Guide · Collets and collet nuts
Collet runout claims: TIR, measuring points and what studies show
How to read a collet's TIR claim: where makers measure it, why the measuring point matters, how runout is checked and what milling research reports.
We may earn a commission if you buy through links on this page. How we make money
Picks in this topic
- Genmitsu ER11 precision spring collet set, 7 or 15 pieces (1‑7 mm; 15‑piece adds 1/8 in and 1/4 in)Check price on Amazon
- Genmitsu ER20 router collet set, 14 pieces (1/16 in to 1/2 in)Check price on Amazon
- Genmitsu ER11‑A collet clamping nuts, 4‑pack (M14 thread)Check price on Amazon
Short answer
A runout (TIR) figure means little unless you know where it was measured. PreciseBits states its readings at the collet face and 1 inch down a calibration blank; REGO-FIX and Haas publish a single figure without a measuring point.[1] No source we found sets a "good" number for hobby routers, so compare only claims made the same way.
Picks in this topicSee the shortlist
- Genmitsu ER11 precision spring collet set, 7 or 15 pieces (1‑7 mm; 15‑piece adds 1/8 in and 1/4 in)Check price on Amazon
- Genmitsu ER20 router collet set, 14 pieces (1/16 in to 1/2 in)Check price on Amazon
- Genmitsu ER11‑A collet clamping nuts, 4‑pack (M14 thread)Check price on Amazon
Runout and TIR in plain words
Runout is how far a spinning tool wobbles off its true center. TIR (total indicator reading) is the swing a dial indicator shows over one full turn. A smaller number means the tool runs truer. The number also grows with distance from the collet, so a reading at the collet face and a reading 1 inch out are not comparable.
What makers claim, and where they measure
| Maker | Claim | Where measured |
|---|---|---|
| REGO-FIX, ER standard line | 10 um TIR | Not stated |
| Haas 04-1180, ER20 1/2 in | 0.0006 in TIR | Not stated |
| PreciseBits Precision-Grade | 0.0004 in at the face, 0.0008 in at 1 in | Collet face and 1 in down a blank |
| CMT ER20 precision | Two tiers, no published runout values | Not stated |
REGO-FIX lists a 10 um TIR for its ER standard collet line covering ER8 to ER50, without saying where it is measured.[2] Haas rates its 04-1180 ER20 1/2 inch collet at 0.0006 inch TIR; the page does not give the measuring location.[3]
PreciseBits grades its ER collets by TIR at the collet face and 1 inch down a calibration blank. Its Precision-Grade, for example, is 0.0004 inch at the face and 0.0008 inch at 1 inch.[1] CMT sells ER20 precision collets in a standard-precision and a high-precision version, but publishes no runout figure with a unit or method for either tier.[4]
All of these are the makers' own claims.
What the standards call for, per PreciseBits
PreciseBits states that DIN 6499 and ISO 15488 call for collet TIR to be read at the collet face and again at a distance of at least four times the clamping diameter.[5] We have not read the standards themselves. If a maker does not say where it measured, you cannot tell whether its figure follows that method.
How runout is checked
PreciseBits recommends checking combined spindle and collet runout with a dial test indicator on a smooth calibration blank, reading at the collet face and again at least 0.80 inch (20.3 mm) out.[6] This reads the whole stack: spindle bore, collet and nut together.
Do this only with the spindle power disconnected, as for any bit change.[7] Never hold an indicator against a powered spindle.
Stock router collets, as PreciseBits reported them
PreciseBits reports that in its checks around 2010 of low-cost routers and trimmers, stock collets and reducers measured about 0.003 to 0.010 inch TIR, while the motor bores were generally under 0.0003 inch.[6] That is one supplier's observation of a small, undocumented sample of older products. It is not a ranking of any current brand.
What milling research shows
The research we found studies metal and micro milling, not hobby wood routing. Read it as background on why runout matters, not as a rule for your machine.
- A 2019 Procedia CIRP paper reports that runout gives a tool's teeth unequal effective radii, so the teeth with larger radii carry more load and wear faster.[8]
- The same study notes that in finishing with very small chip thicknesses, such as finishing titanium alloys, runout can exceed the chip thickness so that at least one cutting edge removes no material.[8]
- A study of micro-milling brass with carbide end mills found that runout strongly distorted cutting-force signals, and that only one flute appeared to cut when feed per tooth was smaller than the runout.[9]
- A 2022 cutting-force model for micro end milling of AISI 4340 steel included tool runout, edge radius and flank wear, and reported good agreement with measured forces.[10]
- A 2017 finite-element study of micro end milling Al6082-T6 aluminum included measured radial runout and reported reasonable agreement with experimental chips and force-signal shapes.[11]
- A 2024 paper proposes a model that estimates micro-milling runout from cutting forces. Its authors conclude it does not yet work for runout calculation in its current form, so treat it as a method under development.[12]
- A 2007 paper in the International Journal of Machine Tools and Manufacture is titled "Runout effects in milling: Surface finish, surface location error, and stability."[13] We could read only its title.
None of these studies gives a runout limit for a hobby CNC collet.
Next
Questions
What is a good runout number for a hobby CNC collet?
No source we found sets one for hobby routers. Compare claims only when the makers state the same measuring method, such as PreciseBits' readings at the collet face and 1 inch down a blank.[1]
Are precision collets worth it?
Research on industrial and micro milling links runout to unequal tooth load and wear, but we found no controlled study of hobby wood routing. A precision collet is a maker's claim you can check on your own machine.[8]