Why a Buyer Needs to Understand GD&T
A sourcing professional does not need to be able to author a tolerance stack-up, but a print that is misread at the quoting stage produces two predictable outcomes: an unrealistically low bid that later becomes a change order, or a part that passes a supplier's own inspection and still fails at final assembly. Geometric Dimensioning and Tolerancing (GD&T) is the language that is supposed to prevent both outcomes, but it is also one of the most commonly misquoted and misinterpreted parts of a technical drawing. Buyers who can recognize the structure of a feature control frame, understand what a datum actually constrains, and know which standard (ISO or ASME) governs the drawing are far better positioned to ask the right questions before a request for quotation (RFQ) goes out, rather than after parts arrive.
Datums: The Reference System Behind Every Tolerance
A datum feature is a real, physical feature on the part — a flat face, a bore, a pin — that is selected because it plays a functional role, typically because it locates the part in an assembly, in a fixture, or against a mating part. The datum itself is the theoretically exact plane, axis, or point derived from that feature; it exists only in the idealized geometry, not on the physical part.
Most functional tolerances are meaningless without a stated reference frame. A hole position toleranced "loosely" relative to the wrong reference system can still cause an assembly failure even though it measures within tolerance in isolation. This is why feature control frames commonly reference not one datum but a hierarchy of up to three: primary, secondary, and tertiary. The primary datum removes the most degrees of freedom (typically it stabilizes the part against a plane, removing three), the secondary further constrains rotation and one additional translation, and the tertiary locks the last remaining degree of freedom. This ordered set is called a datum reference frame (DRF).
For a buyer, the practical takeaway is this: the order of datums in a feature control frame is not cosmetic. [A|B|C] and [B|A|C] describe two different inspection setups and, potentially, two different accept/reject outcomes for the same physical part. If a supplier's quote assumes a different datum order than the print specifies — often because a legacy fixture was reused — parts can be built to a self-consistent but incorrect reference frame.
Buyers should also be alert to datum targets (points, lines, or areas used instead of an entire surface, common on castings and weldments) and to datum feature simulators, the physical gauge elements (a flat granite surface, an expanding pin) that a metrology lab uses to simulate the theoretical datum during inspection. A print that calls out a datum feature which has no practical way to be simulated on the supplier's available equipment is a fixture and CMM programming problem waiting to surface after tooling is cut.
Feature Control Frames: Reading the Box
The feature control frame (FCF) is the rectangular box that carries the actual tolerance instruction. Reading left to right, it typically contains:
The geometric characteristic symbol — one of fourteen symbols in ASME Y14.5 grouped into form (flatness, straightness, circularity, cylindricity), orientation (perpendicularity, parallelism, angularity), location (position, concentricity, symmetry), profile (of a line, of a surface), and runout (circular, total).
A diameter symbol, if the tolerance zone is cylindrical.
The tolerance value, sometimes followed by a material condition modifier.
Datum references, in precedence order, each with its own modifier if applicable.
A frame such as ⌖ ⌀0.25 Ⓜ | A | B Ⓜ | C is read: "the true position of this feature must fall within a cylindrical zone 0.25 mm in diameter, at maximum material condition, relative to primary datum A, secondary datum B (also at maximum material condition), and tertiary datum C." Every symbol in that string changes either the size of the tolerance zone, the reference frame, or both.
For a buyer reviewing a quote, the FCF is where three practical questions live: what is actually being measured, against what reference, and how large is the effective tolerance zone once modifiers are applied — because the last of these is very often larger than the printed number suggests.
Material Conditions and Bonus Tolerance
This is the area most likely to be misquoted, because it directly affects cost through inspection method and effective tolerance.
Regardless of Feature Size (RFS) is the default condition in ASME Y14.5 (and requires no symbol). The stated tolerance applies at whatever size the feature actually is, with no additional allowance.
Maximum Material Condition (MMC), symbol Ⓜ, applies the tolerance when the feature contains the most material (smallest hole, largest shaft). As the feature departs from MMC toward its least-material size, it earns bonus tolerance equal to that departure — the position tolerance effectively grows. MMC is used when the design intent is strictly assembly (mating parts fit together), because the bonus tolerance mathematically guarantees clearance is never worse than at the worst-case condition.
Least Material Condition (LMC), symbol Ⓛ, is the inverse, used less frequently, typically to protect a minimum wall thickness or edge distance.
The commercial consequence: a feature toleranced at MMC can, for an oversized-hole/undersized-pin combination, legitimately accept far more positional deviation than the nominal callout implies. A supplier or buyer who quotes to the nominal number without accounting for available bonus tolerance is quoting to a tighter (and more expensive) requirement than the print demands. Conversely, a quality department that rejects a part for "exceeding position tolerance" without checking actual feature size against MMC may be issuing a false reject. Functional gauging — a go/no-go gauge built to worst-case boundary conditions — is the classical, and often most cost-effective, inspection method for MMC callouts, but it requires the gauge itself to be specified and budgeted at the RFQ stage, not discovered afterward.
ISO GPS vs ASME Y14.5: Where the Standards Diverge
ISO Geometrical Product Specifications (GPS) and ASME Y14.5 share common ancestry and, since the 1990s, have converged substantially, but they are not interchangeable, and a drawing that mixes conventions silently is a common source of dispute.
Default interpretation. ASME Y14.5 treats RFS as the implicit default for size-related tolerances of position and orientation; a modifier must be explicitly added for MMC or LMC to apply. ISO GPS (per ISO 8015 and ISO 2692) takes the opposite default: the independency principle applies unless otherwise stated, meaning size and geometric tolerances are treated as independent of each other unless the envelope requirement (Ⓔ, per ISO 14405-1, roughly analogous to ASME's Rule #1 / Taylor principle) is explicitly invoked. This single difference in defaults is enough to change acceptance criteria on an identical nominal print if the wrong standard is assumed.
Datum reference conventions. ISO 5459 defines datum systems with similar structure to ASME's DRF, but ISO distinguishes more granularly between common datums, datum systems, and mobile/movable datum targets, and uses different symbology and modifier placement in some cases. ISO drawings also more commonly use the term "datum system" where ASME uses "datum reference frame."
Profile tolerancing and general tolerance blocks. ISO GPS makes heavier use of general geometric tolerances applied via a title-block reference to ISO 2768 (for parts without individual GD&T callouts on every feature), a mechanism with no direct ASME equivalent — ASME relies on the assumption that ungoverned features are controlled by explicit general notes or block tolerances that must be separately defined by the drawing's originating company.
Symbol and terminology differences. A small but consequential set of naming differences exists: ASME's "runout" and ISO's use of largely equivalent circular/total runout symbols align well, but terms like "regardless of feature size" have no separate ISO symbol because RFS is the ISO default rather than a stated modifier. ASME's "basic dimension" (boxed dimension defining exact theoretical location) corresponds to ISO's "theoretically exact dimension" (TED), same concept, different label.
Governing document callout. Both systems expect the drawing to state which standard governs, typically in or near the title block (e.g., "TOLERANCING PER ASME Y14.5-2018" or "PER ISO 1101"). A print with no such statement, or with legacy notes referencing a superseded revision (Y14.5M-1994 vs Y14.5-2009 vs Y14.5-2018; ISO 1101:2004 vs ISO 1101:2017), leaves the interpretation of defaults ambiguous, and a supplier is entitled to interpret ambiguity in whichever direction is cheapest to produce.
Questions to Resolve Before Issuing the RFQ
Before a drawing package goes to suppliers, a buyer or sourcing engineer should confirm the following, ideally in writing so the answer becomes part of the quote basis rather than a post-award argument:
Which standard, and which revision, governs the drawing? Confirm the title block statement matches what the supplier's quality system is actually certified and equipped to inspect against.
Is the datum reference frame achievable with the supplier's fixturing? Ask whether the supplier has, or can build, gauges or fixtures that simulate the specified datum features in the specified order.
Are material condition modifiers present, and does the supplier's inspection method exploit or ignore bonus tolerance? A supplier quoting CMM-only inspection may not be capturing legitimate bonus tolerance that a functional gauge would allow, inflating cost or scrap rate unnecessarily.
Is there a default-condition assumption embedded in the print (RFS by ASME default, or independency by ISO default) that the supplier's home standard would interpret differently?
Are general tolerances defined, either via an ISO 2768 class callout or an equivalent house-standard note, for any feature not individually toleranced?
What inspection equipment and reporting format will substantiate conformance — full CMM report, first-article inspection (FAI) per AS9102 or equivalent, functional gauge certificate, or sampling plan — and is that method capable of actually verifying the geometric characteristics called out (some characteristics, like profile of a complex surface, are effectively impossible to verify manually and require CMM or optical scanning)?
Do datum features exist in a practical, accessible location on the finished part, or will finishing operations, coatings, or fixturing obscure or remove the surfaces the datums depend on?
Are there any composite or multiple single-segment feature control frames (pattern tolerancing) whose upper and lower segments carry different datum precedence or looser tolerance — a frequent source of underquoting because the second segment is overlooked?
Resolving these questions before RFQ shifts interpretation risk from the inspection floor, where it is expensive and adversarial, to the quotation stage, where it is a matter of clarification.
Practical Implications for Sourcing
The commercial risk in GD&T is asymmetric: a tolerance interpreted too tightly inflates the quoted price and may eliminate capable suppliers unnecessarily, while one interpreted too loosely produces parts that pass a supplier's own inspection and still fail at assembly, generating cost far downstream of the purchase order. Both risks are reduced by the same practice — treating the feature control frame, the datum structure, and the governing standard as commercial terms to be confirmed explicitly, not technical details to be assumed identical across every supplier and every standard.