How to really tell if catalytic converter is bad

Maintenance/Repairs
21

DO NOT REPLACE A CATALYTIC CONVERTER UNTIL YOU HAVE REPLACED THE BEFORE AND AFTER SENSORS AND DRIVEN IT A WHILE. PERIOD.

That just makes no sense to me. Parts are easily tested and verified as good or failed. You’re advocating doing away with any sort of technical ability and using “swaptronics” as a repair method.

Today I had in a 2002 Toyota for an emissions failure. It had fault code P0420 indicating poor catalyst efficiency. After examining the car for exhaust leaks, checking the fuel and ignition systems for proper operation, and then road testing the car under different conditions, I reviewed the data I had been graphing and recording and came to the conclusion that the catalyst is failing. I am 100% confident that the air/fuel ratio sensor and the oxygen sensor were performing properly. I am 100% confident that replacing the catalyst with a quality new unit would repair the car. I am 100% confident that replacing the sensors would have been wasting $300 of someone’s money.

22

Testing the temperature is a good way to determine if the cat has failed or is near to fail. A good cat can get hot too if there’s a lot of unburned gasoline going into it, overly rich. Overly-lean operation can cause cats, like the rest of the engine parts, to heat up too I think. If one of a pair of cats reads a much higher temperature than the other, that could be caused by a problem in one of the cylinders up-steam of that cat. Common possibilities are spark, compression, fuel injector, head gasket, etc. If the cat is on the list of suspicions for causing a drivability symptom, one idea might be to temporarily remove or bypass it and see if that solves the drivability problem. The car’s computer compares the signal from the pre-cat O2 sensor to the post-cat O2 sensor, to determine if the cat is working. The computer algorithm may use other information from the car’s sensors before making this determination; it might need more than just the two O2 sensors to flag a non-functional cat.

Edit: Many shops have emissions sniffers. That’s another way to tell that the cat might not be working correctly.

23

Interesting discussion, but back to the original question - from the temperatures measured, what we learn is that one cat is oxidizing more CO or HC than the other in the exothermic end of the cat. What we don’t (unfortunately) know is whether that is because one cat is oxidizing more effectively, or whether both cats are working fine but one cat is receiving more HC and CO and perhaps less free oxygen as well.

Could it be spark plugs? It could, but that would not be my first bet. Vacuum leak would be more likely suspect. If you can get a clear shot at all the individual exhaust manifold pipes with your thermometer, try measuring them and see how the temperatures compare. Try to take all measurements the same distance from the head.

If the exhaust pipes run independently from the cat to the tailpipes, feel the exhaust flow and see if the individual bursts of exhaust feel the same on both sides. This will tell you if you have a cylinder on one side that is not firing as strong as the others at idle. If one side feels like a smooth flow with no individual bursts at all, that suggests obstruction.

24
That just makes no sense to me. Parts are easily tested and verified as good or failed. You're advocating doing away with any sort of technical ability and using "swaptronics" as a repair method.

EXACTLY…If you’re doing the work yourself and don’t know the procedures for testing the O2 sensors…then yes replace them first. A good competent mechanic can determine exactly what the problem is…and not just replace parts.

25

I did not say not to test the sensors. I meant do not let the mechanics start you by replacing the cat. If you do not trust anyone to be able to test them for you, which is a very common situation, then replace the sensors before the $2000 cat, as on some cars. And, in most cases, the problem will be solved.

I did not intend to be writing for pros who can very well test the sensors with a graphing scanner. Otherwise I stand by what I said.

We have had a number of people over the years who told us the mechanics and service managers told them P0420 always means the cat has to be replaced.One such case in the last few days. Nonsense.

Over the years, we have had a steady stream of professionals who say they always have to replace the cat when they get P0420. Others have said they almost NEVER have to replace the cat, the sensors usually fix it. Clearly a lot of cats are being replaced when it was not necessary.

I also discovered that many mechanics who insist they always have to replace the cats, also 'REPLACE THE SENSORS WHILE WE ARE AT IT."

26

I don’t know of a single mechanic that would replace any working sensor of any type just because they were performing a certain repair/service.

That would be like telling a customer, " Oh! While I’m doing a coolant exchange, I’m also going to replace the coolant sensors for the computer, radiator cooling fan(s), and dash light/gauge for the coolant temp. Even though they’re working."

Tester

27

Let us not be ridiculous, Tester. Are you the man who asked me why I didn’t replace the transmission every so often some years ago when I explained high-rel maintenance and why I replaced all the light bulbs and similar items every so often?

My computer mouse did something strange and I lost a lengthy posting. That has given me time to think it over. I think I am going to stand pat on this.

I worked in a very high tech factory. We had roughly 300 technicians, and only a handful at any time were what I call Top Guns. But they all thought they were Top Guns. I was at times a Top Gun but only on microprocessors and digital circuitry. At times I was a dog, especially on anything not digital. The difference is I well know when I was a dog. Most dogs have no clue how bad they are. Top Guns know very well when they are on something beyond their ability. (The trick is to work on stuff you are good at, heh, heh.)

We have to deal with reality, not the way things should be. Yes, all mechanics should be able to do what Asemaster and OK and Mike can do. But, in real life, the vast majority, probably well over 95%, of car owners do not have access to that quality of mechanics.

Asemaster, if I come into the shop where you are, will you work on my car? Or will it be whoever is out of work? I am betting on the latter. Top Guns like you only work on a car after the customer gets on the No-Fly list with death threats against the shop.

I have followed this debate on sensors vs. cats, since I started here in the late 90’s. Around half of the mechanics have said they almost never have to replace the cats, the sensors almost always fix the problem. The other half insist they almost always have to replace the cat. But, when you ask, they will tell you, yes, of course, they think you should replace the sensors “while you are at it.” That tells me that most replaced cats were not bad, and that is a lot of money. In the overall picture, the price of mistakenly replacing a good sensor dwindles to nothing in comparison.

The people who have posted that they were told P0420 means the cat is bad, have never said the service manager told them that a graphing scanner showed the sensors were good. They said the P0420 means the cat is bad. I think I am going to stand pat. If a poster says they were told P0420 means the cat is bad, tell the mechanic to replace the sensors and I will drive it around for a while before you hit me for big bucks.

We have even had mechanics on this board talking about cats wearing out. They do not wear out. They can become mechanically damaged, or contaminated. There is nothing to wear out in there. Look up the word catalytic in a good dictionary.

Over those many years, I also read a lot, all over the Web, many car forums (fora?). The results were the same, most times the sensors are the problem. And, those who do replace the cats mostly also replace the sensors.

Yeah, I am standing pat.

My thanks to the mechanics here who are indeed Top Guns. You should be an example to all mechanics. But, I am not holding my breath.

28

But, when you ask, they will tell you, yes, of course, they think you should replace the sensors “while you are at it.”

As often as not an estimate for replacing a catalyst will include a new sensor as part of the job. That’s because I may have to remove the sensor from the old catalyst and install it in the new one or I may need to remove it from a pipe near the sensor to weld in a new one. Often these sensors are rusted in, the threads may gall on removal or the banging and thrashing needed to remove them may damage them. If a sensor is located in a catalyst that is being replaced, I include a new one in the estimate. If the sensor comes right out, I reuse it. If it shows any sign of a struggle, I replace it. I will not spend $40 in additional labor time to save an $80 part. That’s a disservice to my customer.

We have even had mechanics on this board talking about cats wearing out. They do not wear out. They can become mechanically damaged, or contaminated. There is nothing to wear out in there.

Like it or not they do sometimes just go bad. Now I will readily admit that that may have much to do with quality control and construction. It may have to do with software. It may have to do with design. For several years it was common to see mid-size GM cars with a failed catalyst at 80-100K miles with no underlying causes. Ford Escapes seem to suffer rear catalyst failure, with or without misfire issues. And frankly I don’t spend too much additional time searching for sources of catalyst failure on, say, a 15 year old car with 200K miles. Engines are crude and imperfect, things wear out.

True, a catalyst “should” last the life of the car. So should an ignition control module, a fuel injector, a camshaft position sensor, and many other things. But sometimes things just break.

29

If engines continued to put out only the levels of undesirable components that they did when they were freshly broken in, converters might last the life of the vehicle. But in the real world, that doesn’t happen.

Or, if platinum-palladium were cheap perhaps manufacturers would build in enough core to enable converters to last a lot longer… but that isn’t real-world either.

30

Testing should be the first step, not replacing. I think we all agree to that.
Cat converters eventually wear out even under the best conditions, due to erosion of the precious metal coating from its substrate.
Same with O2 and A/F sensors.

31
Cat converters eventually wear out even under the best conditions, due to erosion of the precious metal coating from its substrate.

It must be a very very very long time. I’ve never worn one out…two of those vehicles over 400k miles.

32

The cheap aftermarket converters have thinner coatings and don’t last as long.

33

Since most of us shade tree mechanics don’t have adequate diagnostic equipment, the How old is it? test is often used. If an oxygen sensor is suspect, and it has over 100k miles on it, the ‘parts replacement’ approach to diagnostics is often the most cost-effective. Since a shop mechanic may well use the same technique, it is cheaper for you to use that technique than to pay someone else to do it.

34

Some time ago, we had this same discussion. A few new bits of information this time, but essentially the same opinions.

At that time, I wanted to know because we get postings all over the place and I happened to have an intermittent P0420

I am retired and love to do Internet research if something is important. I can’t even remember how long I spent on that project but it was many hours. There were discussions on boards; Ford PIckup boards; Chevrolet pickup boards; Dodge pickup boards. Porsche boards; Toyota boards, and lots more.

Very consistently, the number of mechanics who said they almost always had to replace the cats was very close to the number who said they almost NEVER had to replace the cat. I did not spend much time contemplating the possibility that the same mechanics always got cars with bad cats, and others always got cars with good cats.

On the surface, I had the impression that most replaced cats were actually good. But, after some thought I realized with no N for each person, I could not come to a very accurate estimate on the number of cats replaced. So, I simply concluded that it had to be at least half the cats replaced were actually good. And probably more than that. It was amazing how many people reported, like the poster on this board, that the mechanics told them P0420 meant bad cat, period.

This is totally consistent with both incompetence and greed. Dummy’s not gonna’ know his cat was okay. Hose it to 'em.

Just now, I spent maybe twenty minutes trying to find out how many replacement cats are sold every year. I did find a cost estimator, which gave a high figure of around $900 or so for parts and labor at the top end. That is much lower than posters here have reported being charged.

I just googled for how many cars in the US. There were figures all over the place. One common figure was around 136 million cars, and 110 million trucks including pickups and SUVs.

Again, lack of data. I don’t know how many of the vehicles have cats, but assume most do. Nor do I know how long an average cat lasts. If ten years, that would be 10 to 20 million cats replaced a year. I certainly hope someone can show me those figures are way high. They would indicate some billions of dollars a year just replacing cats.

And, anything over a very small error figure means car owners are being milked out of billions of dollars by replacing good cats for whatever reason.

That is why I think in the absence of a very good diagnosis like Asemaster has done, the sensors should be replaced before the cat.

Always I find when doing a research project, I find new questions. In this case, I wonder if needing a new cat is responsible for final junking out of bottom feeder older cars, just as a bad motor or transmission is at higher car values.

I agreed above that we base our opinions on our own experiences. I was wrong in agreeing. When I do major research projects I am seeking feedback from large numbers of people, not just the person posting on the topic. I know very well my knowledge is very limited.

35

I do agree with Mike about cats wearing out. the horribly expensive materials the cat screen is made out of is designed to get very hot and burn unburned hydrocarbons. The screen is self-cleaning up to a point. But, when those wires do get covered by contamination, I suspect it doesn’t take long for more crud to coat the screen which then blocks the cat completely.

Once the screen is fully coated, the cat will stop functioning and can no longer clean itself. That will sure act like they have worn out!

Just a few days ago, we had a poster who said he took off his cat and soaked in dish soap water over night, as Scottie KIlmer has suggested. Small pieces of white stuff came floating, which would seem to be a form of contamination.

Which reminds me of another major research project I did. I posted Scottie’s video on putting a gallon of lacquer thinner in your gas tank, and driving it out to clean the cat. Posters here all said it couldn’t work and would destroy your fuel system. I found a lot of discussion on the Web, but most never tried it and said it would wreck your fuel system.

A very few tried it and almost all of them said it worked and did not harm the fuel system, except on one very old Honda. Some said P0420 came back in a few months.

If the cats wear out, why do used cats bring so much for recycle value? Think about it.

36

@irlandes makes some good points about proper testing, and unfortunately I have to admit that the number of shops who automatically replace cats for a P0420 is higher than I would like. But I will also say that a number of mechanics who replace sensors first may be misinformed about whether they fixed the car or not.

Joe may take his car in to a shop, be told that the code could just be caused by sensors, gets them replaced, only to have the engine light come on again a week later. At this point he may figure that shop #1 didn’t know how to properly fix his car and then takes it to the dealer/exhaust/other shop where they replace the catalyst and the car is fixed.

Shops that do a thorough and complete test and diagnosis are up against places like Autozone–and even other shops–that advertise free code scan or free diagnostics. You come to me, it’s going to cost you about $100 for my diagnosis. The discount muffler and brake shop down the street advertises free diagnosis. The thing is, even with the other guys doing nothing more than a code read and recommending repairs based on most probable causes, they’re going to be right as often as not. The dirty little secret is that when they’ve done what they can and the car still isn’t fixed, the car gets sublet to me for diag. They eventually fix the car–on the customer’s dime–and come out looking like the hero for figuring it out, and I’m the guy down the street who’s too expensive to take the car to in the first place.

37

I’ll go slightly off-topic

Scottie Kilmer should probably be involuntarily committed to a mental institution

I would recommend that any professional mechanic does NOT follow the guy’s advice. Do a proper diagnosis and repair. You’ll be better off in the long run

With his “peppy” personality, two things come to mind

He’s high on something . . . at the very least he’s had WAY too much caffeine before filming his videos

He’s so “peppy” he should be the spokesman for pep boys

38

"Small pieces of white stuff came floating, which would seem to be a form of contamination."
Not without an analysis of the composition of the white stuff…

My brother enjoys humiliating people, embarrassing people, and planting disgusting things in the minds of kids. He used to tell little kids that the runny part of eggs was from the farmer clearing his sinuses. He’d tell them that the boogers would get on the feed, the chickens would eat the boogers with the feed, and it would come out in the eggs.

Scotty Kilmer may THINK this white stuff from the converter core is contamination, but that only suggests to me that he never should have graduated from high school.

39

Cleaning a cat with solvents or dish soap will not work. Anything that comes out was not contamination of the active catalytic surface. The contamination is chemically bound to the surface, not stuck on. If the car has been running rich, some junk can be burned off by fixing the problem to get a normal mixture then taking the car for a nice run and getting everything hot. Any coke (unburned hydrocarbons) on the cat surface will burn off, but any contamination remains. The only way to get rid of the contamination is to melt the material, re form it and once in it’s new configuration, get it hot, just short of glowing and the material is now reactivated. The platinum , palladium and other material used for the catalyst is still valuable even when contaminated, but not useful as a catalyst until reactivated. Used cats are torn apart and melted down to make new cats.