How to Inspect Sandvik Coromant Cutting Tools: A QC Checklist for Carbide Insert Deliveries

Who This Checklist Is For

If you're the person who signs off on carbide insert deliveries before they hit the production floor, this one's for you. It's for quality leads, process engineers, and buyers who've been burned once by a wrong grade code and don't want a repeat.

I review roughly 40–60 incoming tooling batches a year at a precision CNC shop. In Q1 2024, we rejected three batches for marking and paperwork discrepancies that, if missed, would have caused a $22,000 rework on a mold project. This checklist takes about 30–40 minutes for a typical batch—or rather, closer to 45 on a day when the paperwork doesn't line up. It's the cheapest insurance we've found against bad tooling entering the line.

The same logic applies whether you're running a conventional CNC cell, a mold shop, or an additive manufacturing testing lab that qualifies specimens. If you accept metal-cutting consumables from a supplier, you have a provenance problem that needs a process, and the process is the same.

Step 1: Match the Paperwork Before You Open the Box

Start with the packing slip against your purchase order. Confirm the Sandvik Coromant article number, quantity, and batch/lot number. If the batch number on the box doesn't match the packing slip, stop there. That one number is your only way to trace the product back through the supply chain, and without it you're trusting a label.

If you asked for a material certificate—and you should for any critical job—check it against the batch number too. The certificate should identify the grade, the ISO designation, and the coating specification. A generic certificate without a batch link is a red flag, not a formality.

I learned never to assume this step is trivial. Last year we received a box labeled GC4225 with a certificate for GC4325 printed inside. Those are different steel-cutting grades. The packing slip and box label agreed with each other, so everyone almost missed it. Only the certificate caught it.

Step 2: Verify the ISO 1832 Designation String

Every Sandvik Coromant carbide insert for turning carries an ISO 1832 designation—the letter-and-number string that defines its shape, clearance angle, tolerance, chip former, and dimensions. So CNMG 120408-QM reads as: C-shape, N-class clearance, M-class tolerance, 12 mm cutting edge length, 4.76 mm thickness, 0.8 mm nose radius, QM chip former.

Per ISO 1832:2012, "Indexable inserts for cutting tools—Designation," the designation string is the international reference for insert geometry and tolerance classification. (Standard confirmed current at iso.org, January 2025.)

Verify the string against your PO before you measure anything. The most common error I see is the nose radius digit: someone orders -08 for finishing and receives -04, which changes surface finish and sits differently in the tool holder. The insert drops in fine. The result is poor. It's a thirty-second check.

At least, that's been my experience with turning insert orders in the 100–600 insert range. If you're buying at much higher volume, add a formal statistical sampling plan to this step.

Step 3: Read the Laser Marking—Don't Just Look at It

Turn off the overhead light and put an insert under a desk lamp at a shallow angle. Sandvik Coromant laser-marks the grade and designation onto each insert. It should be crisp, high contrast, and clear of the cutting edge. If the mark is faint, misplaced, or shows odd characters, treat it as a defect even if the insert measures fine.

It took me about two years and one expensive mold job to understand that the marking isn't branding—it's your traceability anchor. If you can't read the grade on the insert itself, you can't prove that the insert is what the certificate says it is. That's a rejection ground.

(And whether your supplier's marking line uses TRL laser versus CO2 laser doesn't change the acceptance criteria. The mark has to be legible under shop lighting at arm's length. The laser-type debate is a procurement distraction from the actual deliverable, which is a readable, traceable code.)

Step 4: Sample the Dimensions

Pull at least three inserts per batch—or 10%, whichever is higher—and measure:

• Inscribed circle diameter
• Thickness
• Nose radius (with a comparator or profile projector)
• Hole diameter for screw-mounted inserts

For M-class inserts, the commonly cited tolerance is roughly ±0.05 mm on the inscribed circle and ±0.13 mm on thickness. But the governing values are the ones on the manufacturer's drawing for that specific insert. The goal is not to enforce your own tighter tolerance; it's to catch gross deviations that reveal a wrong batch, a mislabeled box, or a counterfeit.

Counterfeit Sandvik Coromant products are a real and growing problem, which is worth keeping in mind when a deal looks too good to refuse. They typically fail on nose radius: a genuine CNMG 120408 measures 0.8 mm, while a fake often lands at 0.6 or 0.9 mm. Newer QC folks fixate on the IC diameter and thickness and skip the nose radius—which is, ironically, the most application-critical dimension for a turning insert.

Part of me wants to trust the authorized distributor entirely. Another part knows that the counterfeit market has improved faster than our trust did. So we verify, every time.

Step 5: Visual Inspection Under Magnification

Get a 10–30x stereo microscope and look for:

• Micro-chipping on the cutting edges
• Coating voids, peeling, or color inconsistencies
• Chip former profile consistency
• Grinding marks on the rake face

The edge condition is the first thing I check. A chipped edge you can't see with the naked eye will cut for half the expected time, then break catastrophically. We once rejected a batch of 8,000 pieces—different brand, not Sandvik—because every visible sample had micro-chips. The vendor argued "within industry standard." Their own drawing said otherwise.

The surprise wasn't the micro-chipping. It was how much batch-to-batch variance appeared on the rake face of otherwise "acceptable" inserts. Never expected that from a premium-brand distribution channel. Turned out the product had been reground by a third party after the original manufacturing step. Ask about reground or refurbished inventory if the price looks too good.

Step 6: Run One Test Cut

Paperwork, designation, marking, dimensions, coating—everything says the insert is right. Now prove it. Mount one insert and run a single pass on the same material and at the same parameters the application calls for.

Check the surface finish, chip form, and edge wear after a short run. Uniform flank wear is what you want. Chipping or built-up edge means something is off—grade, coating, or edge prep.

I once assumed "same specifications" from a trusted distributor meant identical performance across batches. Didn't test it, cut a corner on this step, and ran a 1,200-part job with inserts that broke every dozen pieces. That mistake cost roughly $15,000 in rework and a late delivery. Fifteen minutes of test cutting would have caught it before the first part. That's where I learned that five minutes of verification beats five days of correction.

Step 7: Log It in One Place, Every Time

Record each inspection in a single spreadsheet or tooling management system: date, supplier, batch number, article, measurements, deviations, disposition. Review the log quarterly.

Trend data is the actual point. You'll see which distributors reject at 2% and which at 7%. You'll notice when the same batch number reappears across different PO dates—a sign a distributor is offloading old stock. And when a vendor disputes a rejection, you can pull nearly three years of batch history and show them the pattern instead of having an argument.

I should add that this was the most skipped step for me too. In 2022, I set up our current verification protocol—a five-column spreadsheet—and it took about a month to make the habit stick. Since then, I estimate it has saved us around $8,000 a year in avoided rework.

Mistakes We Still See

Three things to keep in mind, drawn from deliveries that came across our dock this year:

• Counterfeits. Sandvik Coromant is one of the most counterfeited cutting tool brands on the market. If the price comes in 30–40% below typical market, treat that alone as a rejection criterion. Buy through authorized channels and verify batch traceability.
• "Same article number" isn't "same product." We've received the same article number from two authorized distributors with different batch numbers and notably different coating quality. Both passed the paperwork. One failed the test cut.
• Don't stop checking "trusted" suppliers. The delivery you skip the inspection on is the one that will be wrong—and it'll be on the longest-running, highest-volume job you have. Five minutes of verification beats five days of correction.

Add this checklist to your incoming goods protocol, and actually do it on the days you're busy. The 40-minute routine is cheaper than any rework ticket you'll ever write, and the first time it catches a bad batch, it pays for itself for the whole year.