What we are making: a second version. Almost nobody's v1 fits, and the
interesting work is turning that into a number.
How the part in your hand was actually made.
Do it — measure the error
Measure the printed part: overall size, hole diameters, hole spacing, face height.
Put them next to your drawn numbers in a table.
Work out the error for each, with a sign — plus or minus.
Look for a pattern. Are the holes all small by about the same amount? They usually are.
A real first print. Outside dimensions grow; holes shrink.
drawn printed error
overall length 40.0 40.3 +0.3
overall width 20.0 20.2 +0.2
barrel holes 16.4 16.05 -0.35
hole spacing 26.0 26.1 +0.1
face height 30.0 30.1 +0.1
💡 Tip
That pattern — outsides slightly large, holes slightly small — is the same on every machine of this kind, and it is why you add clearance to holes and not to outsides. You have just measured it yourself rather than being told.
Do it — fit it, and be honest
Push the sensor into the holes. Does it go in without forcing?
Does it sit square, or does it lean?
Do the pins have room behind, or are they pressed against the plastic?
Measure a wall at 30 cm, ten times. Once in the stand, does it give the same number every time?
Compare that with ten readings taken holding it in your hand. Which spread is wider?
💡 Tip
That last comparison is the whole unit in one measurement. A sensor in a stand you designed repeats; a sensor in a hand does not. You have just shown that a mounting is part of a measurement, not an accessory to it.
Do it — version two
Change only the dimensions that were wrong, by the amount they were wrong.
Do not redesign. Do not make it prettier. One change per measured error.
Save as team-N-sensor-stand-v2 and hand it in.
Write one line in your notebook: v1 was out by ___ on ___, so v2 changes ___.
📷
images/l2-e3-print-measure.png
Photo: callipers on a printed part, reading the barrel hole diameter
💡 Tip
That sentence is the deliverable. A team that prints a perfect part by luck has learned less than a team whose v1 was 0.4 mm out and who can say exactly what they changed.
End of Unit E
You measured a real object well enough to design around it.
You know the four rules: nothing floats, 2 mm walls, 45-degree overhangs, oversize holes.
You know orientation decides whether a part is strong.
You measured your own machine's error instead of guessing it.
And you turned that error into a specific change, not a redesign.
Challenge optional — only if you finish early
Print the same part twice and measure both. Are the errors identical?