Custom frame design comments/concernces/constructive criticism

9862 Views 66 Replies 10 Participants Last post by sebwiers, Jan 13, 2018

UniHalf

Discussion Starter · Dec 31, 2017

Alright, y'all. Not sure if this belongs here or in technical, but I'm assuming this is okay.

I've got an ambitious project in the works currently. I'm a Mechanical Engineering major in my junior year at Ohio State. I've done a lot with suspension and carbon fiber molding with the Formula SAE team there so I'm not entirely out of my depth here, but we'll see how it goes. I've got an internship in Pennsylvania so over the next 8 months I'm going to start designing and building a motorcycle. First step is drawing up a 3D model of the frame. I'm posting the first draft of the design here because you all have way more experience building/modifying bikes than I do. I'm curious to get your opinions.

Does my design have any glaring errors that I'm missing? What could I improve here? Be harsh if you want, I want to make my design as good as possible. I'm mostly going to be setting it up for road racing/street driving. Good handling and feel is one of my top priorities here.

Diagram Technical drawing Design Drawing Plan

This is my recent iteration.

Thoughts/comments/concerns?

See less See more

Save

1 - 13 of 67 Posts

UniHalf

Discussion Starter · #2 · Dec 31, 2017

Some more detailed views:

Technical drawing Text Drawing Diagram Line

Line Diagram Parallel Drawing Automotive window part

Line art Drawing Diagram Technical drawing Parallel

See less See more

Save

UniHalf

Discussion Starter · #3 · Dec 31, 2017

Stupid mistake #1 that I've just discovered going over these photos: the rear chassis brace between the rear engine mounts would block me from putting a chain on.

Save

UniHalf

Discussion Starter · #7 · Dec 31, 2017

I'm designing the frame around the engine (Kawasaki Ninja 250R) and suspension I have in mind. So the dimensions came directly from my height and the mounting points on that engine.

The CAD is necessary partially because it just is the standard for mechanical engineering at this point. It'll also let me do FEA on the tensile strength and fluid dynamics of the end product.

Criticise the CAD all you want but I get paid partially to know how to use it so it's still to my benefit.

I have engine, suspension, swingarm, steering, fuel tank, battery mounting points that I'm working with on the design. A few of those still need to be e added. What is it that I'm completely missing here that makes this a bad starting place?

Save

UniHalf

Discussion Starter · #13 · Dec 31, 2017

I like the handling of the KTM RC 390, so I'm modeling the front suspension after that bike, so steering head angle 65° and 100mm trail. I'll revise the spring rates because this bike should be significantly lighter.

I'll also most likely be using the wheels from the 390 so 17s on both ends. I'm using those because the wheels are relatively like to keep the unsprung weight down and because the can be bought cheaply.

The rear suspension I have to get a little more creative with, so I have some research to do before I decide that and the swing arm's exact geometry. I'm planning on dual adjustable socks and dampers from Race Tech.

Save

UniHalf

Discussion Starter · #16 · Dec 31, 2017

It'll be fuel injected.

Not needing a saddle fuel tank is a brilliant point. I could have some fun with that, I'm sure.

Anyway the point here is to not create an exact geometry replica. I'm using the engine from a Ninja 250, but that's about all I'm using from it. If I wanted a Ninja 250, I'd go out and buy one. It'd be way less expensive and time-consuming than this. Also computers are virtually all of engineering at this point. This being a terrible design is the reason it's in CAD in the first place. I can revise and revise until it's a good design.

I had been planning 1 1/4" steel tubing. Not sure of the wall thickness yet. Should I go higher than that? I'm thinking 1 1/2" might be overkill and add unnecessary mass.

Save

UniHalf

Discussion Starter · #19 · Dec 31, 2017

woodsman said:
Your using a honda front end aren't you? How are the triples going to play into your KTM geometry? 100mm trail seems excessive, I believe that generally it should run in the 55 to 75mm range. What you do with the rear suspension will affect the front end geometry.

You're probably right here. Especially because it won't be a super fast bike in a straight line. I could lower the trail without too much trouble.

Save

UniHalf

Discussion Starter · #20 · Dec 31, 2017

TrialsRider said:
UniHalf said:

It'll be fuel injected.
Not needing a saddle fuel tank is a brilliant point. I could have some fun with that, I'm sure ...

Click to expand...

Good start because that thing that looks like a fuel tank on a modern GP sportbike is not a fuel tank, it's the air box.
Almost nothing related to frame members is overkill and unnecessary mass, the frame is what holds the bike together and stops it from feeling like it has hinge in the middle. But just thinking something is the right shape or size is not going to cut it, manufactures pay huge bucks to do stress analysis and destructive testing on those things during the design phase, they don't just do it by eye. At least not in this century.

Gotcha. The stress analysis is going to be important to the project too so I can show potential employers I can do that type of FEA.

I'll go back to the drawing board and check back when I have a second draft.

See less See more

Save

UniHalf

Discussion Starter · #26 · Dec 31, 2017

jcw said:
I do agree with the above that you model a dummy engine and build the frame around it, rather than design a frame from scratch and fit the engine in it.

Since you obviously have skills in modeling, I would mock up a very simple engine "dummy block" and wrap the frame around that.

A close tight frame looks better anyway, than a frame with huge unnecessary gaps and spaces.

I'm going to see what I can do about taking a 3D scan of the engine. There are some good programs out there free to students. I'll start the second draft from there.

Thanks for the book recommendation too. I'll grab a copy. I'm sure it'll come in handy.

Save

UniHalf

Discussion Starter · #32 · Jan 2, 2018

DesmoDog said:
Fluid dynamics? On a frame? I doubt it. ;-)

I know you're kidding but I still wanted to mention that I say fluid dynamics because one of my end design goals is less than a certain drag coefficient. So I'll see how that works in the (fairly distant) future.

DesmoDog said:
The Book on design that was mentioned. Absolutely fidn it and buy it. I got a copy back when I wanted to build a track bike and spent countless hours going over it. Very useful. I posted info on it here years ago, if I can find it I'll edit the post with a link.

Matrerials - someone mentioned sticking with typical steel and staying away from Chromoly or whatever. +1000. Do enough research and decide for yourself, but the benefits of using something fancy, especialy on your first frame, are highly overstated IMHO. Somewhere on here I made the argument that using chromoly only resulted in frame that was harder to dent.

I did get a copy of that book and I'm waiting for it currently. I think I'm going to go with a trellis style frame and go from there.

Save

Reactions: 1

UniHalf

Discussion Starter · #40 · Jan 2, 2018

TrialsRider said:
Not sure how to interpret that "less than a certain drag coefficient", is your goal to make it wind slippery to the point of having less than a 1.0 wind drag coefficient? If yes, pretty sure you are going to need access to a wind tunnel test facility, that is how BMW and Moto Guzzi did it.

Wind tunnel testing might actually be a possibility with my connections from Formula SAE if I'm willing to beg, borrow, and steal enough to make it happen. Probably a bit of a pipe dream, but who knows.

Save

UniHalf

Discussion Starter · #44 · Jan 3, 2018

jcw said:
You could design your intake manifold based on fluid dynamic designs...

I plan to

TrialsRider said:
Oh by the way UniHalf, can you even ride a motorcycle and are you any good at it?
If you set out to design a better mouse trap it is a good idea to start by having some mouse killing experience

I /can/ ride a motorcycle. I have an '86 K100 that I ride occasionally. Terrible starter bike. Other than that I've done some test rides on other bikes.

Whether or not I'm good is a completely different question. I'm near mid-Ohio so I want to do a track day there at some point.

Save

UniHalf

Discussion Starter · #51 · Jan 3, 2018 (Edited)

TrialsRider; said:
So if this thread is all about a design exercise to pass a test or graduate from somewhere, I give it 35% grade at most and that is only if it runs and rides like a motorcycle.

I can work with 35%. That means there's a lot of room for improvement.

Once I have the textbook I imagine the test will be easier.

Save

1 - 13 of 67 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.

Top Contributors this Month