Making your own race parts

For me the best part of making your own stuff is not having any design constraints as far as cost of machining. I got a quote for machining these clip-ons at a shop and it was 640$ per side... Nobody would pay me anything close to that but I can make my own projects as crazy as I want.

These are 320 grams for clipons with carbon tubes. If i remember correctly the clip on with cf tube was lighter than just the renthal tube.

 

With how the '03 R6 is, making my own clipons would be really beneficial because I can't use normal clipons, they're too low. Right now I have to use OEM bars, but if I made clipons that aren't as tall as aftermarket ones it'd be fine. Seems like a tough project, I like it...

 

fallrisk, are you using 4th axis, or are you creating fixtures to hold some of these parts? If it's not 4th axis, you're a fixturing mofo. Love the detail on the top of the clamps. It's little things like that which really make the difference, and that part is usually the easiest thing to add in. Are you cutting the clamp slots with an arbor blade, or band sawing them? What CAM package are you using?

 

These little buggers took me a bit to figure out (I'm not the brightest bulb). I've got 4th axis, but that's way over my head for now. I figured out some soft jaws for these, and even though it's 6 ops, it's only something like 25 minutes a pair. I'd bet I could cut that in half if I could figure out how to do it 4th axis.

 

Well set up 3 axis mill with a slotting saw.

The clip-ons were 8 setups so you have to get creative with probing and fixturing. I design with machining in mind so I know ahead of time how creative I want to get with the parts or keep the work to a minimum.

Autodesk Inventor (comes with cam)

 

I've been doing design for quite some time, but only just started to delve into the toolpath and actual machining of it. Funny how you look at things differently when you know you're the one stuck trying to figure out how to hold some weird shape. I wish I learned more quickly, as some mistakes can be pretty costly. So, I have had access to Delcam Powermill, but just couldn't wrap my head around it, so have been using HSMWorks in Solidworks. Now that Autodesk has bought everything (Delcam, AutoCad, Fusion360, Inventor), it seems like things are migrating towards a very similar path. That's a good thing, as Powermill was not intuitive, though is very powerful.

If it weren't for having the probe on my Haas VF-2, I'd probably never get anywhere. I don't understand why anybody who buys a CNC, wouldn't spend the little bit of extra money (relatively) to have probing. Heck, just being able to set all the tools, while you walk away and do something else, is a huge time saver. I wish I had these resources back in the day, when I was actually still racing. I used to make parts for my RZ with a manual Bridgeport, and while they served their purpose, it doesn't compare with what can be done with modern tools. I'd love to have the scanner that R1_Kid has, and a nice 3D printer to use for testing fitment. More toys are more fun, but boy it's a lot harder to learn things than when I was in my 20's.

 

Well, to be fair, it's not like you're trying to learn how to use a bridgeport. Learning CNC is much more complex so give yourself a little credit.

 

The probe is a lifesaver. It also comes in handy as a precision measuring tool for checking parts like triples for straightness.

 

BOO!

 

The scan to CAD issue is pretty engineering industry wide. I've probably seen it come up in 5-6 projects over the past 2 or 3 years in multiple industries. Not a ton of people understand the various steps, and those that do have a leg up as more industries adopt scanning and/or printing technologies. If you're the entrepreneurial type, keep an eye out for opportunities if you aren't already.

As for the rest of this thread, interesting stuff. I've made my fair share of parts, and I still have my old CRF450 supersingle sitting in my garage for whenever I rebuild or find a motor. Now I try to keep a balance between easily repairable (ie widely used parts) and simple, effective designs. Last couple things I made were quick change stuff for the endurance SV and a few steering damper brackets. Anything that hits the ground I usually just buy.

 

Well, I clearly don't know how to use the 'quote' function...lol

 

I don't think those will be rigid enough Very nice work.

 

Lots of machining/cnc porn in this thread. I like.

Are you guys doing any FEM on the structural stuff (triples, clip-ons) or are you just using the trusty gut?

I've been wanting to make some custom head light brackets like those in this thread, but like others have said, the cost of machine time for someone to make them is prohibitive.

 

I've got FEA capability in Solidworks, and run some designs through it. In the end, you've got to know how to set up the simulation, and know what it's telling you. I don't trust the results, so I just leave a wide margin for failure. Yet another thing I need to learn more about.

 

Precisely

Fea Ftw

 

Yup, unless you're modeling things accurately, all you end up with are pretty pictures that have little meaning.

I had an intern set up an FEM for a weldment that consisted of plates welded to square tubes. He modeled the assembly as a unified solid and presented the results in a design review. I had to cut the presentation off as soon as he mentioned his method because the results were meaningless.

I'm weary about designing structural parts without at least doing some hand calcs first.

 

See, this is the difference between being an actual trained engineer, and somebody who's just plunking around figuring it out as you go. Probably why I struggle so much with the FEA software, because there are some basic engineering principles that need to be understood, and I'm sorely lacking. I can do simple FEA setups, like a footpeg design I did. Pretty easy in terms of what's the fixed point, and where the loads are distributed, but that doesn't take into consideration variation in live loads. I mean, you can test a peg to withstand a 400lb. rider, but that doesn't tell you the forces that might result if that rider were up on the pegs and hit a bump at speed. The software can certainly do it, but the knowledge required to do the analysis setup correctly, is where education comes in.

 

I heard a story that Colin Chapman would keep making parts lighter until they broke then he would go back to the previous revision. Iterative design/learning on the fly can yield positive results as long as you're willing to be the test rider .

 

It's great to have when people tell you that you are an idiot for doing it "like that"