Loading…
27
Showerthought: a 0.001 inch error on a 20 inch part is a 0.005 inch error by the time you hit the last cut
I was reading a forum thread about tool deflection last week and a machinist posted that number, and it stopped me cold. I always figured small errors stay small, but they stack up along the whole length of the part. So that sloppy setup I rushed through on a 16 inch aluminum bracket last month? That's exactly why the bolt holes came out 30 thousandths off and I had to scrap it. How do you guys deal with error stacking on long parts, do you dial in the middle or just chase it the whole way down?
1 comments
Log in to join the discussion
Log In1 Comment
ryanh561d ago
That math hits hard when you see it laid out like that. What I've found on longer parts is you gotta indicate the vise or fixture in before you even touch the part, because a few tenths of slop back at the jaws turns into way more at the far end. On anything over a foot I usually dial in the middle and let it run, then check the ends after and split the difference on my next pass if it's still off. Also, spring passes help a ton on finish cuts because the tool is gonna deflect no matter how tight your setup is, so letting it clean up with no feed pressure gets you closer than chasing it the whole way down. Sometimes you just have to accept that a 16 inch bracket is gonna move on you and plan the order of ops so the critical holes get drilled last, when the part is closest to its final shape. What kind of work holding are you running on those long parts.
7