Signs a Catalytic Converter Is Going Bad—and What to Check Before You Replace Parts

If you're reading this because you searched for signs catalytic converter is going bad, what you probably want is a straight answer: replace the converter, or don't? I'm not going to give you a one-size-fits-all answer, because the same warning signs can come from very different places.

For context, I'm a quality and brand compliance manager at an automotive components manufacturer. I review stamped parts, stamping dies, CNC-machined parts, forgings, and aluminum extrusions before they ship—roughly 200 part numbers a year. In our Q1 2024 quality audit, we rejected about 8 percent of first deliveries. Most of those were not bad metal. They were missing material certificates, incorrect batch numbers, or documentation that didn't match the parts. That has shaped how I look at failures: the part itself is rarely the whole story.

Signs a catalytic converter is going bad: what the code doesn't tell you

The clearest signs catalytic converter is going bad are usually:

• An efficiency code like P0420 or P0430
• A rotten-egg or sulfur smell from the exhaust
• Rattling or metallic noise from inside the converter
• Loss of power, especially when the exhaust feels restricted
• Discoloration or heat damage around the converter and heat shield

Those warning signs are real. But I'd be doing you a disservice if I let the list end there. A catalytic converter is part of a larger system. A code that says catalyst efficiency is below threshold does not automatically mean the catalyst material is dead. That tempting shortcut is exactly how a lot of money gets spent on the wrong part.

Scenario 1: You're diagnosing the vehicle

If you're diagnosing a vehicle that's sitting in front of you, I'd treat the symptoms as clues rather than as a final answer.

Take diesel applications as one example. Differential pressure transmitters—the sensors that measure exhaust pressure across a diesel particulate filter—can fail in a way that makes the engine computer think the filter is blocked. The computer responds by running more active regenerations. The extra heat can damage a catalytic converter that was perfectly fine when it left the factory. If you only focus on converter symptoms, you may replace the converter and feel good for a while. Then the same sensor problem can damage the replacement.

I'm not a mechanic, so I won't pretend to diagnose OBD-II codes remotely. What I can tell you from a quality perspective is this: before you remove an expensive converter, remove the ambiguity. Check freeze-frame data, oxygen sensor switching, exhaust leaks, wiring, and whether the engine has been burning oil or coolant. A lower-cost sensor or wiring issue can cause the same warning signs as a dead catalyst.

One more thing I've learned: a failed part is evidence. If you throw it away before comparing it to the supplier's test report, you lose the chance to understand why it failed.

Scenario 2: You're buying replacement parts in small or medium quantities

Now suppose the diagnosis really points to a component that needs replacing. Maybe you're a repair shop stocking oil filter housings, or a small OEM building prototypes, or a fleet buyer ordering a limited run of hub assemblies. The size of your order does not mean you should accept lower quality.

Take an oil filter housing as an example. It looks simple: a housing, a seal face, threads, and maybe a cooler interface. But a porous casting or a seal surface that isn't flat can cause a slow leak after installation. A pressure test at the supplier is cheap. A disassembly, a returned part, and a customer who loses trust are expensive. If a supplier treats a low-volume order as too small to document, they're passing that hidden risk to you.

The same logic applies if you found this article while searching for NSK hub bearings. I think that's a fair search. Hub bearings are safety-related, and the brand name matters. But I don't treat any name—including ours—as a magical guarantee. A bearing is only as good as its raceway geometry, heat treatment, sealing, and traceability. The brand tells you who made it. The batch record tells you whether that specific part meets the spec.

I also won't tell you to trust the NSK website—or any supplier's website—as proof of quality. Websites are brochures. They're a useful starting point, but they don't confirm what came off the line last Tuesday. Ask for a batch number, a material certificate, and a dimensional report. If the response changes when the order is small, that's a red flag.

I've watched suppliers treat small buyers as interruptions, and I think that's a mistake. One of our earliest customers started with a low-dollar sample order—maybe $200, if I remember correctly—before growing into a much larger account. They are not our only customer, but they are a good reminder: a small order can be the start of a serious engineering relationship. The supplier who respects the trial order is often the supplier who respects the warranty claim later.

Scenario 3: You're qualifying a supplier for production volumes

At production volume, the question shifts from “is this part good?” to “can this process repeatedly make good parts?” That's a different conversation.

When I evaluate a supplier, I want to see process controls, not final inspection heroics. For stamped parts, I ask how die wear is monitored, how often first-article inspections happen, and what happens after a die is sharpened. For CNC-machined parts, I ask about capability indexes, tool life management, and fixture control. For forged or extruded components, I ask about heat treatment, hardness testing, and material traceability.

In automotive supply chains, the Production Part Approval Process—PPAP—exists for exactly this reason. It's not a stack of paperwork. It's evidence that the supplier thought through the process before parts reached your dock. If a supplier can't explain their process controls, a beautiful website and a low quotation won't save you later.

Here's something suppliers won't always tell you: a final inspection station can sort out bad parts, but it can't explain why they were made. If a metal stamper relies on sorters instead of die maintenance, the defects will keep coming. I'd rather hear about a supplier's corrective action records than their promises about perfect quality.

So which scenario are you in?

If you're not sure where to start, ask yourself what you're about to do next.

• If you're about to fix a vehicle, focus on system data first. That means sensor readings, freeze-frame data, and physical inspection before you spend money on a catalytic converter.
• If you're about to place a purchase order, ask for lot-level documentation. This applies whether you're buying an oil filter housing, a hub bearing, or a machined housing that holds a pressure sensor.
• If you're about to add a new supplier, verify process controls and part approval documentation before you commit to production volumes.

There's no single answer that fits every situation. The signs catalytic converter is going bad deserve attention, but they deserve the right kind of attention. Sometimes the fix is a new converter. Sometimes it's a faulty sensor. Sometimes the real problem is a supplier who never verified the part in the first place.

From where I sit, the goal isn't to avoid every defect. No process is perfect. The goal is to trace each defect back to a cause and prevent it from repeating. A quality supplier gives you the data to do that. A careless one leaves you guessing.