Problem Description
A bad regrind ruins a gun drill faster than any cutting parameter mistake. I’ve sent drills out for resharpening and got back tools that couldn’t hold size for more than 10 parts. The tip geometry was wrong — the clearance angle was 2 degrees off and the drill rubbed instead of cutting.
Gun drills have complex geometry: tip angle, outer clearance angle, inner clearance angle, chip breaker height, and coolant hole positioning. A regrind service that treats them like standard twist drills will destroy the tool geometry in one pass.
What to Look For
I evaluate regrind services on four criteria: their inspection process, their tolerance capability, their experience with gun drills specifically, and their communication.
| Criterion | What I Check | Weight | Red Flag |
|---|---|---|---|
| Inspection | Do they measure every drill after regrind? | 30% | “We just sharpen and ship” |
| Tolerance | Can they hold +/- 0.5 degrees on tip angle? | 30% | No documented capability |
| Experience | How many gun drills do they regrind per month? | 25% | Under 50 drills |
| Communication | Do they call when a drill is worn beyond limits? | 15% | They sharpen anything sent |
I weight the criteria because not all are equally important. Inspection and tolerance account for 60 percent of my evaluation — a service that cannot measure cannot produce consistent results. Communication is weighted lower, but I still consider it essential. A service that regrinds everything you send without feedback is dangerous. I want a partner who tells me when a drill is not worth resharpening. A carbide gun drill has about five good regrinds in it before the carbide is gone.
I also look at the regrind equipment. The best services use a CNC tool and cutter grinder with a dedicated gun drill program. Manual grinding on a universal tool grinder depends too much on the operator’s skill for that day. I ask about the machine brand during the initial phone call — if they hesitate, they are likely grinding by hand.
Inspection Checklist
When I get reground drills back, I inspect every one before it goes into production. I use a 20x magnification comparator to check the tip geometry. I follow this checklist:
| Check | Acceptable Range | Tool | Failure Action |
|---|---|---|---|
| Tip angle | Within +/- 1 degree of spec (e.g., 117-119 for 118 deg) | Optical comparator | Return batch if over 10% out of spec |
| Outer clearance angle | 8 - 12 degrees | Optical comparator or gage | Reject individual drill |
| Inner clearance angle | 12 - 18 degrees | Optical comparator | Reject individual drill |
| Coolant hole clear | No grinding swarf visible | Compressed air blow-through | Reject drill, check batch |
| Tip diameter | Within 0.02 mm of nominal | Micrometer (0.001 mm resolution) | Reject individual drill, check grinder setup |
| Edge condition | No chips, burrs, or burn discoloration | Visual at 20x | Reject individual drill |
| Chip breaker height | Within 0.05 mm of spec | Depth micrometer | Return batch for adjustment |
| Margin condition | No damage on OD margin | Visual at 20x | Reject drill — margin damage causes oversize holes |
I reject drills that fail more than one of these checks. I have sent entire batches back when the coolant holes were packed with grinding debris. The worst batch I received had coolant holes packed solid in 60 percent of the drills — the regrind service was not cleaning out the grinding sludge before shipping.
I keep a reject log that I share with the regrind service monthly. The log tracks the number of drills rejected and the reason. If the reject rate exceeds 5 percent for two consecutive months, I schedule an on-site visit to the regrind facility.
Cost Comparison: Regrind vs. New
The economics of regrinding depend on the drill size and the number of regrinds you get. Here is what I see in practice:
| Drill Diameter | New Drill Cost | Cost per Regrind | Regrinds per Drill | Cost per Hole (reground) | Cost per Hole (new only) |
|---|---|---|---|---|---|
| 8 mm | $120 - $180 | $25 - $40 | 5 | $0.08 - $0.15 | $0.60 - $0.90 |
| 12 mm | $180 - $280 | $35 - $55 | 5 | $0.12 - $0.20 | $0.90 - $1.40 |
| 20 mm | $300 - $500 | $50 - $80 | 4 | $0.20 - $0.35 | $1.50 - $2.50 |
| 30 mm | $500 - $800 | $75 - $120 | 4 | $0.30 - $0.55 | $2.50 - $4.00 |
The math is clear — regrinding saves 75 to 85 percent of the tool cost per hole. But these numbers assume the regrind service delivers consistent geometry. A single bad regrind that causes a drill to break in the hole costs more than the entire batch of regrinds saved. I track this in my data collection and troubleshooting system, where each drill breakage event gets logged against the regrind batch number.
Communication Checklist
I expect the regrind service to communicate with me proactively. Here is what I put in the service agreement:
- Notification when a drill is worn below the minimum diameter for regrind (they call before sharpening it)
- Measurement data with every batch return — as-ground tip angle, diameter at tip, and clearance angle for each drill
- Photo of the tip at 20x for any drill that looks unusual during their inspection
- Monthly reject summary showing the number of drills they rejected before shipping
- Lead time notification if their backlog exceeds 3 business days
I have found that services willing to provide this level of communication also do better quality work. The act of recording and sharing the data makes them more careful.
Regrind Frequency
I track regrinds per drill with a serial number system. Each drill gets a unique ID stamped on the shank, and I record the number of regrinds and the material it’s run on. For a 10 mm carbide gun drill in 4140 steel, I typically get 5-6 regrinds before the diameter drops below minimum.
When a drill has been reground more than 5 times, the carbide thickness at the tip gets thin. The remaining carbide can’t handle the cutting loads and the tip chips on the first pass. I retire drills after 5 regrinds regardless of appearance.
| Regrind Number | Typical Stock Removed | Expected Tool Life (m of cut) |
|---|---|---|
| 1 | 0.3-0.5 mm | 100-150 m |
| 3 | 0.8-1.2 mm total | 80-120 m |
| 5 | 1.5-2.0 mm total | 50-80 m |
Building a Relationship
I send consistent volume — 20-30 drills per batch — to one regrind service and give them clear specifications. I provide a drawing with the tip geometry, diameters, and tolerances. Good services keep these specs on file and reproduce them every time.
I also ask for the measurement data with every returned batch. A record of the as-ground tip angle, diameter, and clearance tells me if the process is drifting. If I see the tip angle trend from 118 degrees to 117 degrees over four batches, I discuss it with them before it reaches 115 degrees.
I have a preferred regrind service and a backup. The backup gets used only when the primary is at capacity. I send the backup a small batch first (5-10 drills) and run them through my full inspection checklist before committing to regular volume. The backup relationship takes longer to develop, but having one prevents panic when the primary service has a lead time problem.
The regrind geometry interacts with coolant delivery at the cutting edge — I have covered this in the coolant additives overview. A drill with inconsistent clearance angles needs more EP additive to prevent edge wear, and I adjust the additive concentration based on the regrind quality data.
Key Takeaways
- I evaluate regrind services with a weighted scoring system: inspection and tolerance account for 60 percent of the decision, experience for 25 percent, and communication for 15 percent.
- I use an 8-point inspection checklist on every reground drill, including tip angle, clearance angles, coolant hole condition, tip diameter, edge condition, chip breaker height, and margin condition.
- Regrinding saves 75 to 85 percent of the tool cost per hole compared to buying new drills, but a single bad regrind that breaks in the hole costs more than the savings.
- Five regrinds is the maximum for carbide gun drills — after that the carbide at the tip is too thin to handle cutting loads.
- I track drill breakage events against the regrind batch number in my data collection system to identify problem batches fast.
- I maintain a primary and backup regrind service, with the backup qualified on a small batch first before committing to regular volume.
- A service that sharpens everything without feedback is dangerous — I expect proactive communication including measurement data and reject notifications.