Infill: strength versus cost

Stiffness versus infill, on a curve that flattens fast

Infill percentage and stiffness are related, but not in a straight line, and the shape of that curve is the whole argument for printing at 15% instead of 60%.

Going from a hollow part to a small amount of internal structure makes a large difference to how rigid a part feels. Going from already-dense to fully solid makes a difference most people cannot detect by hand, at roughly triple the material and a large increase in time.

The practical version: below about 15% the part feels weak and flexible. Between 15% and 30% is where most functional parts belong. Above 50% you are mostly buying mass.

Walls do more work than infill

The cheapest way to make a part stiffer is not infill at all — it is perimeters. Wall thickness contributes disproportionately to bending stiffness, because material at the outer surface resists bending more effectively than the same material in the middle. This is the same reason a hollow tube is stiffer than a solid rod of the same weight.

Raising wall count from two to four often does more for a functional part than raising infill from 20% to 60%, and it costs considerably less material and time. If you are only going to change one setting on a part that flexes, change the walls.

What the numbers look like

A 180 g model at 100% infill, with a quarter of that weight in walls and skins, a spool at $0.022 per gram, and machine time at $0.20/h:

Infill	Filament	Material cost	Approx. time	Machine cost	Total
10%	58.5 g	$1.29	5.4 h	$1.08	$2.37
20%	72 g	$1.58	6.0 h	$1.20	$2.78
40%	99 g	$2.18	7.4 h	$1.48	$3.66
70%	139.5 g	$3.07	9.6 h	$1.92	$4.99
100%	180 g	$3.96	11.8 h	$2.36	$6.32

Read the two ends together: moving from 10% to 100% multiplies material by 3.1 and total cost by 2.7, for a stiffness change that is dramatic at the bottom of the range and barely perceptible at the top. Print times are indicative and depend on your profile — use the infill cost calculator with your own slicer times.

Patterns matter as much as percentage

Two prints at the same infill percentage are not the same print. The main patterns behave differently under load:

• Grid and lines. Fastest to print, weakest in shear, and grid crossings can cause nozzle bumps.
• Gyroid. Isotropic in all three directions, printable without crossings, and good at absorbing compression. Usually the best default for functional parts.
• Cubic and triangle. Strong in the vertical direction, more material per unit of stiffness than gyroid.
• Concentric. Follows the shape, fast, poor in any direction other than in-plane.

If you are tuning a functional part, gyroid at 20–25% with an extra perimeter usually beats grid at 40% on stiffness, material and time simultaneously.

How to decide without guessing

A repeatable procedure that takes one afternoon and settles the question for a whole product line:

1. Print the part at your current settings and test it the way the customer will.
2. Keep the infill and add one perimeter. Test again.
3. If it is good enough, stop — you have just made the part cheaper.
4. If not, raise infill by 10 points and test again. Repeat once.
5. When it passes, record the settings next to the cost from the calculator so the next quote for that part is not a guess.

The point of the sequence is that you spend material only where testing says you need to, instead of building in a safety margin you cannot measure.

Important assumptions

The filament figures here use a simple approximation: the model's weight at 100% infill, multiplied by the shell share plus the remaining share scaled by the infill percentage. Real slicers add supports, brim and purge, and the solid share varies with wall count and geometry.

Print times are indicative. Use your own slicer's times in the calculator rather than these, because time estimates vary substantially between machines and profiles.