IN01 · Reading a small length

Vernier Caliper & Micrometer

Why does a finer scale not automatically give a more accurate length? A finer scale resolves smaller changes, but a shifted zero moves every reading together. Check the closed jaws before measuring, look straight at the scale and repeat the contact. Repeating a biased reading does not remove its bias.

Slide a jaw or turn the thimble, read the matching graduations and check the zero. Measure a cylinder, a thin foil or a bore, then compare repeated readings.

Model assumptions

Rigid specimens and parallel measuring faces are held at twenty degrees Celsius. The caliper uses twenty vernier divisions over nineteen millimetres; the outside micrometer advances half a millimetre per turn, with fifty thimble divisions. Contact deformation, screw backlash, thermal expansion, jaw thickness and alignment errors are omitted. Scale-plane separation gives a geometric parallax shift. Seeded reading scatter is a teaching assumption, not a manufacturer specification. Standard uncertainty includes repeat scatter, a conservative resolution floor and the saved zero calibration; uncorrected parallax and later zero drift remain systematic errors. The blind specimen is synthetic. Apparatus proportions and scale close-ups are schematic.

Saved simulated measurements

Read, correct, repeat

Indicated length

12.35\,\mathrm{mm}

After zero correction

12.35\,\mathrm{mm}

Mean of this series

12.35\,\mathrm{mm}

Standard uncertainty

12.35\,\mathrm{mm}

Inspect a saved reading

Raw scale divisions and their frozen zero correction
No.	Main scale	Division	Zero	Corrected

Calibration, uncertainty & model

Repeated readings estimate scatter. A resolution floor and the saved zero uncertainty are added in quadrature. Viewing bias and a zero changed after calibration can still shift the mean; they are not hidden inside that uncertainty.

u_c=\sqrt{\frac{s^2}{n}+\frac{d^2}{12}+u_0^2}

The rigid specimen and parallel faces are at twenty degrees Celsius. Contact deformation, backlash, thermal expansion, jaw thickness and alignment errors are omitted. The bore uses ideal thin inside faces; the outside micrometer only measures outside dimensions. These synthetic records are not instrument certification.

Reveal the synthetic reference length

12.35\,\mathrm{mm}

The reference is separate from the recorded readings and is never used by the mean or uncertainty calculation.

Start with the essentials

Focus question: Why does a finer scale not automatically give a more accurate length?
One-sentence intuition: Resolution separates small changes; calibration checks where the scale starts.

Core mathematical model

Two scales with slightly different spacing

20d_v=19d_m,\qquad d_m-d_v=0.05\,\mathrm{mm}

The main scale has one-millimetre divisions. Twenty slightly shorter vernier divisions span nineteen millimetres. Each successive line gains another twentieth of a millimetre.

The screw turns angle into travel

L_{\rm indicated}=0.5N+0.01j\quad(\mathrm{mm})

Count complete half-millimetre sleeve intervals, then add the thimble fraction. Fifty thimble divisions make one turn. A half-millimetre sleeve mark matters as much as a whole-millimetre mark.

Remove a measured zero offset

L=L_{\rm indicated}-z_0

The closed faces provide a zero-length reference. Its indicated reading is saved separately for each instrument; changing the instrument later does not rewrite it.

Uncertainty of a repeated result

u_c=\sqrt{\frac{s^2}{n}+\frac{d^2}{12}+u_0^2}

The sample scatter s and count n give the repeat contribution. The scale division d retains a conservative quantization floor, and the last term describes the saved zero. Uncorrected viewing bias and later zero drift are outside this estimate.

Common difficulties

Five identical readings can still be biased

Typical misconceptionIf every reading agrees, the result must be exact.

Better mental modelAll readings share the same zero and viewing direction. The repeat scatter can vanish even when the mean is displaced. The resolution floor therefore remains.

Run the experiment

01
Find the matching line
Choose Read a cylinder, then Scale close-up. Open the jaw slightly and bring it back into contact.
What to observe: The main scale gives the whole millimetres. The green vernier line gives the remaining fraction. A reading can only be saved after contact.
02
Check a drifting zero
Increase Zero offset in Instrument errors & seed. Keep a reading, choose Close & check zero, then select the cylinder again and save another.
What to observe: The earlier reading keeps its old calibration. The later one subtracts the newly checked zero. More repeats alone cannot remove that shift.
03
Measure from the inside
Choose Open into a bore. Drag the bright jaw outward until it reaches the inner wall.
What to observe: The jaws stop at the bore walls. For an outside measurement, the contact stops a closing motion instead.
04
See why the viewing direction matters
Choose Look from the side and switch between Scale close-up and Parallax side view. Set Viewing angle back to zero.
What to observe: The physical opening stays fixed while the two separated scale planes shift in apparent alignment. Repeating from the same wrong angle preserves the bias.

Vernier Caliper & Micrometer

Read, correct, repeat

A matched line gives the fraction

Core mathematical model

Two scales with slightly different spacing

The screw turns angle into travel

Remove a measured zero offset

Uncertainty of a repeated result

Common difficulties

Five identical readings can still be biased

Run the experiment

Find the matching line

Check a drifting zero

Measure from the inside

See why the viewing direction matters

Related experiments and articles

How physics instruments measure