In November 2024, a 2011 Kia Soul pulled into my shop with a burned-out passenger headlight and a check-engine light on. The owner asked for a quick Kia headlight bulb replacement quote. It looked like a 20-minute job. I almost handed it to my tech and moved on.
Then I saw the connector. Melted.
The previous “quick fix” had used a cheap off-brand bulb that ran too hot. That small job turned into a wire-harness repair. And the check-engine light turned out to be a much better lesson than the headlight ever could have been.
Quick background: I’ve been running a ten-bay independent repair shop for about 12 years now. I’ve personally made—and documented—enough mistakes to total maybe $7,000 in wasted parts and labor. Some were classic beginner errors. One of the most expensive was assuming a catalytic converter was dead just because the code said so.
So let me tell you what I do now when a car throws a P0420, and how the same Kia ended up with a MOOG K200705 stabilizer bar bushing kit and a MOOG EV471 tie rod end.
How to Test a Catalytic Converter Before You Replace It
The Kia had a P0420 code: catalyst system efficiency below threshold. If you’ve ever had that code pop up on a scan tool, you know how easy it is to jump to the most expensive conclusion.
I did that myself in my first year, on a 2005 Honda CR-V. I quoted a catalytic converter, ordered it, installed it, and watched the same code come back a month later. That mistake cost roughly $800 in parts and labor—and a customer’s trust. After that, I stopped letting a code do my thinking.
Testing a catalytic converter doesn’t take a dyno or a lab. It takes a scan tool, an IR thermometer, and about 45 minutes. Here’s the version I use:
- Warm the car up completely. Catalyst efficiency means nothing on a cold engine.
- Put a scan tool on live data and watch the upstream and downstream O2 sensor voltage.
- Hold the engine at 2,500 to 3,000 RPM for about 30 seconds.
- If the upstream sensor is cycling and the downstream reading is flat, the converter is doing its job. If the downstream moves with the upstream, the converter isn’t cleaning the exhaust like it should.
- Use an IR thermometer on the inlet and outlet. A working converter usually runs hotter on the outlet side—often 100°F or more under load.
I also check for exhaust leaks before I touch a converter. A leak upstream of the O2 sensor can pull fresh air into the exhaust stream and confuse the readings. It’s not complicated. It just has to happen before you order a $400 part.
One more habit: I scan the freeze frame before clearing anything. It tells me what the engine was doing when the code set—RPM, load, coolant temp, fuel trims. If the freeze frame shows a lean condition and high load, I look for vacuum leaks and fuel delivery first, not the cat.
On the Kia, the O2 readings were moving together. That pointed at the converter, but I still didn’t order one. Then I looked at fuel trims and saw +14% at idle. There was a vacuum leak, and one of the spark plugs was worn enough to cause a small misfire. Unburned fuel was overworking the converter, making it look inefficient. The converter wasn’t completely dead; it was exhausted.
To be fair, there are times when the converter really is dead. I’ve replaced plenty of them. But I’ve also seen P0420 turn out to be a damaged exhaust pipe, a lazy O2 sensor, and once a broken wire in the harness near the connector. That last one would have made a perfect “new converter” victim: same code, same bank, same lazy reading. The wiring was the actual problem.
I fixed the vacuum leak, replaced the plugs, cleared the code, and told the customer to drive it for a week. The check-engine light didn’t come back. The surprise wasn’t the P0420. It was how many fixes pointed in the wrong direction before I actually looked at the data.
The Clunk I Almost Missed
With the engine running properly, I took the Kia for a test drive. The owner never mentioned any suspension noise. But every time I turned left into the parking lot, I felt a clunk through the steering wheel and heard something from the front passenger side.
Back on the lift, the stabilizer bar bushing was shot. The tie rod end had play when I pushed on it by hand. I showed the owner and gave him an honest choice: cheap no-name parts, or parts I’d put on my own truck. He said, “Do what you’d do on your own.”
That meant a MOOG K200705 stabilizer bar bushing kit for the front stabilizer bar and a MOOG EV471 tie rod end for the steering linkage.
I chose MOOG for a boring reason: fit and predictability. The bushing kit came with the bracket and bushings already lined up. The tie rod end had a grease fitting. No fight, no modification, no return trip to the parts store. On a small repair with a modest budget, that predictability saves real money.
The Cheap Suspension Part Mistake I Made in 2018
I wasn’t always this careful.
I want to say it was 2018—though I might be mixing up the exact year. I ordered a set of bargain sway bar bushings to save a customer about $25. I checked the parts myself, approved the job, and sent the car home. The clunk came back about three weeks later. That $25 “savings” cost us $190 in labor, a second set of parts, and a customer who had every right to be annoyed.
The cheap part wasn’t even visibly broken. It was just too soft, and the bar moved around under load. I learned the hard way that suspension parts are not a price comparison game. The total cost includes the redo, the downtime, and the credibility hit. And to be fair, some cheap parts are fine in non-critical spots. Steering and suspension is not the place I gamble anymore.
Small Orders Don’t Mean Small Expectations
I should add something that might sound unrelated: our shop isn’t a big fleet. We place small orders all the time. $50 here, $120 there. Some suppliers act like you’re wasting their time if you’re not buying in bulk. The good ones don’t make you feel that way—even when the order is a single tie rod end.
That’s one reason I keep going back to brands like Moog. They’re available through the same distributors, the catalog data is clear, and they don’t treat the little guy like an inconvenience. Small doesn’t mean unimportant. It means potential. The customer who came in for a $20 headlight bulb is the same person who called me two months later about a transmission service on his delivery van.
Last winter, I got the same reminder outside the automotive world. In January 2025, my uncle’s heating system started making noisy pipes. The first thing I suspected were the thermostatic radiator valves in the old hydronic setup. It turned out the system had air in it and the pump was weak. The valves were fine. A symptom is not a diagnosis.
Thermostatic radiator valves aren’t car parts, but the lesson is the same: confirm the failure before you replace the part. Whether it’s a valve, a catalytic converter, or a suspension bushing, the part gets blamed first. A good repair order happens when you prove the part is actually the problem.
What That Kia Taught Me
By the time we handed the keys back, the Kia had:
- A proper Kia headlight bulb replacement, with the melted connector repaired instead of hidden
- New vacuum lines and spark plugs that fixed the false P0420
- A MOOG K200705 stabilizer bar bushing kit
- A MOOG EV471 tie rod end
The bill wasn’t massive. The customer wasn’t angry. The car drove better than it had in months. There’s something satisfying about a repair like that—especially when the job started as something so small it almost went to someone else.
So here’s what I repeat to my younger techs: test things, stop guessing, and don’t judge a job by its size. If a small repair is worth doing, it’s worth doing with parts that don’t come back to haunt you.
And if you ever ask me how to test a catalytic converter before you order one, I’ll tell you the same thing I tell everyone: give me 45 minutes and a scan tool. I’ll show you exactly what I mean.