Short answer
For two failures out of six
A fuel-borne cleaner has a plausible mechanism against light surface deposits, and against a catalyst that has simply never been hot enough. It has none at all against a melted, broken, oil-fouled converter, or against a fault that was never the converter. Those four are the common cases.
OilAndMiles.com - Do catalytic converter cleaners work? Search the question and you get lists: best five, best seven, ranked and scored, almost always by a page that earns a commission on the click. What those lists don't do is the thing worth doing first: ask what a liquid poured into a fuel tank could physically change about a ceramic block in the exhaust.
The answer isn't nothing, and it's not much either. It depends entirely on which of the six ways a converter fails you're actually facing, and for four of them a bottle isn't a cheap first try, it's a delay.
We rank no products here and we name none. This site doesn't recommend a brand it hasn't compared under controlled conditions, and we have run no such comparison. What follows is the mechanism, which you can check yourself, and what it does and doesn't permit.
Do catalytic converter cleaners work on what is actually wrong?
Only on two of the six things that go wrong, and one of those two is free. A fuel additive goes into the tank, through the pump, through the injectors, and burns in the cylinder. Whatever survives that leaves as exhaust gas and passes over the catalyst.
So the question isn't whether the chemistry is clever. It's whether a substance arriving as a gas, at exhaust temperature, can change the thing that is wrong.
Against a film of soot sitting on the washcoat, there's a plausible answer: raise the temperature or change the combustion slightly and surface deposits can burn off. That is a real mechanism and it's the one every product on the shelf is describing.
Against a channel that has melted shut, there's no answer at all. The geometry has changed. Nothing that arrives as a gas reopens a fused ceramic passage, and nothing reassembles a substrate that has broken into pieces.
| What is wrong | Can a bottle reach it? | Why |
|---|---|---|
| Light soot on the washcoat | A cleaner might help | Deposits sitting on the surface can burn off at temperature. This is the only failure here where a fuel additive has a plausible mechanism |
| A converter kept cold | Drive it, don't dose it | Short trips leave the catalyst below working temperature. A sustained run at speed does what the bottle claims to, for the price of the fuel |
| Melted or collapsed honeycomb | Nothing reaches it | The channels have closed, and there's no chemistry that reopens a fused ceramic passage |
| Broken substrate | Nothing reaches it | The block has come apart. A liquid poured into the fuel tank can't reassemble it |
| Oil or coolant on the metals | Fix the leak first | The catalyst is coated in something that should never have arrived. Cleaning it while the supply continues is buying the same afternoon twice |
| A fault that was never the converter | The common case | A misfire, a lean mixture or a failing downstream sensor, the one after the converter, produces the same code. A bottle changes none of them |
The case where the free version works better
One of the two failures a cleaner might help is a converter that has never been hot enough to do its job. Short trips, cold starts, and a car that never gets a sustained run leave the catalyst below its working temperature for most of its life.
The fix for that is a sustained drive at speed, and it costs the fuel. That's also precisely what the instructions on the bottle tell you to do while the product is in the tank, which makes the honest comparison awkward: the drive is doing work the bottle is being credited with.
Diesel owners will recognize the pattern from the particulate filter lamp, which asks for the same thing and asks for it explicitly.
The four cases where it can't help
Melted or collapsed honeycomb. Usually from raw fuel burning inside the converter after a misfire, which is what a flashing check engine light reports. The channels are closed and no additive opens them.
Broken substrate. If it rattles, the ceramic is in pieces. That's mechanical and it's terminal.
Oil or coolant contamination. The precious metals are coated in something that should never have reached them. Even if a cleaner lifted some of it, the leak that put it there is still running.
It was never the converter. This is the largest group and the most expensive misunderstanding. A misfire, a lean mixture, an exhaust leak ahead of the sensor, or a failing downstream oxygen sensor all produce the code people buy cleaner for.
Our page on bad catalytic converter symptoms sorts which symptom points where.
What the data says
There's a federal program whose entire purpose is to test products like this, and its results are all public. EPA's Aftermarket Retrofit Device Evaluation Program, the “511 Program”, exists to “generate, analyze, and disseminate technical data on the effectiveness of such products”, and it covers fuel additives by name.
It has published 92 test reports since 1970, of which ten are named fuel additives, and every one of those ten was tested between February 1972 and January 1982. The newest report of any kind on the list is from October 2005, and no product sold today as a catalytic converter cleaner appears on it.
The reason is in the program's own first line: it is voluntary. A manufacturer has to ask to be tested. And EPA states what passing through it doesn't buy, in words worth carrying to a shelf: “EPA does not approve, certify, endorse, or register any products that pass through this voluntary evaluation program… Any claims by a manufacturer to the contrary are false.”
EPA, Evaluation Program for Aftermarket Retrofit Devices and its Test Reports list, read at source September 16, 2026. The counts are of the published list itself and can be recounted from that page.
Worth knowing
The strongest argument against reaching for a bottle isn't that it fails. It's what it costs in time.
A converter that is genuinely blocked is restricting the engine every mile it is driven, and a car still inside 8 years and 80,000 miles is inside the federal emissions warranty on that part. Two tanks of additive is two tanks of not making that phone call.
Why this page ranks nothing
Every page that ranks for this question is a list of products with a score beside each one. Almost none of them describe a test. Where a test is described, it's usually one car, once, with a code that cleared.
A code clearing after an additive proves less than it appears to. The monitor that sets it needs a specific drive cycle to run at all, so a code can clear because the car was driven differently, not because the catalyst changed. Any honest test of this product category needs the same car, the same fault, a control, and a measurement that's not the code itself.
We haven't run that test. Until we do, naming a winner would be repeating a marketing claim in our own voice, and the whole point of the fluids section here is that we don't do that.
| What is wrong | Can a cleaner reach it? | What actually helps |
|---|---|---|
| Light surface deposits | plausibly | heat, from a sustained drive or from the additive |
| Catalyst never reaches temperature | the drive does, not the bottle | a sustained run at speed |
| Melted or collapsed channels | no | replacement, after fixing the misfire |
| Broken substrate | no | replacement |
| Oil or coolant fouling | no | fix the leak, then assess the converter |
| A fault elsewhere | no | read the code and find the cause |
| What fails | What you notice | What is happening |
|---|---|---|
| It stops converting | No driving symptom at all | The honeycomb is intact and gas flows freely. What has gone is the chemistry, and the only thing that notices is the downstream sensor |
| It stops flowing | You feel this one | The honeycomb has broken up, melted or filled in. The engine breathes through a restriction, and it worsens the harder you work that becomes |
| Both at once | The usual end state | A converter that melted stopped converting on the way. Whichever symptom you noticed first, expect the other |
Will a catalytic converter cleaner clear a P0420 code?
Only if the code came from the one failure a bottle can touch: a catalyst that has never been hot enough to work, on a car that only does short cold trips. That one sometimes clears, and the sustained drive the label asks for is doing most of the work.
It won't clear the other four. A melted or broken honeycomb, a contaminated one, and the largest group of all, the codes where the converter was never the fault, are untouched by anything in the tank. A P0420 code reports a measurement rather than a part, which is why the bottle and the code are answering different questions.
What to do before you buy one
Five steps for deciding whether the bottle is worth opening, in the order that costs least. If you're holding a bottle and a code, this is where to start. 5 steps below:
- Read the code first. Free at most US parts chains. If it's not a catalyst efficiency code, the bottle was never going to be relevant.
- Take the free version of the treatment. A sustained run at speed, as the bottle instructions describe anyway. If a cold catalyst was the problem, you have just fixed it for the price of the fuel.
- Check age and mileage. Inside 8 years and 80,000 miles on a 1995 or newer car, the converter is covered by the federal emissions warranty and additives are beside the point.
- Rule out the impostors. Misfire, mixture, exhaust leak, or the downstream sensor, the one after the converter. All cheaper than a converter and all capable of setting the same code.
- Then decide about the bottle. It is a reasonable last cheap step against light deposits and a poor first step against anything else.
Conclusion
Two of six, and one of those two is free
A bottle has a mechanism against two of the six things that go wrong. Light surface deposits, and a catalyst that has never been hot enough to work. Against a melted, broken or oil-fouled converter it has none, and against a fault that was never the converter it changes nothing at all.
One of those two cases is free. A converter that never reaches temperature needs a sustained run at load, not an additive, and that costs a tank of fuel you were going to buy anyway. The other five cases need the fault found first, which is the part a bottle can't do.
Oil & Miles works from documents and counter prices rather than estimates, across repair costs and symptoms.
Common questions
Asked most often
Do catalytic converter cleaners actually work?
Against light surface deposits, plausibly. Against a melted, broken or oil-contaminated converter, no, because the problem is geometry rather than chemistry. Against a fault that was never the converter, which is the common case, they change nothing.
Will a cleaner clear a P0420 code?
Sometimes the code clears afterwards, which isn't the same thing. The monitor needs a particular drive cycle to run, so a code can clear because the car was driven differently during the treatment rather than because the catalyst changed.
Can you unclog a catalytic converter without replacing it?
If the blockage is soot and the cause is fixed, heat can help. Once the ceramic has melted or broken up, nothing poured into the tank reaches it, and no amount of driving reopens a closed channel.
Which catalytic converter cleaner is best?
We don't know, and we won't guess. Naming one would mean repeating a marketing claim in our own voice. A fair test needs the same fault, a control and a measurement that's not the code itself, and we haven't run one.
Can a cleaner damage anything?
Used as directed it is unlikely to. The real cost is time: a genuinely restricted converter is holding the engine back every mile, and two tanks of additive is two tanks of not addressing it.
Is an Italian tune-up the same thing?
It's the free half of the treatment. A sustained run at speed brings the catalyst to temperature, which is exactly what the bottle instructions ask you to do while the product is in the tank.
What people ask next
- Catalytic Converter Theft: What Actually Stops It 13 min
- Catalytic Converter Replacement Cost: When It's $0 19 min
- What Is a Catalytic Converter? Three Metals and a Honeycomb 16 min
- P0430 Code: Bank 2 Location and What Actually Fails 18 min
- Bad Catalytic Converter Symptoms: Two Failures, Two Answers 16 min
- P0420 Code: The Converter Is Usually the Victim 21 min
Sources
- EPA, Sale and Use of Aftermarket Catalytic Converters, read at source, for how a three-way catalyst works and what the precious metals do, and for the rules governing replacement. US Environmental Protection Agency, read August 26, 2026
- EPA, Evaluation Program for Aftermarket Retrofit Devices and its Test Reports list, read at source, for the program's scope, its voluntary basis, the wording on endorsement, and the counts of published reports. US Environmental Protection Agency, read September 16, 2026
- 40 CFR 85.2103, the federal emission performance warranty, read at source, for the 8 year or 80,000 mile coverage on a catalytic converter as a major emission control component. Cornell Legal Information Institute, read August 26, 2026
How this was compiled
The article is built from the mechanism outwards: what a substance that arrives at the catalyst as a gas could change, and what it couldn't. That produces a shorter and less flattering answer than a product list does, and it's checkable by anybody who reads it, which a score out of ten isn't.
What we didn't do
- We tested no product, so we name none and rank none.
- We give no success rate. Every figure in circulation for this product category traces back to a manufacturer or to a single anecdote, and repeating either would be inventing evidence.
- We didn't measure catalyst temperature or back pressure on any vehicle.
- We carry no affiliate link on this page and wouldn't take one for this subject.
Prices are ranges compiled in August 2026 for the US market and vary by region, engine and shop rate. Prices from $10 up are rounded to the nearest dollar, so $14.13 reads as $14 and $14.52 as $15. Below $10 they're exact. This article is general information, not advice about your specific vehicle, see the disclaimer.