5-Step Quality Checklist for Automotive Metal Parts: A Dayco Inspector’s View

Who This Checklist Is For

If you're specifying or procuring automotive metal parts—stampings, forgings, extrusions, or CNC-machined components—you've probably run into the same problem I see every week: what looks good on paper doesn't always hold up on the line.

I'm a quality compliance manager at Dayco. I review roughly 200 unique part numbers a year, from prototype stampings to production runs of 50,000 units. In Q1 2024 alone, I rejected close to 12% of first deliveries because the spec didn't match what we agreed on. Not defects, exactly—but inconsistencies that would cause fitment issues downstream.

This checklist covers the five things I check on every order. It's not exhaustive, but it'll catch the problems I see most often.

Step 1: Verify the Material Cert—Not Just the Grade

Everyone checks the material grade. But I've learned never to assume that 'same grade' means 'same properties.' Last year, a vendor sent us DC04 deep-drawing steel for a progressive die stamping part. The cert said DC04. But the actual yield strength was 30% above spec—it work-hardened too fast and cracked in the second draw station.

What to do: ask for the mill test report (MTR) and compare actual values against your internal spec—not just the grade standard. Tensile strength, yield point, elongation: those numbers vary within grade ranges, and that variance matters for multi-step forming (stamping + coining + tapping, for example).

If the vendor hesitates on the MTR, that's a red flag. In my experience, the good ones send it without being asked.

Step 2: Confirm GD&T on the Die vs. the Part Print

Here's the one most people miss: the drawing tolerance applies to the part, but the die tolerance needs to be tighter. If your part print calls out ±0.2 mm on a hole location, the die should hold ±0.1 mm—or you'll drift out of spec as the die wears.

I assumed this was common practice. It's not. In 2022, we got a batch of 8,000 stamped brackets where the hole location was within print tolerance on every part—but the Cpk was 0.8. The vendor's die was built to the same ±0.2 mm tolerance as the part. Statistically, that's a recipe for scrap.

What to check: ask for the die design tolerance and compare it to the part tolerance. If they're equal, expect drift. For progressive dies with multiple stations, each station should have its own tolerance stack-up documented.

Step 3: Surface Condition Under Storage Conditions

This one cost us $22,000 in rework. We ordered 15,000 stamped steel brackets with a zinc-plate finish. Parts looked perfect on arrival. Three weeks later, in our warehouse (temperature-controlled, but with seasonal humidity swings), white corrosion spots appeared on about 30% of them.

The plating was technically within ISO 9227 salt spray spec—just barely. But 'passes salt spray under lab conditions' doesn't mean 'survives three weeks in a warehouse with condensation.'

What I do now: specify minimum plating thickness AND a humidity exposure test (ASTM D1748 or equivalent). I also ask for the vendor's storage protocol between plating and packaging. That's the part that's rarely documented.

Step 4: Hardness Consistency Across the Run—Not Just at Setup

For forging and heat-treated stamping parts, the first-piece hardness test is usually fine. The problem is consistency across the production run. In a 10,000-piece order, the furnace temperature can drift, quench medium can degrade, or the cooling rate can vary with ambient conditions.

Our protocol: we take hardness samples at the start, at 25% increments, and after any line stoppage. If the range exceeds ±3 HRC, the vendor has to investigate and re-sort the batch.

One vendor argued that ±5 HRC was 'within industry standard.' Maybe for decorative parts. For a suspension bracket that sees cyclic loading, ±5 HRC means some parts are borderline and some are out of spec. We sent the batch back. Now every contract has the ±3 HRC clause.

Step 5: Run a Fitment Trial Before Settling on Surface Finish

Surface finish is one of those specs where 'better' isn't always better. For aluminum extrusions that get anodized, a very smooth surface (Ra 0.4 µm or better) might look nice, but it can cause poor adhesion for the anodizing layer. We ran a blind test: same part with Ra 0.4 µm vs Ra 0.8 µm finish. About 85% of our assembly team preferred the visual of the smoother finish—until we showed them the anodizing peeling after 200 hours of salt spray.

What to do: specify finish based on function, not appearance. For stampings that get welded, a rougher finish holds weld primer better. For CNC-machined sealing surfaces, you want a consistent low Ra, but not mirror-polished (which can cause galling). Ask the vendor what surface finish they typically run for that material—they'll know the sweet spot for downstream processes.

Common Mistakes I Still See

A few things that slip through even with a checklist:

• Assuming first-article approval covers the whole run. It doesn't. Process drift happens. Schedule mid-run checks.
• Not checking the packaging spec. Parts arriving with edge dings because they shifted in transit—that's a packaging spec problem, not a part quality problem. (Should mention: we specify foam inserts for finished stampings with critical edges.)
• Over-relying on CMM reports. A coordinate measuring machine tells you dimensions. It doesn't tell you whether the part will assemble under torque, or whether the burr direction causes adjacent parts to bind. Use physical fitment gauges for critical interfaces.

To be fair, most vendors deliver good parts most of the time. The gap is in the edge cases—the high-volume run, the seasonal storage change, the subtle tool wear that accumulates over 50,000 hits. That's what this checklist is designed to catch.