Design Sprint Goals

This design sprint is intended to be a good starting point for new team members to learn about engineering concepts and how to develop a design from (nearly) the ground up. Other goals include:

• Making Self-Driven Decisions

Finding flaws in your design on your own and determining ways to fix them.

• Documenting Design Decisions

Keeping track of what changes you made and why you made them to your design.

• Working Independently in a Team

Being apart of a bigger project while taking on the responsibility of an aspect of it.

And the potentially more nuanced goal of this sprint is to design a functional trailing arm! Note that we’re going to be designing a trailing arm that will go on THE LEFT SIDE of the car. It isn’t important in the actual design, but makes the loading conditions easier to describe later on.

Topics Covered

• Design Concepting

• Computer-Aided Design

• Design for Manufacturing (DFM)

• *Design for Assembly (DFA)

• Finite Element Analysis (FEA)

*Not guaranteed, but the opportunity is there!

Background Information

Let’s start at the basics, what is a trailing arm suspension? It’s commonly seen in the rear suspension of bikes as seen below.

The above example uses a bellcrank to change the direction of the force, which we won’t be using due to added complexity and weight. There’s an explanation for what a bellcrank is and how it works below, but it’s not necessary for the design sprint.

 Explanation of the Bellcrank

As mentioned above, a bellcrank is used when force needs to be redirected. A bellcrank can be described by 3 points/nodes. An “input”, “output”, and rotation point.

The rotation point is fixed, and will not move, but the input and output points can rotate around the rotation point.

For example, let’s say a force is applied on the vertical linkage going downwards. to solve for a reaction on the horizontal linkage you’d use the sum of moments about the rotation point. This means that the force on the output linkage may not be the same as the input linkage because mechanical advantage can be designed into a bellcrank by adjusting the distance of the other two points from the rotation points.

The key thing to note about a trailing suspension are:

• There is a point where the trailing arm pivots around

• There is a point where the wheel is attached

• There is a point where the shock is attached

The point where the wheel is attached is where the all the forces “enter” the system, and the shock and pivot points will counteract those forces.

The blue circle represents the wheel which isn’t to scale, but gets the idea across (hopefully).

However, the sketch above is only a side view of the wheel.

In the top view show above the system becomes a bit more complex. We can see an additional branch/arm the connects to the pivot axis. This is to help distribute forces which you’ll see soon enough.

I’ll quickly note here that only one of the rear wheels will be powered (ie. have the motor in it). I want to create two trailing arms which are mirrors of one another, and for the side that will not have the motor we will create a spacer to make up the distance between the mounting face to the rim.

At this point, start thinking about what can be changed in the concept. For instance:

• How far does the wheel need to be from the pivot point?

• How far does the shock mount need to be from the pivot point?

• Do all the points need to lie in the same line? (Can the shock be moved vertically?)

• Does trailing arm need to be horizontal? (At what angle is it optimized?)

If you’re getting a bit overwhelmed, that’s normal! This is likely the first time you’re approaching a design problem in this way, so it takes some time to get used to, and also feel free to ask questions! Don’t think you’re bothering us, when you ask questions, you show you’re interested and we want to help build that interest!

Now, back to the technical stuff. To answer a lot of these questions we need to set some design constraints.

Design Constraints

You’ll probably learn some different terminology on this, but I’m using it as the values we know that we should design around. A big part of designing parts is finding number that you need to work around, but it’s very time consuming, so I’ve done the tedious stuff so you don’t! So, I’ll list them out here, and I’ll add in sections that show how I determined these values. You don’t need to know it, but I want to feed people’s curiosity where I can, and it stands to show a good example of documenting your design decisions.

Maximum Vertical Displacement for the Wheel: +/- 9 cm

Maximum Angular Displacement for the Trailing Arm: 10 degrees from the rest position

 Reasoning

From McMaster Carr (a good supplier, but expensive) we can see that swivel bearings usually are rated for around 20 degrees of swivel. We don’t have a specific bearing specified, we’ll need to judge our movement and loading requirements before we can do that.

So for now, we’ll limit ourselves to +/- 10 degrees with the thought that reducing it is better

However, this assumes the bearing will be mounted in such a way where the main axis of rotation is along the Z axis. If you find a different configuration more advantageous, go for it!

Angle between the Pivot and the Wheel Mount: Greater than Zero at Max Extension

At max compression the shock should be tangent to a circle concentric to the pivot point.

Motor Mounting Pattern

If this doesn’t make any sense, it means the motor has 6 equally-spaced threaded holes (M8 thread, so the bolts are 8 mm in diameter) along a circle of diameter 64 +/- 0.1 mm

Loading Cases From the Wheel (must pass all four loading cases)

Positive x is rear to front

Positive y is right to left

Positive z is up

 Reasoning

The regulations define the loading condition in terms of acceleration; 1G Steering, 1G Braking, 2G Bump (https://www.americansolarchallenge.org/ASC/wp-content/uploads/2021/12/ASC2024-Regs-EXTERNAL-RELEASE-A.pdf, Appendix F, F.2). We can see how load transfer effects the loading conditions on the wheels. Since the trailing arms are used on the rear of the car we’ll specifically look at those values.

The above screenshot is from a spreadsheet I developed that will calculate the the load distribution under a 1G brake and 1G steer. In a braking scenario there is more weight loaded to the front however. So I made some adjustments to the calculator and we can see that the rear should expect around 111 kg of mass.

So by the 2G bump case, we should expect an upward of force of around 2177.82 N

How the steering and braking cases impact our loading conditions are through the friction between the tire and the ground. If the car was turning with a centripetal acceleration of 1G, it would require a force equal to it’s weight. This force would be supplied by the friction from the tire, which is calculated by the coefficient of static friction and the normal force. Based on generally accepted values, the coefficient has a value less than 1, which means it cannot produce a force equal to the the weight of the car.

So now we need to make a decision. The car needs to be safe, but we don’t want unrealistic loading conditions either. Based on our priority to have a race-worthy car rather then a high performing car, we’ll assume the higher loading which means that the expected loading is:

Loading Condition	Load	Notes
2G Bump	2177.82 N Upward	Will be applied in all simulations
1G Acceleration	1088.91 N Forward	Will create higher compressive stresses in the part
1G Brake	784.8 N Backward	Will create tensile stresses in the part
1G Steer Left	1088.91 N Inward
1G Steer Right	1088.91 N Outward

There two additional to consider on top of this; the application of a safety factor, and the location of the loading.

At the time of writing this, the highest safety factor that needed to be applied in the previous car was 2 and therefore will be applied to this sprint. However, if this changes I’ll add it to the end of this section. So the new loading conditions are:

Loading Condition	Load
2G Bump	4355.64 N Upward
1G Acceleration	2177.82 N Forward
1G Brake	1569.6 N Backward
1G Steer Left	2177.82 N Inward
1G Steer Right	2177.82 N Outward

By the regulations, the loading condition is applied to the contact path of the tire (where the tire makes contact with the road). Because of this separation between the application point and the wheel mounting point, there will also be moments generated around the wheel loading point. The unloaded diameter of the wheel is 557 mm, so, there will be a moment from the acceleration/brake condition and the steering conditions. The moment arm for the 2G bump case comes from the geometry of the wheel assembly which is measured to be 64.1105 mm. There is also pneumatic trail which would create more moments around the mounting point, which should not be neglected. Based on the tire specifications under Sources and some comparison to other rolling resistance coefficients, the rolling resistance coefficient given is likely not a unitless coefficient, but a measurement of the pneumatic trail as rolling resistance is the force required to overcome the moment created due to a non-uniform loading at the tire. Units are not provided, and thus will be assumed as millimeters. Meaning the pneumatic trail is 3.02 mm at it’s worst (larger moment arm, larger moment).

Basically from here calculate all the moments created, and you’ll see them at the top of the dock!

Also, if you read this far please verify my calculations (I’m doing them at 2:30am), first person who does I’ll give a really cool sticker.

Lastly, we’ll design with the following shock in mind:https://www.royaldistributing.com/bronco-rear-gas-shock-for-yamaha-au-04409.html

We will not be using this shock (or at least it’s very unlikely), so be sure to keep track of your decisions to justify where the shock mounting point goes.

In case website link doesn’t work, the important dimensions are:

Design Criteria

Earlier we looked at constraints, specific numbers that limit what solutions are possible. Criteria are a bit more continuous rather than restricted. The best way to explain it is to apply them to this sprint.

Mass - Since it takes more energy to move an object with more mass, we want to minimize the mass of our car to minimize energy consumption. Therefore concepts that use less material will be better than those with more material.

Cost - We don’t run on an infinite budget, money is a limited resource. Since we can only spend money once, design that cost less will be better. Since you may not know much about materials and manufacturing methods, start off with “blocking” out the part you made. Bigger blocks are more expensive and smaller blocks are cheaper. The explanation is in the expansion below.

Timeline

Week 1 - Concepting

Sept 11 - Sept 17

The idea’s pretty simple, come up with ideas! In concepting we want to come up with as many possible solutions as possible that can solve the problem we have. In our case, the problem is how do we support the wheel given all the information in Design Constraints.

What also might be useful to consider is how to make the part. If you take the best design possible that can’t be manufactured to a machinist, they still can’t make it. We won’t have the training/info session on manufacturing methods until later in the sprint, so only consider it at this stage if you have the time.

Deliverables

2 - 3 Concepts with Sketches (Digital or on Paper)

Week 2 - Computer-Aided Design (CAD)

Sept 18 - Sept 24

Now that we have some ideas on how to solve our problem, we need to pick one to develop further. Developing multiple ideas in parallel is very time consuming and not advised, but we won’t judge if you want the extra practice.

With a concept selected, we need to digitize it! This is where CAD comes into play, by this point we’ve done the training session on how to CAD a part and I want you to do that with your part, but the part needs to be fully defined. In case it isn’t mentioned in the training session, having a fully defined part means that all dimensions needed to define a part are present. In SolidWorks at the bottom right corner you can see if you part is fully defined.

An extension to this part of the design sprint would be to make your design parametric. A parametric design means that the dimensions of the part are defined “externally” from the file so it can automatically update if something needs to be changed. In essence, it makes it easier to tweak your design by not needing to dig through your feature tree to change a dimension.

Deliverables

1 Concept in SolidWorks Fully Defined

Week 3 - Manufacturing Design (DFM)

Sept 26 - Oct 1

With a design digitized, it time to start thinking about how it’s going to be made. Like I mentioned before, a great design is actually pretty bad if it can’t be made. Now that you’ve learned about different manufacturing techniques, it’s time to start specifying material and manufacturing techniques.

I hope at this stage you are looking at your design to make changes to make it easier to manufacture. We can’t purely rely on “if there’s a will, there’s a way” when we want to make something, and this means something need to give, either our bank account, or our design. We rather our design whenever possible.

A hint/idea I’ll throw out there is to see if your design can be made using multiple parts. I’ll try to bring example trailing arms during the presentation, but hopefully you can consider your design not as a single, solid piece of metal, but smaller more manageable chunks. Notices how this reduces costs as well, smaller blocks of metal are cheaper per unit volume than larger pieces of metal.

Deliverables

Bill of Materials with a manufacturing plan along with any changes to your design in CAD.

Week 4 - Static Structural Simulations (FEA)

Oct 2 - Oct 8

We have a design we can make, but will it hold up to the forces we need it to? Again, the idea is simple, but the execution is a lot harder. Hopefully you’ll understand the basics of SolidWorks FEA to run the simulation, but if you’re having any trouble with it, feel free to reach out!

Maybe after running your simulation, your part fails, but does the stress distribution make sense? Maybe it passes, but does the stress distribution make sense? I really want to reinforce the fact that we can’t accept whatever we get from the computer blindly. If they make sense, and your part is failing, then your design needs to be tweaked.

Some common changes that might help:

If there’s a very high stress concentration in a sharp corner, add a filet to get rid of the stress singularity.

If through a thickness the part is failing, make the section thicker to increase the cross-sectional area.

Deliverables

A passing simulation with a stress distribution that makes sense along with any changes to CAD.

Final Review - Oct 22nd

This is where we get to put all your work together and see what others came up with! Everyone’s solution to the problem will be different, and their approach is something you can learn from. What we’ll do for the final review is combine the small groups on each time slot to see more solutions!

The date is not a typo, but it’s considering reading week and midterms for most of you. Which also means that you don’t need to finish the FEA for the 8th. But, I imagine during reading week and midterm week you have better things to do than figure out why a simulation isn’t working, but I let that be up to you to figure out.

Info Session Schedule

*Won’t be recorded, but training materials will be uploaded to Confluence.

All will be happening from after Mech General (~1:15pm) to 2:30pm (hopefully). After the training session we’ll start doing reviews!

Review Schedule

We’ll be making small groups of around 3 people in which you will be paired up with either myself or Shangheethan. We’ll try to spend about 15 minutes on each of your designs, but we’re hoping it becomes a bit of a discussion on what the strengths were of designs and where there’s room for improvement.

Fill out this form to say what preference you have for review time on each day. I broke the form up into each review session. I’ll close the forms the Thursday before the review sessions. Also, you can let us know if you want to be paired with anyone in the review session!

Sept 17th || Sept 24th || Oct 1st || Oct 22nd

Sources

Credits

This design sprint was heavily inspired by the one created by Aidan Lehal, Min Qian Lu, Kevin Bui, and Emily Guo! Big shoutout to them for the hard work they put in!

W22 Design Sprint