First- or third-angle: a mirrored bracket and a foreshortened 48.5 mm dimension
A cast bracket is designed in Germany and machined at a shop in Ohio. The drawing shows a front view and, placed to the left of it, a second view; there is no projection symbol anywhere on the sheet. The Ohio shop read the left-hand view as a left-side view and machined the first-off part; the four Ø12 mounting holes came out on the opposite side to the CAD model, and the shop insists the drawing is wrong. Separately, the inclined web of the bracket is at 30° to the base and carries a 48.5 mm dimension placed in the front view parallel to the slope; the same view carries the 30° angle and a 42 mm overall horizontal dimension, which the shop measured as 42.0 on the print. A fillet is dimensioned R6 to the sharp corner where the web meets the boss, with no note. Twenty-five castings have already been drilled. The shop wants either a corrected drawing or written authority to rework.
1. Decide which projection system the drawing actually uses, explain how the view placement proves it, and state exactly what must be added to the sheet before any rework.
Read the worked answer
The drawing is first angle (ISO 128-30). In first angle the object sits between the observer and the projection plane, so the view obtained by looking at the right-hand side of the object is projected onto a plane behind it and, when unfolded, is placed to the left of the front view. The left-placed view is therefore the right-side view. The Ohio shop applied third-angle habit (left-placed view = left-side view), which mirrors every depth-direction feature - exactly the symptom with the Ø12 holes.
Independent checks on the sheet:
- In first angle the top view goes below the front view; in third angle it goes above. The position of the third view settles the question without any symbol.
- Section cutting-plane arrows must be consistent with the view directions, and hidden-line placement on the side view should agree with the front view's depth features.
What must be added:
- The ISO first-angle symbol (truncated cone, small end toward the observer, front and side views of the cone) next to the title block, plus the note 'Projection: first angle, ISO 128-30'.
- A general note that dimensions are in millimetres and the drawing is third-party-supplied, with a revision block entry recording the clarification.
- Since the part is handed, mark it 'right-hand' or 'left-hand' and add a distinguishing asymmetric feature with its own dimension so a mirrored reading is caught before machining.
Rework authority is only defensible after the corrected sheet is issued and the shop confirms receipt; the error is a projection-reading error, not a dimension error.
2. For the 30° web, which value is the true length, is the 48.5 mm dimension measurable as placed, and what should the sheet show instead?
Read the worked answer
If true length L = 48.5 mm at 30° to the base, the horizontal projection is L·cos30 = 48.5 × 0.866 = 42.0 mm - precisely what the shop measured. So 48.5 is the true length of the inclined face and 42 is its foreshortened projection in the front view.
A dimension placed in a view where the feature is foreshortened cannot be measured or inspected; the 48.5 as drawn is a non-inspectable dimension and will always be disputed at the bench. Three acceptable fixes:
- Best where the face carries holes or a machined pad: add an auxiliary view projected perpendicular to the inclined face (see the section case on auxiliary views) and place 48.5 there, with 30° remaining in the front view.
- Simplest where the face is plain: dimension the horizontal run 42 and the angle 30° plus the web thickness; the 48.5 is then obtained by calculation and should be shown as (48.5) reference, in parentheses.
- Rotation: a true-length dimension obtained by rotating the feature into the projection plane, clearly annotated, is acceptable only if the company standard permits it; many shops will not accept it for inspection.
Also verify that the 42 mm overall dimension is to the same datum as the 30° and that the 48.5 is not being used as a chain dimension - if it is, the tolerance stack changes between the two representations. The mirroring defect and the foreshortened dimension are separate faults and must be corrected in one revision.
3. The R6 fillet is dimensioned to the theoretical sharp corner with no note. Is that acceptable, and what does it change dimensionally?
Read the worked answer
Dimensioning a fillet radius with its leader pointing at the sharp theoretical intersection is common practice, but only if the drawing makes the convention explicit. As issued, the sheet shows a sharp corner with a radius leader and no note, which allows two readings: the corner is at the dimensioned position and the R6 is added outside it, or the arc is tangent to both faces and the theoretical corner is dimensioned.
The geometric difference is not trivial. For a 90° corner filleted R, the arc surface lies R(√2 − 1) ≈ 0.414R inside the sharp corner - with R6 that is about 2.5 mm of positional difference at the corner. On this bracket the fillet meets a boss whose position is presumably toleranced, so a 2.5 mm ambiguity can consume the whole positional tolerance and cause an interference with the mating part.
Correct treatment:
- Draw the theoretical sharp corner as a thin continuous line and the fillet arc as a thick continuous line; leader the R6 to the arc.
- Add the note 'Dimensions to theoretical intersections unless otherwise stated; fillets R6'.
- If the fillet is non-structural (a casting radius), put it in a general note and keep it out of the dimension chain entirely, so the machined faces are dimensioned to each other rather than through the cast surface.
- On the cast drawing, show the unmachined fillet as a casting radius with its own general tolerance; do not hold a machined datum to it.