Hollow Bracket Wall Thickness After Shell
A designer models a cast-aluminum mounting bracket as a solid extrusion 180 mm long, 90 mm wide and 45 mm deep, then applies the Shell command with a uniform thickness of 4 mm and selects the two 90 x 45 mm end faces as removed faces. The part mass, checked in iProperties, is 18 percent higher than the target of 1.35 kg. The designer suspects the 4 mm wall is too thick but cannot simply reduce it, because the two M8 tapped bosses on the top face must retain at least 8 mm of local material and the part must still pass a 6 mm minimum-wall rule for the casting supplier. After changing the shell to 3 mm, the model fails to rebuild and Inventor reports a self-intersecting face near the internal fillet where the side wall meets the bottom flange.
1. Explain why reducing the shell thickness caused the rebuild failure, and describe a modeling approach that keeps the 6 mm rule while allowing the wall to be thinner.
Read the worked answer
The failure is geometric, not a units or tolerance problem. A uniform shell offsets every retained face inward by the thickness value. Where the original solid had a small internal corner radius, a 3 mm inward offset leaves only a few millimetres of material before the offset surfaces converge; if the original fillet radius is close to or less than the shell thickness, the offset surface crosses itself and Inventor cannot build a valid face. Reducing thickness did not create the problem - it exposed a corner that was always too tight for the offset.
Diagnosis steps: use Inspect > Zebra or the Curvature display on the failing region, and check the original fillet radius with Measure. If the radius is under roughly twice the intended wall, the offset will pinch.
Options, in order of preference:
- Enlarge the internal corner radius in the base sketch or the fillet feature so it exceeds the shell thickness by a safe margin, then keep the 3 mm shell.
- Split the shell into regions: use Split or a sketch-defined face boundary so the thin 3 mm wall applies only to the large panels, and model the boss region as a separate thickened pad. This preserves the 8 mm boss requirement without thickening the whole part.
- Model the wall as a revolve or sweep of a closed profile rather than a shell, giving explicit control of thickness at each station.
Verify against the supplier rule by sectioning the part and measuring the thinnest section, not by trusting the shell value alone, since draft and fillets change actual wall thickness.
2. The mass is still over target after the wall change. List the checks you would run before altering geometry again, and explain what each tells you.
Read the worked answer
Before removing more material, confirm the measurement itself is trustworthy.
- Material assignment: open iProperties > Physical and confirm the material is the correct aluminum alloy and that its density matches the supplier datasheet. A default generic material can be several percent off.
- Volume vs. mass: read the volume in the same dialog. If volume looks plausible but mass is high, the problem is density, not geometry.
- Override check: look for a manually overridden mass value, which silently ignores geometry changes.
- Feature suppression: confirm no cosmetic or construction solids are included. Check that the part contains only the intended solid bodies - a leftover body from a failed boolean adds mass invisibly.
- Boss and rib inventory: measure the actual boss wall and rib thickness. Ribs are a common hidden mass source because they are modeled thick for strength and never revisited.
- Draft and fillet contribution: large fillets add volume; confirm they are sized to function, not habit.
Only after these checks would I change geometry. If the mass is genuinely high, the highest-leverage change is usually reducing rib thickness or shortening a boss, because these are localized and do not affect the casting wall rule. Document the final measured minimum wall and mass in the model so the next revision starts from verified numbers.