Making your own race parts

You need to do something thats universal which is hard because most parts arnt. I think vortex, woodcraft. Flat and larger foot pegs are a great idea.

 

Ok, if I had a hi - Q DMLS printer, I'd titanium the shit out of it...

 

Few things I've done in the past year. Latest project was a replica of the motogp yamaha triple. So many parts, so little time to mill...

 

Go for it make all the parts. It will make you a better engineer even if they fail. You did yourself a great service by joining FSAE. Just go for a ride or stare at your bike and find problems or ugly parts.

I designed/made a full catalog of prototype parts for my 01' R1. Here is the sprocket cover and the process from my facebook post. I'm going to make a run at making/selling aftermarket parts. My old R1 has everything now it on to a modern bike.





"I figured I could share the process for bringing an idea into reality. All added up this was 15ish hours of work to go from desire to make a sprocket guard to finished billet Aluminum prototype.

The process starts with measurements of the mounting and counter-shaft sprocket area and putting those measurements into a CAD program (computer aided drafting)

Once the measurements are checked then I iteratively sketched out multiple designs and layouts until settling on a finished design in CAD.

To do a final verification of the design. The CAD model is printed out in ABS plastic on my 3D printer. It's bolted to the bike and checked for fitment.

Now it's on to metal. From this point on there can be no mistakes one oversight, one wrong math problem, or missed decimal point, means lost time, a crashed cnc machine, a scrapped part, and starting over all over. Some times its not so dramatic as a crashed machine sometimes it's the part was over cut by .005 of an inch. It still means starting over from scratch.
The cad model is used to create a G-code computer program using a CAM (computer aided manufacturing) program. At the same time fixtures to hold the part while it's being machined are designed. Specific tools to make the part are loaded into the machine (9 tools for this part) the vice and or fixtures are aligned to a machine axis. Origins and each tools specific length are inputted into the CNC machine control. Finally the G-code program is sent to the cnc control, and the CNC machine cuts the part. Hopefully turning a block of raw metal into a useful peice of art."

 

That Triple looks great

 

I also went the extra step of 3D scanning my R1.

 

id love to know more about this. setup the scanner on both sides and run it? how much cleanup did u have to do on the model?

 

I really like that Sprocket Cover, I want to make some for KTM LC4 engines....

 

I've been doing this setup on most of my bikes lately- oem sprocket cover gets yanked and I run a home made aluminum bracket for the universal Bazzaz QS sensor adjuster mambo-jambo thingy with zip ties

 

Like this:

 

And on this one too:

 

Question about printed parts:
Can the plastic have characteristics of Delrin, specifically, it's thermal expansion and self-lube properties?

 

This thread actually made me start an account, after probably 8 years of lurking. There is nothing cooler than being able to make your own race parts, even if there are similar things out there to do the job. Anything you buy that somebody else is making, was made to fit a price point, with an eye towards sales. When you're building stuff for yourself, it's your time, so you can get crazy with things. Sometimes it's because you can't easily source the parts you need. Case in point, ever try to buy an intact gauge pod for an RZ350? I think you can get a Unicorn for less. So, I made one out of billet aluminum. I'm thinking to have this done by someone as a one-off, would probably cost $2-3k. For me, it was just a spare chunk of 6061....and a lot of time.

BTW, really fine work, fallrisk and R1-Kid. What type of equipment are you using for those results? Scanner?

 

The actual 3D scan of the bike was the easy part. Converting that scan data into useful CAD data was the challenging part. I scanned the individual major parts of the bike (frame. engine, bodywork, swingarm, ect) with a Creaform GO!scan3D scanner then. After cleanup and conversion into dumb solids/ parametric solids. everything was assembled in CAD, so I could design parts in context of the motorcycle. It made the exhaust design very straight forward.
I used the project to learn Geomagic Design X software. There is probably 6 hours of scanning and 40-60 hours of CAD work. It would be much less now that I know what I doing.



Thanks Britt. I would make, and give you one if had a KTM to design off of. My biggest challenge is getting ahold of bikes to design/make parts for.

 

Damn, I could use one of those!
Will trade scans for parts...

 

Appreciate the offer,
I am still learning (re-learning) Cad/Cam and my CNC...I will get there, working gets in the way of fun sometimes..

 

R1, did you buy the Creaform for your business, or is it just for personal use? I can't get past the purchase price, even on the GO!, let alone the HandyScan. I've been using a NextEngine since they first came out, and though it has it's limitations, the workflow is much quicker than Geomagic. It really doesn't do too badly if you're not working on something super tight tolerance (scans are +/- .009"), but anytime you have something that has sharp edges, or holes, it takes quite a bit more work to collect the data, and I wouldn't use it to reverse engineer a close tolerance part. You're absolutely right about converting the data though. Getting the scan is the easy part.

 

How about some ridiculously expensive headlight brackets (if you were to pay someone to make them).

 

700$...