But They’re Hard to Make!
Posted on 06 June 2017
We’ve mentioned several times how “tricky” it is to make our dividers. I imagine some of you find that odd (or maybe just whiny) - I understand. If I weren’t so immersed in making them, I’d be skeptical, too.
I mean, the tool has a total of four parts. (Five if you include the off-the-shelf washer – I don’t.) They are made using a single material (high-carbon steel) that isn’t even anodized, and there isn’t a single laser. Not one.
This is a world where you can buy a toaster oven with artificial intelligence, 12 kinds of safety lockouts and a Wi-Fi link to NASA for like $13 and free shipping. As I write this, some kid in Nebraska just made a full-scale model of Versailles or something with a 3D printer and dial-up Internet.
So what could possibly be so “difficult” about making two shapely sticks with a trumped-up dowel joint?
So first – the fine print. What we mean, really, is that they’re incredibly difficult to make within the confines of anything remotely resembling “affordability.”
Now just wait a second. Before you erupt all over the comment section about my use of the term “affordable,” know that I’ll talk more about why they cost what they do in the next entry. Until then, if we could just push “pause” on the whole pitchfork-gathering and effigy-torching thing, I’d be eternally grateful.
So let’s cover a couple of the first-order issues that make these difficult to produce.
First is just the overall shape. The dividers are curved on five out of their six exterior faces. We did sneak in about 2” of flat area at the top of each leg. Still, all that curvature creates significant workholding challenges, and a real deficiency of convenient reference surfaces for precisely putting parts in a CNC mill. No matter how the process is ordered, you’re going to have to grab some curved sections sooner or later and maintain some fairly tight repeatability tolerances while you do.
In practice, this adds up to the need for solid and creative fixturing to hold the parts at every stage. Or, for instance, at the very tip of each leg we’re talking about pieces of steel that have a cross section that’s 3/32” square and sticks out about 4” from the relative stability of the upper section, where the primary reference faces are. Add in the tapered chamfers on each face, you also have less than 1/16” of actual real estate (curved, of course) available for a clamping surface at those tips.
In short, these parts are long and thin, tough to hold, and susceptible to vibration and flexing.
The second issue is the number of setups these require on the mill. Ideally, you want to do everything in a single setup, because every time you remove and re-orient a part it adds (or multiplies) errors and inconsistencies. In practice, you rarely can make a tool in one setup, but by ordering things to work the most critical interdependencies together, things are usually manageable.
The dividers have rather a number of critical relationships like this. And to make things work properly, that means a lot of tradeoffs. In the end, we found we needed a minimum of three operations on the legs, and we have some pretty tough tolerances to hit across all the setup changes.
All of these things are doable. This is the sort of thing machinists, designers and manufacturers deal with every day. But every complication carries costs – in time, money, man-hours, fixturing and in how much error the process can tolerate.
Change out a cutter a bit too late, tighten clamps in the wrong order, or miss a metal shaving when cleaning up fixture plate – any of those errors cause problems. And those problems are cumulative. If the part isn’t secured perfectly, vibration increases. That makes cutters wear faster at best. More wear equals more vibration, and both add up to more force needed in each cut, and more force means more heat. More heat means more distortion of parts. More distortion leads to more vibration and higher forces and more heat. In other words: a vicious cycle.
How does all this play out? The dividers can absolutely be made. But it’s a procedure that’s sensitive to tiny errors. As long as the operator is competent and focused, things go fine. But it doesn’t take all that much distraction, lost sleep or overworked tooling to turn a batch of dividers into scrap. And quite possibly shred a few hundred or thousand dollars in tooling.
All these details can be hard to explain to someone who doesn’t do this sort of work. In most cases, it’s not worth explaining this in detail to customers.
In which case, it’s usually much easier to just say: It’s complicated.