The wall can be grandfathered. The receptacle I ship into it can’t

The thing I had backwards is the thing a lot of buyers have backwards. An existing non-TR receptacle sitting in an old wall is usually fine to leave alone — older installations get grandfathered, and nobody’s forcing a rip-out. So I’d assumed “old building, replacements, no big deal, send standard.”

Replacement is a different question from leave-it-alone. Under NEC 406.4(D)(5), when a receptacle is replaced at a location that requires tamper-resistant receptacles, the replacement has to be a listed TR type. The grandfathering protects the old device that stays in the wall. The moment I ship a new one to go in its place, that new one follows current code for that location. My carton wasn’t being judged against 1990; it was being judged against the edition the inspector was holding.

There’s one exception I now keep in my head, because it’s the one that actually saves money: a non-grounding receptacle replaced like-for-like with another non-grounding receptacle is not required to be TR, under 406.4(D)(2)(a). That’s a narrow case — old two-prong, no ground, swapped for the same — and I don’t blur it with a normal grounding-type replacement. Everywhere the location is on the 406.12 list, the replacement is TR. Everywhere the location is on the 406.12 list, the replacement is TR.

“15A duplex” is not a finished line

What I learned the expensive way is that a replacement line carries more than amperage and a face style. Each of these is a separate “must-be,” and one doesn’t buy another. The table is the checklist I now run before I send an order up to my supplier or accept one from a customer downstream.

The line I wrote named two fields and left four to chance. On a location that required TR, “left to chance” got answered at the jobsite, and the answer came back to me.

A carton of non-TR units can be electrically perfect and still be the wrong stock

Here’s the part that keeps me careful now, because it’s where my money sits. The twenty thousand standard units I had were fine. Listed, working, nothing wrong with them electrically. That was exactly the trap: a good product felt like the right product, and it wasn’t.

A non-TR carton aimed at a TR-required replacement is the wrong SKU, not a defective one. And wrong-SKU stock is the worst kind to hold, because it’s hard to push in either direction. My supplier won’t take it back as defective — there’s no defect. My downstream customer won’t take it for the job — it won’t pass. It just sits on my shelf as money I already spent. The cheap non-TR lots that float around the market are tempting for exactly the reason they’re risky: the price looks right because the SKU is the wrong one for where TR is now required.

So at intake I read the carton, not just the invoice. If my supplier marks to my own SKU, I check it against what the order actually needs. If there’s no SKU and they default to a full product name on the box — something like “15A TR Standard Receptacle” — I read that line, because that’s where TR either is or isn’t, and the box is the last place I can catch it before it’s mine.

When a customer wants to return a TR outlet that’s “hard to plug into”

The other return I see runs the opposite direction: a downstream customer pushes back a TR receptacle because a plug won’t go in, and calls it defective. Before I accept that one back, I check the mechanism, because the resistance is the feature.

A TR face has spring-loaded shutters that only open when both slots are pressed at the same time, which is what a normal plug does and what a single object can’t. A blade going in at an angle, a bent prong, or pressure on one slot won’t open it — that’s the shutter doing its job, not failing at it. The fix is usually a straight, even insertion. A unit that won’t take a properly aligned plug, with both blades square, is the one that’s actually wrong, and that’s the one I’ll route back up to my supplier and ask for their plug-insertion test record on. The same plug-fit behavior is worth understanding alongside what plug-retention testing actually covers, so I’m not eating returns that were never defects.

What I can lock on the paper, and what the inspector decides on the wall

I’ve drawn a clear line around what’s mine to control. I can put TR — or WR, GFCI, AFCI — into the purchase order I send up and the confirmation I send down. I can require the marking on the box and ask for the listing and the test records. That’s all paper I touch.

What I can’t do is decide the location. From a PO line alone I can’t tell whether that address falls under 406.12 in the edition the local inspector enforces — adoption varies, and the AHJ has the final read at the jobsite. So I don’t promise compliance; I promise the device and the spec I shipped. When a replacement order reaches me without TR, WR, GFCI, or AFCI stated, I quote it as standard, non-TR — that’s my default, and it’s exactly how twenty thousand standard units were correct against my own silence. Now I flag the blank field back to whoever sent the order before anything ships, because the cost of asking is one email and the cost of guessing was a holiday-week scramble.

The checklist I run before a replacement order closes

Before an order leaves, I walk the location to the fields. What’s the location, and is it on the 406.12 list — if so, TR. Damp or wet — WR. A 210.8 area — GFCI. A 210.12 circuit — AFCI. Then the basics that were never in question: rating, face, color. Every one of those becomes a written field on the line, on both the order I place upstream and the order I confirm downstream, so the two match and nothing gets answered by assumption in between.

The wording that would have saved me was boring: “15A duplex, TR, listed, [color]” instead of “15A duplex, standard.” One field. That field was the whole difference between a clean delivery and three thousand emergency units flown out before a factory locked its doors. This is the same discipline that runs through a full GFCI and receptacle sourcing decision map — the location names the requirements, and the requirements belong on the line.

Sources

• NFPA 70, National Electrical Code — official standard reference for NEC article numbering, adoption, and code context:
  NFPA 70, National Electrical Code
• NEC 406.4(D) public code summary — receptacle replacement rules, including 406.4(D)(2)(a) for non-grounding receptacle replacement, 406.4(D)(5) for tamper-resistant replacements, and 406.4(D)(6) for weather-resistant replacements:
  Receptacle Replacement — NEC 406.4(D)
• NEC 406.12 public code summary — locations where listed tamper-resistant receptacles are required, plus exceptions such as receptacles more than 5½ ft above the floor and certain dedicated-appliance locations:
  Tamper-Resistant Receptacles — NEC 406.12
• NFPA tamper-resistant receptacle guidance — background on the purpose of TR receptacles and the safety reason for shuttered receptacle faces:
  NFPA Tamper-Resistant Receptacles
• ESFI tamper-resistant receptacle guidance — safety background for tamper-resistant receptacles and child-protection use cases:
  ESFI Tamper-Resistant Receptacles
• NEMA receptacle background — TR receptacles use an internal shutter / interlocking mechanism that opens when hot and neutral blades are inserted together, while a single object does not release the shutter:
  NEMA Connector Background
• Local AHJ guidance and adopted code edition — final enforcement depends on the code edition adopted in the project jurisdiction and the local authority having jurisdiction at the jobsite.