When a drawing calls for MQL machining instead of conventional flood cooling, the buyer's question shifts from "can you hit the spec" to "how will residual heat change my part between the last cut and the final go/no-go gauge." For small precision turned parts, that gap shows up as 0.005 mm dimensional drift and a surface texture that does not match what a coolant-lubricated run would produce.
MQL machining effect on tolerances
MQL delivers 2 to 10 ml/min of mist lubricant at the cutting zone, compared with 300 to 2000 ml/min in standard flood cooling. With less heat extraction, the workpiece runs warmer, and thermal growth on a 40 mm brass shaft can push an outer diameter 0.008 mm above nominal if the operator does not compensate mid-cycle. On ±0.005 mm bores in stainless steel bushings or copper insert nuts, the shift is smaller but still measurable once the part cools to ambient.
The practical fix is to state the acceptable thermal allowance in the RFQ. If the drawing already carries a 0.003 mm tolerance on a critical bore, ask the supplier whether MQL will be used and what in-process check they run before the part reaches room temperature. At Weeda Precision we note MQL-compatible materials and recommended check intervals on the inspection report so the buyer can verify dimensional stability without ordering a second CMM run on every lot.
MQL machining for small batch parts
Prototypes and first-article samples of 5 to 20 pieces are where MQL shines most. Lower coolant volume means shorter machine setup, no wastewater handling for a run that never scales, and a 2-day faster quote turnaround on a 50-piece sensor-housing batch in 6061 aluminum. The buyer gets a clean part with minimal coolant residue, which matters when the next step is anodizing or a press-fit into a medical device housing.
On repeat orders above 500 units per year the calculation changes. Tool wear under MQL accelerates on titanium and superalloys, so amortizing a carbide insert over 1,200 pieces versus 300 moves the per-part price. Specify annual volume and expected drawing revision cycle in the RFQ; if the drawing jumps from revision B to C mid-run, MQL tooling may need re-calibration that flood cooling would absorb without downtime.
Before sending an RFQ that references MQL machining, add one line to the cover sheet: material grade, the single critical tolerance zone, and whether the part will be 100% thread-gauged or sampled at 10:1. That sentence keeps the quote and the drawing on the same page and trims a week of back-and-forth.