7-Step Quality Checklist for Valeo Automotive Parts: Precision in Stamping, CNC & Forging

Who This Checklist Is For

If you're sourcing stamped, forged, or CNC-machined automotive parts from a supplier like Valeo—or evaluating a potential partner for a new program—this is for you. Specifically, it's for:

• Procurement engineers validating first-article inspections
• Quality managers setting acceptance criteria for high-volume runs
• Design engineers checking that production parts match the 3D model

I've been reviewing deliverables from metal forming suppliers for over four years—roughly 200+ unique part numbers annually across stamping dies, CNC brackets, and forged steering components. In 2023 alone, I rejected about 12% of first deliveries due to either dimensional drift or surface finish issues. Below is the checklist I now use on every new part. It has seven steps.

Step 1: Verify Material Certificates Against the Spec

This sounds obvious, but you'd be surprised how often a supplier ships parts with the wrong alloy. When I first started managing vendor quality, I assumed that if a supplier like Valeo listed a part as made of 6061-T6 aluminum, it would always be 6061-T6. Two years ago, a batch of UPF63R oil filter housings came in made from a less ductile 6005-T5 alloy—the material certificate showed the exact spec, but the packaging label said something else. We caught it only because I insisted on cross-referencing the heat number on the certificate against the one printed on the shipping pallet.

What to do:

• Request the mill test report for each batch. Match the heat number on the certificate to the physical parts.
• For steel parts, check the carbon content and tensile strength against the ASTM/AISI spec called out on your drawing.
• For aluminum extrusions (like those used in intercooler brackets for the Ecoboost Mustang), verify temper—T6 vs. T5—because it affects final strength.

One time saver: ask your supplier to pre-certify that all material certificates will follow a consistent format (heat number, date, mill name). Without that, you'll spend hours cross-referencing documents for every 10,000-part order.

Step 2: Dimensional Check — Focus on Critical-to-Function Features

Not every dimension on a drawing matters equally. A flange hole that clears a bolt? It has ±0.5 mm tolerance. But a bore that aligns a starter motor shaft in the Valeo starter motor catalogue? That's likely ±0.05 mm. I use a CMM (Coordinate Measuring Machine) for the latter and a simple go/no-go gauge for the former.

Three things to check in order:

1. Datums and reference points — If your part's 3D model uses Datum A, B, C, ensure the CMM program measures relative to those, not just a convenient edge.
2. Wall thickness variation — Common in deep-drawn stampings. A variation of 0.1 mm can cause a part to buckle during assembly. I've seen this on oil filter adapter housings (the UPF63R variant).
3. Surface profile — For forged parts like steering knuckles, use a contour gauge. Industry tolerance is usually ±0.25 mm (this was accurate as of mid-2024).

Step 3: Surface Quality — The Brand Perception Check

Here's where my initial mindset shifted completely. I used to think surface finish was purely cosmetic—something you'd only care about for exterior trim pieces. Then I ran a blind test with our engineering team: same forged bracket, one with a uniform matte finish (Ra 1.6 µm) and one with visible tool marks (Ra 3.2 µm). Over 70% of engineers rated the smoother part as 'more reliable' even though both met structural specs. The cost difference per piece was about $0.15.

For color-critical parts (like certain trim-level aluminum extrusions), we reference the Pantone Color Bridge guide. Industry standard color tolerance is Delta E < 2 for brand-critical colors—beyond that, even a layperson sees the mismatch. For functional parts that don't require color, I still check for burrs, scratches, and witness marks. A sharp burr on a CNC-machined mount for an Ecoboost Mustang intercooler can cut into a silicone hose within 500 miles.

Step 4: Functional Fit — Assembly Test

Dimensional data doesn't always predict assembly behavior. I've received parts where every critical dimension was within tolerance (±0.05 mm), yet they still wouldn't mate properly because of cumulative tolerance stack—three features that each bent 0.03 mm in the same direction. This is especially important for multi-part assemblies like wiper blade sets (the Valeo Silencio X.TRM 577849 set, for example, involves a blade, a spoiler, and mounting clip that must align within tight limits).

We now run a sample assembly test on every first article and on every 500th piece in production. It adds maybe 10 minutes per batch, but it has caught stacking issues that geometry reports alone missed.

Step 5: Packaging & Label Verification (The Overlooked Step)

Part of me thinks packaging is the most neglected quality gate. I have mixed feelings about spending extra on robust packaging—on one hand, it increases per-unit cost; on the other, a damaged packaging can let moisture or debris ruin 8,000 units in storage conditions. This happened to us in 2022 with a shipment of aluminum die-cast parts for water pump housings: the lack of desiccant caused surface oxidation that required manual rework.

What to look for:

• Label data matches the part number exactly (including revision level). A label that says 'UPF63R' but with a dash number that points to a different version is a red flag.
• Protection against corrosion—VCI paper or coating for ferrous parts; separate compartment for aluminum and steel to avoid galvanic corrosion.
• If the parts are destined for a just-in-sequence line, ensure the dunnage is compatible with your robot grippers or kitting stations.

Step 6: Documentation — Certificate of Compliance & Test Reports

No part ships without a Certificate of Compliance (CoC) that references the purchase order, part revision, and material spec. For safety-critical items (steering, braking, cooling systems), I also require a dimensional certificate showing CMM results for at least 5% of the batch. The Ecoboost Mustang intercooler brackets—those are safety-critical because a failure could cause a hose pop-off on a boosted engine. We request a CMM report for every 200th piece (this, I learned, is overkill for most parts, but for intercoolers it's justified).

One nuance: check the measurement date against the production date. If the CMM report is from three weeks before shipment, it might have been taken on a different batch. (Not that we've been burned by that, but it's happened at competitor plants.)

Step 7: Final Visual & Dimensional Sampling

The last step is a random sample pulled from the shipment—not from the production line. I take 5-10 pieces from different pallets and test them against the most critical dimensions. This catches two things: mix-ups in the final bin (a different part accidentally added) and handling damage during palletization. In Q2 2024, this sampling caught a single batch of stamped oil filter housings that had a 0.15 mm burr on the sealing face—the CMM data had passed, but the part had been scratched during packaging. That issue would have caused a leak on a UPF63R filter. The rework cost us about $0.20 per unit, which is far less than a field replacement.

Common Mistakes & Final Notes

Three mistakes I see repeatedly from buyers:

1. Skipping the material certificate cross-check — Relying on the supplier's word instead of verifying heat numbers. It's easy to do, but has cost me a $22,000 redo on one batch of forged tie rods.
2. Ignoring how parts will be installed — A water pump bracket that fits perfectly on the bench might not clear the timing cover when installed. If you're sourcing parts for a water pump replacement, always ask the supplier to dry-fit it to a reference jig that simulates the engine bay. Case in point: how to bleed a water pump properly on an Ecoboost engine—you need access to the bleed screw. If the bracket covers it, you've got a redesign on your hands.
3. Not setting acceptable quality limits (AQL) upfront — Many contracts say 'parts must meet spec,' but they don't define what sample size triggers a whole-batch rejection. I use an AQL of 1.0 for critical dimensions and 2.5 for cosmetic ones (based on ANSI/ASQ Z1.4). Without that, you end up arguing over a single out-of-spec part that maybe could be reworked—or maybe not.

One last thing: the checklist above is based on my experience up to early 2025. Metal forming technology—especially in progressive dies and CNC—evolves fast. The tolerance capabilities I saw from Valeo three years ago (circa 2022) have tightened since then. Always ask your supplier for their current CMM capability study, not the one from the last RFP cycle.