I figured this would be a good place to ask since there quite a few talented and experienced machinist in here.

These components are for an FSAE Carbon Fiber A-arm suspension. They will be made out of aluminum most likely. The tabs will probably be machined in (they aren't modeled that way because these are just a crude concept sketchs). These are for mounting the push/pull rods.


The first CAD model is the ideal method to save weight. The CF tube ID will slip over the insert rods.





The second CAD model was an attempt to make machining more reasonable. The OD of he CF tubes will insert into the holes of the piece.






I will shed more weight from these models and refine the designs later but I am more concerned about manufacturability at this point. What do you guys think?


.

 

With some small changes, the first model is definitely doable on a 3-axis mill. It would take a bunch of setups, but it can be done.

--Ferdi

 

Thanks for the reply. What small changes would you recommend? Also how do you make the round inserts on a mill?

 

I was thinking it was all one piece. If the round bosses are separate, I'd make them on a lathe.

The sharp inside corner between the round bosses is pretty much a no-no. Also making the outside walls straight instead of curved would help.

--Ferdi

 

They bosses are one piece attached to the bearing insert.

Thanks for the advice. Once again, just concept sketches... Small issues like that will be addressed when later in the design phase.

 

You just have to stand the piece up to make the round bosses.

--Ferdi

 

How well can you hold circularity with that method? That dimension is pretty critical to maintain a consistent and tight tolerance bond gap for the epoxy.

 

On a CNC, not an issue.

--Ferdi

 

Completely possible. Like what he said though, it needs a few small changes.

Other than that, it's going to have to be fixtured like 4 different ways. Plus it's going to be pretty costly.

 

Definitely a no-no. In any material this notch point will make for disaster.

 

You would need a 5-axis CNC tho right?

 

why not make the eyelets on top a seperate peice with an allen head titanium through bolt from the bottom?

or weld it on after then heat treat it.

 

Not being a programmer but dealing with crybaby programmers daily, either look 3axis doable once you make the walls straight, as been mentioned.

Hell, you could also do it on a bridgeport depending on how much you value your time. I'm amazed at what some of the old-timers in the prototype shop can do with manual machines and proper tooling & fixturing.

 

The shop i work at makes a ton of little parts like that. 95% of the time it's more than one piece. I'm usually welding it, or someone is, then it goes out for heat treatment and stress relief. Sometimes even goes back into the machine for finish work.

With the proper weld and a good process it should be just as strong as a billet unit.

 

Yeah, anything is possible.

Coming from the manufacturing world, if you want to make a bunch and have them be strong, lathe the steel tubes, stamp or laser cut the tabs, machine the Y shaped boss on a burn table followed by a 3-axis mill (burn-face-flip-face-bore) then weld the three together.

But again, for a one-off piece (or I'm guessing two-off), it's all doable as one piece.

 

The only issue I see with welding is finding a capable aluminum welder. Our team has members who can weld steel well but I don't know how confident they are in their Al welds. Making these parts out of steel would offset any weight savings seen from CF tubing compared to the original steel A-arms.

The other option I am looking into is making them out of titanium and having them welded.



Ya, it as amazing what some machinist are capable of. Unfortunately I am not one. lol. These would either need to be simplified or we have access to a 3 axis CNC mill at school.


I am drafting a couple multi-piece designs as well to make them more manufacturable in house by myself or another student. I will post these up when I get them finished.



Another option I'm looking into is titanium rapid prototyping if we can find a vendor to sponsor us.



Thanks everyone for the input.

 

This is all depending on number of parts...
You're going to pay loads less on welding a few pieces together than you would trying to CnC one piece....

Unless you have someone to program and run a CnC for free.

 

Nope, 3-axis CNC is enough. Like Erat already said, it'll just take 4-5 setups to get them done.

--Ferdi