By definition, a catalytic converter should last the lifetime of the vehicle. It can last this long because a catalyst is something that, just by being present, causes a reaction to other materials without itself being a participant in the reaction. However, real life has proven that the life span of a catalytic converter varies as greatly as the life span of the vehicle itself.
One catalytic converter may not fail in 200,000+ miles, while another won’t even make it out of the vehicle’s base warranty. But one thing is for sure, they aren’t going away as long as vehicles are powered by fossil fuels. So let’s talk some about how they work, then move on to spotting one that is misbehaving.
As you know, there are multiple styles of cats — Conventional Oxidation Catalysts (COC), Three-Way Catalysts (TWC), and combination TWC+COC. Catalysts are fed a diet of harmful emissions — Hydrocarbons (HC), Nitric Oxide (NOx), and Carbon monoxide (CO). They digest these harmful gasses and expel less harmful as benign material. Oxygen (O2) is very important to this process. Without it, none of the harmful gasses will be converted.
Hydrocarbon: Inside a converter, the hydrogen atom is torn from the hydrocarbon molecule creating one free hydrogen atom and one free carbon atom (see figures 1 and 2). The hydrogen atom then joins with one free oxygen atom and another free hydrogen atom to form water (H2O). The one free carbon atom joins with two free oxygen atoms to form carbon dioxide (CO2), which has less harmful effects than carbon monoxide.
Carbon Monoxide: Carbon monoxide is combined with a free oxygen atom to create carbon dioxide. Carbon dioxide is less harmful to breathe than carbon monoxide, although you still can’t breathe it. CO2 still impacts the environment because it is a green house gas.
Nitric Oxide: The nitrogen atom is ripped from the oxygen atom. The free nitrogen atom is free to combine with another free nitrogen atom to make nitrogen (N2), a harmless gas. The free oxygen atom is then free to combine with two free hydrogen atoms to make water (H2O).
If you do the math, you’ll see that the molecules being converted are not bringing enough oxygen atoms with them to complete the conversions (see Figure 1).
Using only the oxygen atoms that are found within the gasses being converted, two carbon atoms are left out. Therefore, two oxygen molecules are needed to combine with the two free single carbon atoms to produce two carbon dioxide molecules in the case of three-way catalysts.
COC cats only convert HC and CO. So in the case of a COC cat, even more oxygen is needed from a source outside of the harmful molecules that are being converted (see Figure 2).
So where does the extra oxygen come from? On some systems, it’s pumped and piped into the exhaust. So on these systems, it is critical that the A.I.R. pump, tubes and valves are functioning properly. On systems without an A.I.R. pump, it comes from the leaner combustion events. This type of cat will have a layer of material called Ceria inside it. The Ceria will store extra oxygen molecules during the lean events (when the O2 sensor is below .45 volts). During the richer events (when O2 voltage is higher than .45 volts), the Ceria gives up those stored oxygen molecules to the exhaust stream. Having a healthy fuel control system, that is able to provide that predictable “lean-rich-lean-rich” cycling, is critical.
Hmmm, Curious… What killed the cat?
Usually, when a catalyst fails it is due to something other than the converter itself. Misfires and improper fuel ratios melt them down and clog them up. Oil, antifreeze, sealants, additives and cleaners can poison them. Impacts and deep water puddles can bust them up. If you replace a converter, you need to spend a little time with the vehicle afterward (if it is not evident beforehand) to determine if the vehicle will destroy another one.
M-I-S-F-I-R-E is how you spell death for a catalyst. Excessive hydrocarbons with excessive oxygen are what a misfire feeds to a cat. Excessive heat in the cat, to the point of meltdown, is what you get in return. If the openings of the honey comb inside are fused together, then this is a big red flag that further investigation of the vehicle’s ignition system, mechanical condition and fuel system needs to be performed.
Black or grey carbon soot clogging the ends of the honeycomb is a big indication of over fueling or oil burning. A lot of times when people think of over fueling (and oil burning) causing that type of condition, they falsely assume that clouds of smoke would also be evident at the tail pipe as well. That is not always true.
OBDII has done a lot to help alert the driver to problems such as a sluggish O2 sensor, but the driver still has the option to ignore the light until it is time for a new emission’s sticker. In a non-green state, people can ignore that check engine light indefinitely. So the cause(s) of a failed catalyst may have started a year or more before the catalyst became restricted.
Take a look at the sluggish O2 sensor in the capture labeled sluggish O2 in Figure 3. Do you think the catalyst is being fed proper proportions of pollutants and oxygen?
Over a long exposure time, bank 2 catalyst will certainly have a shortened life as a result of receiving an exaggerated rich/lean cycle. Controlling the amount of oxygen to the converter is also a way of protecting the cat (see Figure 4).
Take a look at where I parked the cursor — at the 11 second mark, in the data recording in Figure 4. What do you see? I had just went from cruising at highway speed to letting off the gas pedal, dropping from RPM numbers around 3800 to around 1600 and the PCM commanded a rich condition. The engine was not under a load in this shot. The O2 sensors didn’t cause it, they jumped right up to a high voltage with the high trims. There was no throttle demand for more fuel, so why did the short term fuel trims go high to around 10%? The PCM commanded a catalyst cooling strategy. It “knows” that after a sustained load and higher RPM, like that on the engine, that the converter must be very hot. So, to cool it down it enriched the fuel mixture. But wait a minute… wouldn’t that heat up the cat more? No. By enriching the fuel mixture, the PCM effectively cut off the cat’s oxygen supply briefly. By doing so, it cut the fire off inside the cat to give the cat about a two second cool down period to knock a few degrees out of it.
About two seconds of excessive fuel delivery stifled the reaction inside the catalyst long enough to have a cooling effect. If it wasn’t obvious before, then it should be crystal clear now as to exactly how sensitive a converter is to extended right and lean cycles that a sluggish O2 can cause. The fuel system and ignition system must be in proper working order for the catalyst to function properly.
Signs of a sick cat
You can certainly thank OBDII for increasing your catalytic converter sales. The PCM tests it, and if it is not up to par, it will tell on it. Still, in some places the customer can ignore the light, but even in green states, converters get replaced for nothing more than a check engine light because the customer doesn’t want to look at it.
One of the most common converter failures is simply to fail the PCM’s efficiency testing. To do this testing, the PCM compares the actions of the down stream O2 against the actions of the upstream O2. If the downstream O2 is cycling too many times when compared against the upstream, then a catalysts efficiency code may be set. For some vehicles, the results of this testing can be seen in Mode $6 (see Figure 5).
Looking at the Mode $6 example #1, you can the PCM’s converter test results highlighted in yellow. You can see that the PCM set a fault code for bank 2 catalyst efficiency low because that downstream O2 “switched” at a rate of 76% that of the rate of the upstream O2 sensor. The limit provided is 66%. This is actually a lot more involved of a test than just taking a little sample of one O2 against another. Because the driving conditions play a variable, the PCM performs many calculations against many samples before arriving at this decision. So it is not advisable that a technician attempts to monitor a brief sample of O2 signals and decide on his own if a cat is good or bad. However, they can become pretty obvious when viewed that way if the cat is completely dead.
Let’s look at that Mode $6 shot gain (Figure 6). This time I also highlighted the O2 sensor’s tests in green. Notice anything?
The upstream O2 sensor for the very same bank as the failed catalyst, bank 2, is weak. The O2 sensor test results show the O2 sensor barely made it above the threshold voltage of .44 volts when its time limit was up and the PCM took a snapshot of the sensor’s voltage. So, we see that we have an O2 sensor that is almost ready to set an O2 sensor slow response code. That O2 passed its test by just a little bit and is on the same bank as a catalyst that failed its test by just a little bit. So, is the O2 sensor bad enough to cause the cat to fail? Possibly.
If you’ll also look back at the cat test results, but for bank 1 this time, you’ll see that bank 1’s cat isn’t doing so hot either, it barely passed, yet its upstream O2 is testing well into the good. It is possible that the catalyst for bank 2 would also barely pass the test, at least for a little while, if it had the benefit of a healthier upstream O2 on its side. At the very least, we can see that one cat has failed the test (which is what the code already told us) and the other is about to fail testing (recheck on the horizon), as well as an O2 sensor that is about to fail (another recheck on the horizon).
Testing for a restricted cat can be both simple and tricky. Probably the best method to test for a restricted exhaust has some of the most hit-and-miss published specifications out there. An exhaust back pressure gauge is one of the best tools for this task, but finding a specification to reference your readings to can be tricky. After some digging through symptom flow charts, I managed to find a generic specification from Ford that says no more than 3 psi at idle, and no more than 8 psi at WOT. My personal experience has found that there should be no back pressure at idle, just a little needle bounce at best. And if I see more than 4 psi at a no-load 3,000 RPM, I’m suspicious.
I took a back pressure reading on a 2003 E250 4.2L van. The gauge popped right up to a reading of more than 5 psi at idle.
That was taken from the driver’s side upstream O2 sensor port. When measured from the passenger side, no back pressure could be found on the gauge through 3,000 RPM. Testing both sides like that proved that only the left cat was restricted. The right-hand cat, as well as the muffler and rest of system, were proven good in that manner.
But what lead up to testing the back pressure? The symptoms were the classic power loss and hissing sound. Misfire codes were set for bank 2 (driver’s side). Mode $6 shows misfire occurring only on bank 2.
Cylinders 1, 2 and 3 had no misfires showing in Mode $6 (figure 7), whereas cylinders 4, 5 and 6 (all on the same bank) were each showing the same number of misfires. In this case, simple reasoning indicates that the misfires counted were a symptom of the restricted cat, not the cause.
Of course, with this high of a number of misfires, each of the effected cylinders had a misfire code to match; P0304, P0305 and P0306. It also had a P0316 for misfire at start up.
When monitoring the O2 sensor and fuel trim PIDs, the restriction on bank 2 makes its mark clearly (figure 8).
Note the point where I parked the cursor. Notice the opposite standings of the short term fuel trims. Bank 1 is far positive while bank 2 is far negative. This is a classic condition of an engine with only one half of the engine effected by a restricted cat.
One side is flowing normally and on side is not. However, the PCM assumes both sides are flowing normally and that all of the air flowing through the MAF sensor is being distributed to both sides of the engine evenly. This ends up causing the good side to have too low of a base injector pulse will calculate for it, when compared against the air flow that is truly occurring on that side. Conversely, the restricted side receives a base injector pulse that is too high when matched against the air flow that is actually occurring on it. So the good side is too lean and the restricted side is too rich. When the O2 sensors pick that up, the fuel trims are then sent in opposite directions. Knowing that makes it pretty obvious that an exhaust back pressure testing will not only prove fruitful, but also provide a suggested outcome of the test before the test is even performed.
The effects of too much back pressure can be very nasty nowadays with all of the plastic parts on the engine. Like on this pressure differential sensor and EGR port seen below.