A bad batch of steel is the most comfortable explanation for a cracked swing hanger, and it is the explanation that falls apart first. Send the broken part to a lab that handles the sort of failure analysis nc manufacturers order when a product comes back in pieces, and the fracture surface usually points somewhere less convenient than the supplier. A metallurgical examination of that surface, paired with the plating chemistry, names the mechanism, and the mechanism decides whether the fix is a new supplier or a new drawing. Those two answers cost wildly different amounts. One is a phone call and a credit note; the other is tooling, a revised spec, and a season of sales you do not get back.
Picture a family-owned playset maker in North Carolina shipping around 8,000 backyard swing sets a year, mostly through independent dealers. Three hangers come back cracked in the same quarter, all from different states, all installed by different people. The temptation is to blame the plating shop, order a new lot, and move on. The parts that came back are the only witnesses you have, and they will not stay useful forever once they sit in a drawer collecting rust.
Hydrogen Embrittlement Explained Without Jargon
Zinc plating is a chemical bath, and a bath does things to steel beyond leaving a coating on it. During cleaning and plating, hydrogen works its way into the metal, and if it is not driven back out by a low-temperature bake soon after the parts come off the line, it collects at the grain boundaries and waits. Put the part under load and it cracks at a stress it should have shrugged off. That is what metallurgists call hydrogen embrittlement, meaning the steel got brittle from absorbed hydrogen rather than from anything you can see on the outside. Hard fasteners are the most vulnerable, which is awkward, because hardness is often exactly what the drawing asked for. The case we see most often is not one bad ingredient but a stack of small decisions: a hardness spec at the top of its range, a plating line running a hot pickle, and a bake that either happened late or did not happen at all.
The timing is a tell, and it is free to look at. Embrittlement cracks tend to appear early, within days or a few weeks of the set being loaded, while fatigue cracks arrive after a season or two of a child pumping a swing to full height. Under an electron microscope the two look nothing alike, one splitting between grains and the other marching across them in beach marks. That is why the broken hanger matters more than the paperwork. The fracture surface keeps a record nobody in the room can argue with.
Ask A Metallurgist These Questions First
Before you authorize the work, find out what the report will actually tell you. A lab that is comfortable being pinned down on method is a lab that will still be comfortable if a lawyer reads the same report two years later. Ask these, and listen for whether the answers include numbers.
- What fracture mode are you going to identify, and how? A good answer names intergranular versus transgranular cracking and says which imaging will show it.
- Was this hardware baked after plating, and can the supplier produce the record? A good answer asks for a temperature, a hold time, and how many hours after plating it happened.
- What hardness is the part, measured rather than assumed? A good answer comes back with a number and says whether it sits in the range where embrittlement gets likely.
- If we pull ten more hangers off the shelf, what do you expect to find? A good answer commits to a prediction before the test, not after it.
- What in our design made this failure easier? A good answer is willing to criticize the drawing, not only the batch.
None of this is exotic laboratory work. The failure analysis NC manufacturers commission for a part like this usually lands as a written report with photographs, a named root cause, and a remediation section you can hand to both the plating shop and the engineer who drew the part. Expect visual inspection first, then sectioning, then microscopy and chemistry. The report is the deliverable, and a report that names a mechanism is worth five that describe an appearance.
Corrosion gets an inspection schedule wherever a code writer has bothered to think about it. Rimkus Consulting Group published a review in April 2026 noting that the 2026 edition of NFPA 25 requires annual internal inspections of dry-pipe, preaction and deluge valves, and that where nitrogen generators, vapor corrosion inhibitors or negative-pressure systems are used to hold off dry pipe corrosion, those systems have to be inspected, tested and maintained to the manufacturer’s instructions. Nobody has written an equivalent schedule for a swing hanger in a back garden. That gap belongs to the manufacturer, which is the real reason to spend money on one failure: the findings tell you what to check on the next 8,000.
Evidence Ends The Batch Versus Design Argument
Before any of the laboratory work matters, the equipment comes out of service. The U.S. Consumer Product Safety Commission’s 2025 public playground safety handbook requires that hardware found cracked or broken be taken out of use immediately and stay out of use until it is properly repaired, and a hardware store bolt of roughly the right size is not a repair for a rated structural fitting. Children stay off the set until a professional has fitted the specified replacement part. Tell the customer that in the first sentence of the call, before you say a word about testing or warranties.
What I cannot tell you is how many cracked hangers are in service right now, on sets nobody has looked at closely. No registry tracks backyard equipment failures, warranty claims capture only the fraction of parents who bother to file, and I have never seen a credible multiplier for the rest. I have testified about a broken part that had been quietly splitting for most of a summer before anyone mentioned it. So the honest position is that three returns are three reports, not three failures, and you size your response to the mechanism rather than to the count.
That is what the investigation buys you. If the answer is embrittlement, you change how the parts are processed and you add a bake record to your incoming inspection, and the design stays. If the answer is a stress concentration at a radius the drawing never should have allowed, no supplier on earth fixes it for you. The argument between a bad batch and a bad design is only unresolvable while both sides are guessing, and steel is not a hard witness to question. Get the part read, then spend the money once.