The master lead screw sets the pitch of every thread a lathe cuts, so its accuracy is the accuracy of every screw, micrometer and nut made afterward. It is built by cutting a thread on the best available lathe, measuring the travel of a nut along it against gauge blocks, and lapping with a split nut charged with fine abrasive until the pitch error is within +-0.01 mm per 100 mm, a target the shop must also be able to measure.
Part of Precision Tool Making
Making, measuring and correcting the screw that fixes thread pitch, and where its error budget goes.
The Lead Screw Is Where A Length Reference Enters The Machine
Flatness and angle can be generated from nothing. Length cannot; it must be copied from a reference. A lead screw is the point in the chain where that happens: its pitch, one fixed length per turn, repeats along its whole length, and every thread cut with it inherits that pitch. If the pitch is wrong by 0.1%, every screw is wrong by 0.1%, and every micrometer made from those screws reads 0.1% off.
The carrier’s recipe is: start with a ground and hardened steel bar, cut the thread on the best available lathe, measure the pitch error using gauge blocks and a dial indicator, correct by lapping with a split nut charged with fine abrasive run along the screw, re-measure and re-lap until the error is within +-0.01 mm per 100 mm. The loop it breaks, a lathe needing a lead screw and a lead screw needing a lathe, is broken by the lapping step, which corrects the screw instead of copying the lathe’s error.
What You Need
- The best lathe you have and a hardened steel bar (Lathe Fundamentals)
- A split nut that can be clamped around the thread, abrasive and light oil
- Gauge blocks for a stack of the length being checked (Gauge Blocks)
- A dial indicator on a stand, and a steady temperature
Method: Cut, Measure, Lap, Re-measure
Step 1: Cut the thread and harden the screw.
Step 2: Mount a nut on the screw and stand a dial indicator against the nut or carriage.
Step 3: Step the nut along the screw and compare each position with a stack of gauge blocks of the matching length. Record the difference at every 100 mm.
Step 4: Lap with a split nut charged with fine abrasive, running it along the screw so the high spots wear away.
Step 5: Re-measure, in several places along the screw and at one position over a full turn.
Step 6: Stop when the record shows the error inside the target, and keep the record.
Worked Example
What the error budget costs. Suppose the pitch error stays at +-0.01 mm per 100 mm, accumulating in proportion to length. A 25 mm micrometer screw inherits +-0.01 x 25 / 100 = +-0.0025 mm. A pitch error over one turn of a 0.5 mm pitch is 0.5 x (0.01 / 100) = 0.00005 mm, which is invisible; what matters is the accumulation, not any one turn.
Measuring it. To check 100 mm you need blocks totalling 100 mm. With block accuracy of +-0.002 mm each (the carrier’s lapping target), a stack of four 25 mm blocks can be off by up to 4 x 0.002 = 0.008 mm in the worst case. That is 80% of the +-0.01 mm allowance, so the checking stack nearly consumes the tolerance it is meant to verify. Two blocks of 50 mm bring the worst case to 0.004 mm. Use the fewest, largest blocks you have.
Heat. A 100 mm length of steel grows 0.012 x 0.1 = 0.0012 mm per degree (Precision Measurement). A temperature difference of 5 degrees between screw and blocks is 0.006 mm, 60% of the allowance. Screw and blocks at the same temperature expand together and cancel; a screw warm from the lathe measured against cool blocks does not.
Wear. A lead screw wears most where the carriage is used most, so the pitch is worse in the middle of its working range than at the ends. Measure at several positions, not only from one end.
Common Mistakes
- Taking the target as met without a record. The +-0.01 mm per 100 mm figure is the carrier’s; unmeasured, it is only a hope.
- Measuring warm. Heat looks like pitch error.
- Stacking many blocks. Their errors add.
- Testing only whole turns. An error that repeats once per turn, from the screw running off-centre, hides there (this is a general feature of screws, not a figure from the wiki).
The Tolerance Is One Source’s Figure And The Lapping Rate Is Unsourced
The tolerance of +-0.01 mm per 100 mm appears only in the carrier article, and no second source in this wiki confirms it. The claim that lapping with a long nut averages local pitch errors out of the screw is general workshop reasoning, not a line in the wiki, and the wiki gives no rate or number of passes. Treat the procedure as qualitative and rely on your own measurement record.
Safety
Lathe, abrasive and a screw that cannot be stopped from where you stand
Cutting the thread puts a spinning bar beside a tool, and the lapping nut runs on a rotating screw charged with abrasive; keep sleeves and hands clear, and know the stop and the coast-down. Abrasive on the screw gets into the air and onto the skin: lap wet where possible and see Dust and Respiratory Protection. A hardened screw is brittle; do not strike it.
What’s Next
- Learn next: Thread and Screw Standards - what the screw’s pitch is used for
- Combine with: Micrometer Construction - the instrument that inherits the screw’s error