GreatLight CNC Machining Factory logo
CNC Machining
Rapid Prototyping
Materials
Industries
News
About GL

Get Instant Quote

Thread grinding process guide

Internal Thread Crushes: How to Improve Grinding Efficiency

Internal thread grinding stalls for a handful of repeatable reasons. The wheel is too hard, the spindle runs too slow, the coolant never reaches the arc of contact, or the dresser is left too long between passes. This guide walks through the checks and settings we use on the floor to improve the grinding efficiency of internal thread crushes without giving up thread form. It is written for process engineers and shop supervisors running hardened nuts, aerospace fittings, and hydraulic bodies.

Hardness vs. wheel gradeSpeed and infeed rangesDressing cadenceCoolant delivery
Internal thread processing setup to improve the grinding efficiency of internal thread crushes
Quick answers

Key takeaways

Match wheel grade to hardnessOne grade softer than you think on hard threads. Too hard glazes and rubs.
Raise wheel speed firstGoing from 30 to 45 m/s often cuts cycle time more than raising infeed.
Coolant must hit the arcThrough-spindle or high-pressure delivery at 15 to 40 bar.
Dress on a count, not a feelingEvery 8 to 15 parts on a hard thread, sooner if form drifts.
Fix the fixture before the wheelRunout above 0.01 mm shows up as chatter on the thread flank.
Section 1

Why Internal Thread Grinding Loses Time

Internal thread grinding is a low-speed, small-arc operation. The wheel contacts only a short section of the thread flank, so the same grit stays in cut far longer than it would on an external cylindrical pass. That is why the same wheel that cuts freely on a shaft can load and rub inside a Ø20 mm nut. The first step to improve the grinding efficiency of internal thread crushes is to accept that contact geometry, not spindle power, sets the ceiling.

Most lost time falls into four buckets: wheel specification, spindle and work speed, infeed strategy, and coolant delivery. Dressing and fixturing follow close behind. In our shop, a single change on the first two buckets often moves cycle time by 20 to 35 percent on hardened 4140 and 17-4PH work. Nothing exotic is required. The gains come from setting values that match the hardness and the thread depth.

A useful baseline before any change: log actual cycle time, wheel wear per part, and surface finish on the flank for one full shift. Without that record you cannot tell whether a new wheel grade helped or whether the operator simply ran fewer roughing passes. Ten minutes of logging saves a week of guessing.

One warning applies to every item below. Do not change two variables at the same time. Grinding is a system. Move wheel speed, run 20 parts, then move infeed. Otherwise the data is noise and the operator stops trusting the numbers.

  • 1
    Contact arc is shortHeat and chips concentrate in a small zone.
  • 2
    Wheel grade dominatesIt decides whether grit stays sharp or glazes.
  • 3
    Log before you changeCycle time, wear, and finish for one shift.
Section 2

Wheel Specification and Grade Choice

For hardened steel above 45 HRC, start one grade softer than the wheel you would pick for external grinding. Aluminium oxide in a vitrified bond is still the workhorse. For 17-4PH and other stainless grades, a white or pink alumina grain with a more open structure sheds chips instead of packing them into the pores. Cubic boron nitride earns its cost when the batch is large and the thread is deep. It holds form for many more parts and cuts the dressing time that dominates a short cycle.

Grain size is the next lever. A 60 to 80 grit wheel gives a better finish on the flank but removes stock slowly. A 46 grit wheel cuts faster and leaves a finish around Ra 1.6 to 3.2 μm, which is usually fine for a roughing pass. Use a two-wheel strategy when the thread tolerance is tight. Rough with 46 grit, then finish with 80 to 100 grit at a lighter infeed to reach Ra 0.8 to 1.6 μm.

Bond hardness is where many shops go wrong. A wheel that is too hard will not self-sharpen. The grit dulls, the wheel glazes, and the operator compensates by pushing harder. That raises force, pushes the thread off size, and burns the flank. If you see a shiny wheel face and blue marks on the thread, go one or two grades softer before you touch any speed.

Wheel diameter and width need to match the minor diameter of the thread. Too wide a wheel overloads the contact zone on a small bore. Too narrow a wheel means more passes and more wear per part. Keep the wheel width within roughly 60 to 80 percent of the thread length you are grinding in one pass.

  • 1
    Hardened steel above 45 HRC
  • 2
    Stainless and 17-4PH
  • 3
    Long runs, deep threads
Section 3

Speeds, Infeed, and Pass Strategy

Wheel speed is the cheapest gain available. Many internal grinders run at 28 to 32 m/s because that is where they were set years ago. Moving to 38 to 45 m/s thins the chip, lowers force per grit, and usually improves finish at the same time. Check the wheel rating first. Vitrified wheels are commonly rated to 50 m/s or higher, but the spindle and the quill must be rated for the same speed before you push it.

Work speed for internal thread grinding sits far below external grinding. A common range is 15 to 30 m/min for hardened steel, with harder material at the low end. Running the work too fast on a small bore builds heat faster than the coolant can remove it. The thread grows, the gauge reads oversize, and the operator chases the size down all shift.

Infeed strategy matters more than total depth. Take 60 to 70 percent of the stock in two or three roughing passes at 0.02 to 0.04 mm per pass, then finish with two light passes at 0.005 to 0.01 mm. A single heavy pass raises force, deflects the quill, and produces a tapered thread. Spark-out at the end of the finish pass, two to four seconds with no infeed, lets the wheel spring back and cleans up the form.

Watch the spindle load meter, not just the sound. A steady load near 70 to 80 percent of rated power with a clean finish means the settings are working. A load that climbs part after part means the wheel is dulling and the dressing interval is too long.

  • 1
    Wheel speed
  • 2
    Work speed
  • 3
    Rough then finish
  • 4
    Spark-out
Section 4

Coolant Delivery and Dressing Cadence

Coolant has to reach the arc of contact, and inside a thread that is hard. Flood coolant from the side rarely gets there. Through-spindle delivery, or a high-pressure nozzle aimed at the entry of the thread, at 15 to 40 bar makes a visible difference on deep bores. Oil-based coolant handles heavy stock removal and gives better lubrication on the flank. Water-based coolant cools better and is easier to clean, which matters when the part goes to inspection.

Filter the coolant and keep the concentration in range. Fine swarf from a thread pass recirculates fast and scratches the flank. A 10 to 20 μm filtration step pays for itself on parts with a Ra 0.8 μm requirement. Check concentration weekly and top up with the same product, not a different brand. Mixed coolants separate and leave residue in the thread root.

Dressing controls both form and cutting behavior. On a hard thread, dress every 8 to 15 parts. On softer material, 20 to 30 parts may be fine. Use a small depth per dress pass, 0.01 to 0.02 mm, with two or three spark-out passes at the end. A heavy dress breaks the grain too early and the wheel wears unevenly across the thread form.

Track the number of dress passes against the parts produced. If the count drops while the wheel still looks sharp, the thread form is probably drifting, not the dressing. Gauge a part at the front, middle, and end of the run to catch a taper before the whole batch is scrap. On a 1,000-part order, that check takes seconds and protects the run.

  • 1
    Pressure
  • 2
    Filtration
  • 3
    Dress depth
Do this in order

Seven Steps to Improve the Grinding Efficiency of Internal Thread Crushes

Run one change at a time and keep 20 parts between changes.

  • 1
    Log the baselineRecord cycle time, parts per dress, wheel wear, and flank finish for one full shift before touching any setting.
  • 2
    Check fixture runoutMeasure TIR at the work spindle and the part. Above 0.01 mm, fix the fixture before changing the wheel; chatter will mask every other gain.
  • 3
    Pick the wheel gradeFor steel above 45 HRC, go one to two grades softer. For stainless, switch to white or pink alumina with an open structure.
  • 4
    Raise wheel speedMove from 30 to 38 or 45 m/s if the spindle and wheel ratings allow. Run 20 parts and recheck finish.
  • 5
    Reset the infeed planTwo to three roughing passes at 0.02 to 0.04 mm, then two finishing passes at 0.005 to 0.01 mm with a two to four second spark-out.
  • 6
    Fix coolant deliveryAim high-pressure coolant at the thread entry at 15 to 40 bar. Filter to 10 to 20 μm and keep concentration in range.
  • 7
    Set a dressing countDress every 8 to 15 parts on hard threads, 0.01 to 0.02 mm per dress pass. Gauge front, middle, and end of run for taper.
Settings reference

Starting Settings by Material and Thread Size

Ranges are starting points, not guarantees. Adjust for your wheel and spindle rating.

Material and sizeWheel speedWork speedDress interval
4140, 45 to 55 HRC, M12 to M2038 to 45 m/s18 to 25 m/minEvery 10 to 15 parts
17-4PH, M10 to M1635 to 42 m/s15 to 22 m/minEvery 8 to 12 parts
Hardened tool steel, above 58 HRC30 to 38 m/s12 to 18 m/minEvery 8 to 10 parts
303 or 304 stainless, M16 to M3040 to 45 m/s20 to 30 m/minEvery 20 to 30 parts
Inconel, small bore28 to 35 m/s10 to 15 m/minEvery 6 to 10 parts
Aluminium or brass, coarse thread45 to 50 m/s25 to 35 m/minEvery 30 to 40 parts
FAQs

Questions We Get From Process Engineers

Can I use CBN on a small internal thread?

Yes, if the bore and the wheel width allow it. CBN holds form far longer than alumina, so dressing time per part drops. The trade-off is cost and the need for a stiffer spindle. On a short run of 20 parts, alumina is usually cheaper overall. On a 2,000-part release, CBN often wins.

Why does the thread size drift during the run?

Size drift almost always comes from heat, not from the wheel. Work speed too high, coolant not reaching the arc, or a dull wheel all raise temperature. The part grows, then shrinks after cooling and the gauge reads oversize. Cut work speed by 20 percent and check coolant pressure before you touch the infeed.

How do I know the wheel is glazing?

Look at the wheel face after a pass. A glazed wheel is shiny and the pores look filled. The thread flank may show blue tint or a polished, smeared finish. Go one or two grades softer, or shorten the dressing interval. Raising speed alone will not fix glazing.

Is high-pressure coolant worth the retrofit?

On bores deeper than about 1.5 times the diameter, yes. Flood coolant rarely reaches the contact zone there, so heat builds and the wheel loads. Through-spindle or high-pressure delivery at 15 to 40 bar is the single biggest coolant gain we see on deep internal threads.

What tolerance can internal thread grinding hold?

On the equipment we run, we hold ±0.005 mm on thread dimensions with a finish of Ra 0.2 to 0.8 μm when the process is set up for fine finishing. Coarser roughing passes land between Ra 0.8 and 1.6 μm. Both are checked before shipment.

Should I grind the thread before or after heat treatment?

Grind after heat treatment when the thread tolerance is tight or the material distorts during hardening. Grinding before hardening risks losing the form in the furnace. If you must grind before, leave 0.10 to 0.15 mm of stock and plan a finish pass after hardening.

Send Us Your Thread Drawing

Upload the part and we will return a quotation with a free DFM review within 12 hours. Grinding and machining runs from one prototype to 10,000+ parts, with 100% inspection before shipment.

12-hour quoteNo minimum order100% inspection

Follow our work

More From the Shop Floor

We publish setup notes, tooling trials and inspection data from the factory floor.

FacebookTikTokYouTubeLinkedInInstagramThreadsPinterest

Trusted by engineers and manufacturers worldwide

Tesla Ford Motor Company BYD Auto Denso Magna International Boeing Airbus Medtronic KUKA FANUC