If your machinist wouldn’t fixture from that datum, don’t tolerance position to it.

That line might sound a little blunt, but it explains a huge percentage of the position‑tolerance problems we see in the real world.

Acting as an exact coordinate, Position is the total permissible variation that a feature can have from its "true" position.

True position is one of the first GD&T symbols engineers reach for when a part needs to assemble cleanly. Hole patterns, pins, bearing bores, fasteners — position is the right tool for all of them. But position only works when the datum reference frame matches reality. And “reality” means how the part is actually held, machined, inspected, and assembled.

When those things don’t line up, you end up with parts that technically pass inspection and still don’t work.

Here’s why.

A position tolerance doesn’t locate a feature by itself. It doesn’t say “put this hole here.” What it really says is, “this feature is allowed to vary this much relative to these datums.” The datums establish the coordinate system. Position just controls how much deviation is allowed inside it.

That’s an important distinction, because it means all of the intelligence in a position callout lives upstream, in the datum choices.

If the datums don’t reflect how the part is constrained in the real world, the position tolerance can be perfectly applied and still miss the design intent.

This usually shows up in one of two ways.

1. Either the part assembles only if you force it

2. Or the assembly works but only when you mix and match parts until you find a “good set.”

In both cases, inspection reports look fine, which makes the problem even harder to diagnose.

A common pattern looks like this: an engineer selects a nice, flat surface or a convenient outside edge as the primary datum. On the drawing, it makes sense. It’s clean, easy to dimension from, and it’s accessible for inspection. But on the shop floor, that surface is never used to locate the part. It’s not stable enough, it’s not functionally relevant, or it simply doesn’t make sense to fixture from.

So the machinist does what machinists always do. They fixture from the surfaces that are repeatable, rigid, and meaningful. Inspection, meanwhile, measures from the drawing datums. Assembly locates from whatever interface actually constrains the part in the product. Now you’ve got three different “truths” about where the part is supposed to be.

All three are defensible. None of them agree.

Real Life Example

We’ve seen this with a machined aluminum plate that had a four‑hole bolt pattern used to mount it to a larger assembly. The drawing called out a tight position tolerance on the hole pattern, referenced to an outer profile edge as the primary datum, with a secondary datum on an adjacent face.

On paper, the holes were beautifully controlled. They also passed a CMM inspection without any issues.

In reality, the plate mounted against a machined pocket in the mating part. The pocket floor and a perpendicular wall constrained the plate during assembly — not the outside edge that was used as the datum on the drawing. The outside edge didn’t touch anything functional. It was just “there.”

During machining, the part was fixtured from the same surfaces that mattered in assembly: the main face and a perpendicular wall. That meant the hole pattern was consistent relative to the functional interfaces, but it could float slightly relative to the outside profile.

Inspection said the holes were in position. Assembly said otherwise.

The fix wasn’t tightening the tolerance. It was changing the datum scheme so the primary and secondary datums matched the way the part was actually located in the assembly. Once that happened, the same hole pattern passed inspection and assembled cleanly — with the same machining process and, in some cases, looser tolerances.

This is why tightening position is often the wrong response when something doesn’t fit.

Final Thoughts

When position problems show up late, the instinct is to clamp down. Smaller numbers feel safer. But if the datum reference frame allows unintended shift or rotation, no amount of tightening will fix the root cause. You’re just forcing manufacturing and inspection to work harder to maintain a flawed reference.

In contrast, when datums are chosen correctly, position becomes incredibly forgiving. Bonus tolerance at MMC actually behaves the way it’s supposed to. Variation stacks predictably. Parts interchange instead of being hand‑selected.

A simple gut check catches most of these issues early: would a machinist actually fixture this part from the datum I’m calling out?

 If the answer is no, that datum probably shouldn’t be controlling position.

Another good check is to think about how the part loses degrees of freedom in the assembly. Which surfaces stop translation? Which ones stop rotation? Those surfaces usually want to be your primary and secondary datums. When the datum reference frame mirrors the assembly constraints, position does exactly what engineers expect it to do.

Position is a powerful symbol, but it’s unforgiving. It exposes weak datum choices immediately — sometimes not until parts are already on the floor.

If you want position tolerances to work for you instead of against you, spend the extra time on the datum strategy. It’s almost always cheaper than chasing tighter numbers downstream.

If you’re interested in digging deeper into GD&T, tolerances, and how they translate to real CNC machining and inspection, the HPPI Knowledge Base goes into practical detail on these topics. And if you’re ever unsure how a datum scheme will play out on the shop floor, bringing manufacturing into the conversation early can save a lot of pain later.

Design intent only works when everyone is measuring the same part.

And if you made it this far, here are some more resources that may help on the datum and technical drawing side of things: 

• Anatomy of an Engineering Drawing
• 10 GD&T Best Practices from our Engineering Team
• Tolerances Cheat Sheet 📄
• Technical  Drawing Checklist 📑

Was this email forwarded to you? Sign up here to receive our monthly newsletter. 

We created Tolerance Stack to help engineers who design with microns in mind and understand that manufacturability is just as critical as innovation.

Whether you're optimizing a part for 5-axis milling, selecting materials for a Class III medical device, or navigating the complexities of GD&T, Tolerance Stack delivers the insights you need to make confident, production-ready decisions.