First Angle vs Third Angle Projection Explained (2026)

First angle and third angle projection arrange the same orthographic views on opposite sides of the front view. In third angle the top view sits above the front view; in first angle it sits below. Nothing else changes: same geometry, same dimensions, same tolerances. The only reliable way to tell which one you are holding is the truncated cone symbol on the sheet, and the part of that symbol that matters is not the part most guides point at.

The short answer

Both methods are orthographic projection: the object is viewed square on, along axes at right angles, and each view is drawn true to scale with no perspective. What differs is where the resulting views are placed on the sheet.

Third angle projection imagines the projection plane between you and the object, as though the object sits inside a glass box and each face is drawn on the glass nearest to it. Unfold the box and every view lands on the side it was seen from. Look down at the part, and the resulting view goes above the front view.

First angle projection puts the object between you and the plane, so each view is projected through the object onto the plane behind it. Unfold that and every view lands on the opposite side. Look down at the part, and the resulting view goes below the front view.

The symbol, and what most guides get wrong

Every drawing that uses either method is supposed to declare it with a symbol, normally in or beside the title block. The symbol is two views of a hollow truncated cone: a pair of concentric circles, which is the cone seen end on, and a trapezoid, which is the cone seen from the side.

The symbol is not an arbitrary badge. It is a small drawing of a cone projected using the very method it names, so it demonstrates the convention rather than just labelling it.

If you memorise “circles on the left means third angle” you will be right about half the time and confidently wrong the rest, which is worse than not knowing. The taper is the invariant.

What each method does to the layout

The rear view is the one exception people trip on: it sits at the far end of the horizontal row in both methods, so it is not a reliable tell.

Why there are two conventions

The names come from the four quadrants formed by a vertical and a horizontal plane. Put the object in the first quadrant, above the horizontal plane and in front of the vertical one, and you get first angle. Put it in the third, below the horizontal plane and behind the vertical one, and you get third angle. The second and fourth quadrants are not used because the unfolded views would overlap.

Both survived because both were entrenched before anybody standardised anything, and because each has a defensible argument. Third angle is usually called the more intuitive of the two, since a view appears where you would expect it. First angle has its own logic for large assemblies, where the views unfold away from the object in the direction you walk around it. Neither argument is strong enough to have killed the other in a century of trying.

Why country is not a reliable guide

The usual rule of thumb is that first angle is European and third angle is American. As a prior it is fine. As a decision procedure it is not.

• First angle is the ISO default and normal across Europe, India, China and much of Asia.
• Third angle is standard in the United States and Japan, and is what ASME Y14.3 specifies.
• Mixed in practice: Canada and Australia, where published sources genuinely disagree with each other, which tells you something in itself.

More importantly, drawings travel and templates travel further. A CAD template copied from a parent company, a supplier detailing to their own house standard, a legacy drawing reissued without review: any of these puts a first angle sheet on an American desk or the reverse. That is precisely why both ISO and ASME require the symbol. The symbol exists because the country rule is not trustworthy.

How to tell in ten seconds

1. Look for the cone symbol, usually in or next to the title block. Read the taper. Away from the circles is first angle, toward them is third.
2. No symbol? Look for the words. Many sheets spell it out in the title block as THIRD ANGLE PROJECTION or as a projection field.
3. Still nothing? Find an asymmetric feature. A boss, a step, a counterbore, anything that exists on one side and not the other. Work out where that feature appears in the second view. If the feature you can see on the right of the front view appears on the right of the side view, you are in third angle.
4. Symmetrical part with no symbol? Then it does not matter for making the part, but ask anyway, because the next revision may not be symmetrical.

What breaks when it is read wrong

For a symmetrical part, nothing at all, which is exactly why the mistake survives long enough to get expensive. The failure mode is specific: the part comes out mirrored along the axis that the misread view controls.

• A step on the left of a shaft is turned on the right.
• A counterbore lands in the opposite face.
• A keyway ends up on the wrong end.
• A bracket is made as its own mirror image and will not bolt to anything.

Every dimension on the drawing was still correct, and the part still measures right. That is what makes it a quiet failure: inspection against the dimension list passes, and the problem only surfaces at assembly. It is the same category of error as a DXF that imports at the wrong scale, where the geometry is fine and one global assumption was not.

The defence is boring and it works: state the projection on every sheet, put the symbol next to the title block, and when you receive a drawing without one, ask before you cut. For the wider set of checks a sheet has to pass, how to read a technical drawing covers the view conventions and how to dimension a technical drawing covers what goes on them.

What the standards say

The point worth carrying away from the standards is the one they agree on: neither method is correct in preference to the other, and the obligation the standards impose is not to pick a side but to declare which side you picked. A drawing that states its method is unambiguous in either convention. A drawing that does not is ambiguous in both.

Where the pictures in this article came from

Worth being explicit, since it bears on whether you should trust them. The two diagrams, the symbol and the view grid, are drawn by hand as vector graphics. They carry the article's factual claims, so they are exact by construction rather than by luck.

The photographs and the drawings beside them are real output: a photograph of a part, and the dimensioned orthographic drawing our own pipeline projected from it. We built TechDraw AI to do exactly that, which meant deciding early which convention to emit and being able to produce either. If you want the route rather than the theory, JPG to CAD and image to DWG walk it through, and getting dimensions from a photo is honest about what a photograph cannot tell you: a flat image carries shape and proportion, the scale comes from one measurement you supply, and the projection method comes from the template you are drawing to.