engineering should drive it, not someone's emotions...
To me, weight loss is a consequence of good design, not a goal unto itself.

Maybe I wasn't clear enough, but I'm not advocating a weight target I'm advocating a set of perfomance targets for the completed design.

How much it costs and how long it takes to complete the project will depend on how aggressive the performance targets are. However, you will find that once you analyse the engineering principles you need to follow to achieve your performance, the total mass is a very important factor, if not the most important. This is due to a) physics, b) the competition rules.

Regards, Ian

[Composites Guy]

KB58:

Weight loss is a goal that won't be achieved without a VERY focused intent from the outset. Low weight is a good ENGINEERING based goal.... there is no EMOTIONAL aspect to that decision. A lightweight car WILL be faster than a car with excess (baggage) weight on it. Systems on your car should be designed as a whole for low weight and simplicity. Then each part should be designed for low weight, given the constraints of neccessary strength and deflections. The fact that RMIT makes a darn good car at less than 340 lbs means it is possible...

A captain deciding to make a 360 lb car when others are making 400 lb cars should be saluted! Its a good goal, and an attainable one! Of course he must assess the abilities of himself and his teammates and make sure that everyone is onboard for some very hard work.

Affordability is NOT a determining factor for FSAE. If all you can afford is 15lb junkyard wheels something is wrong. You CAN, and SHOULD find more money. That money is out there and if your team spends 5000 hours designing and building parts and 5 hours getting money... something needs to change.

If it takes an extra $5000 to get you from 400 lbs to 360lbs then find the $5000. If your systems engineers can only 'safely' build a 400 lb car then find some new engineers.