How to Verify Aisin Automotive Parts: A Hands-On Checklist

If your job includes incoming inspection for automotive components, this is the checklist I wish someone had given me years ago. I'm a quality compliance manager at an automotive metal parts company. I review roughly 200 unique parts a month, and in Q1 2024 I rejected just over 6% of first deliveries because of spec drift. Not because the parts were crazy wrong, but because somebody skipped a step.

This checklist is for anyone who has to confirm that Aisin automotive parts - or any precision stamped and machined parts - are correct before they go on a line or into a customer shipment. It works for new vendor approval, first article inspection, and the occasional 'trust me, it's the same part' situation.

Who This Checklist Is For

Use it when:

• You're receiving prototype or pre-production parts.
• You're comparing a new supplier's sample against an approved supplier's part.
• You're putting an Aisin AZ6 transmission or a stamped strut tower into a build and need to verify the part before installation.
• You've been burned by a mismatch between the drawing and the actual part. (Happens more than you'd think.)

There are six steps. Step four is the one most people skip, and it's usually the one that catches the expensive surprise.

The 6-Step Verification Checklist

Step 1: Verify the Part Number Against the Drawing, Not Your Memory

I assumed a casting number was updated because the drawing revision was newer. Didn't verify the actual part. Turned out the supplier had used the older pattern. The number looked right at a glance, but the old pattern had a different boss height that interfered with the mating bracket.

So the first step is boring but critical: read the P/N on the part, the P/N on the label, and the P/N on the drawing. All three need to match. If there's a revision level, check it too. A part that matches a superseded revision is still a rejected part.

Step 2: Check Critical Dimensions with the Right Tool

For a stamped part like a strut tower, you don't get away with a caliper on the mounting face. The hole pattern, edge radius, and material thickness all need the right fixture. If your shop has a CMM, use it. If it doesn't, use a hard gauge that's been calibrated.

One thing I see a lot: people measure the easy things and skip the hard ones. A flange width is easy. The radius inside a deep drawn feature is hard. But the radius is where cracks start. So if the drawing says R6.0, don't accept R5.8 just because 'it looks close.' (Not that anyone would do that, but I've seen it.)

Step 3: Trace Material and Process Certifications

The most frustrating part of incoming inspection is the same issue recurring despite clear specs. You'd think written dimensional requirements would prevent mistakes, but interpretation varies.

So check the material cert against the heat number stamped on the part. Check the process certs - heat treat, plating, or coating. If the lot number on the cert doesn't match the lot number on the box, treat it as a red flag. 'Equivalent' without traceability is not enough.

This matters even more for transmission components. If you're sourcing an Aisin AZ6 transmission, the internal gears and synchros are heat treated. A missing cert means you don't know the hard case depth. That's a no-brainer reject.

Step 4: Fit-Test Before You Sign Off (the Step Most People Skip)

Looking back, I should have made fit-testing mandatory much earlier. At the time, we trusted the supplier's PPAP. It wasn't enough. The part fit on paper, but not in the actual assembly because the locating dowel was 0.2 mm off.

Fit-testing catches what dimensional measurement can miss:

• For a stamped strut tower, bolt it to the fixture with the exact fasteners you'll use in production. Hand tight isn't good enough.
• For an Aisin AZ6 transmission, if it's a replacement unit, check the input shaft spline and the bellhousing bolt pattern. Don't just look at the overall length.
• For any cooling system component, use a radiator vacuum fill kit to verify the system holds vacuum. A $30 kit can save you from a $300 comeback. If the kit cannot hold 25 in. Hg for ten minutes, either the part or the O-ring is leaking. That's a deal-breaker.

People skip this step because it takes time. But the time is way less than the cost of launching a line with a bad part.

Step 5: Check Packaging, Labels, and Lot Traceability

The part can be perfect and still get rejected if the label says the wrong thing. I still kick myself for not documenting a vendor's verbal promise to fix a labeling issue. If I'd gotten it in writing, we'd have had grounds to recover the cost of sorting 8,000 pieces.

Check:

• The label P/N matches the PO and the part marking.
• The lot number is legible and traceable to a production date.
• Packaging protects the part without causing freight damage. (Stamping edges can scratch themselves if they're loose in a box.)

Traceability isn't just paperwork. Per IATF 16949, identification and traceability must be maintained through production. If the supplier can't show that, the rest of the quality system is up in the air.

Step 6: Make the Call Based on Evidence, Not Hope

The hard part isn't deciding to reject a part. The hard part is deciding what to do when the deadline is tight and the deviation is 'small.'

In March 2024, we paid an extra $400 to rush a replacement from a reliable supplier. The alternative was accepting a batch where the zinc coating was slightly thin. The thin spots would have rusted in storage, and the customer would have seen it. $400 was nothing compared to the scrap and credibility we would have lost.

Uncertainty is a premium. You either pay for certainty now or pay for consequences later.

So when a part is questionable, don't use 'we need it now' as the tie-breaker. Use evidence. If the evidence doesn't support the part, the premium you pay for certainty is a better deal than the discount you get for doubt.

Common Mistakes I See

• Skipping the fit test because the first article measurement report looks fine.
• Using the same gauge for a new part that you used for an old one, without checking the drawing.
• Accepting a 'same as before' answer from a supplier without a new sample. 'Same as before' usually means 'same as last time I guessed.'
• Not photographing the part before you return it. Photographs make the reject reason concrete and help the supplier fix it faster.

A Note on 'Where Is Radiator Springs Located?'

Every week, our site gets a search for 'where is Radiator Springs located.' That's the fictional town from Pixar's Cars, not a real place. It's based on several Route 66 towns in California and Arizona.

Why does this matter in a quality checklist? Because it's a good reminder to verify location claims. A part can have 'Aisin' cast into it and still be a copy, or be a genuine Aisin part made in a different plant than the label suggests. Check the factory code. Check the country of origin. If the location on the label doesn't match the stamping code, ask before you accept it.

Bottom line: whether you're looking for a fictional town or a real transmission, don't assume. Verify the part, the spec, and the source. That's the only way to avoid the kind of mistake that costs a lot more than a rush fee.