I need to specify a wire part that is bent in three dimensions. After the wire is formed by Supplier A, it is coated with Plastisol by Supplier B.

The Supplier A will use a check fixture for the unfinished part. The wire geometry matters. We don't care how thick or even the coating is, as long as it meets a minimum thickness.

The simplest way to specify the part seems to be identifying an ideal 3-D path for the centerline of the wire and allow some variation vs. that ideal shape.

Specifying the profile of the edge of the coated part isn't what we need to control. Can I specify a line profile for the centerline of the wire?

If not, how do you recommend I specify the geometry of the bare wire underneath the coating?

Mark Anderson
Senior Mechanical Engineer
Anesthesia Machines

Maybe use the position symbol (the boundary idea) to control the 3-D space of the wire. It sounds like that will best represent what you're trying to gage.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
http://www.gdtseminars.com

Thanks for your reply.

The wire needs to thread a 3-D path between and around stationary obstacles that sit in the same 3-D space. It's a 3/8" diameter wire guard that protects electrical connectors from being bumped and broken off.

Any "boundary" I define would have to snake around in 3-D space. I could specify a surface profile, but that would be tied to the surface of the coated part (or surface of the bare part). It seems much more intuitive to specify the path of the centerline of the wire.

If this can be done with a "position" symbol, please provide more detail.

-- Mark

I often design custom checking fixtures instead of the finished part. So my CAD model is for reference only. The checking fixture(s) might check form, function, or both. Here is one of the previous discussions:

http://www.eng-tips.com/viewthread.cfm?qid=313733

Peter Truitt
Minnesota

I love GD&T as much as anyone, but in this case, I would treat that wire the same way I would treat a single line bent pipe spool, and forget the GDT altogether. Not saying you couldn't come up with a method, but since it is coated anyway, why bother?

Unfortunately you can't tolerance a surface to a theoretical centerline, either by traditional or GD&T methods. I have done coated parts before in a couple of ways. First method, I did a drawing showing both the virgin and coated surfaces, and put profile tolerances on both; datuming can be a challenge. Second method, I put a tolerance on the thickness of the coating; this was valid because ultrasonic testing or resistance testing could be used to validate the thickness.

Jim Sykes, P.Eng, GDTP-S
Profile Services www.profileservices.ca
TecEase, Inc. www.tec-ease.com