Catalyst Control Function

[CCycle]

I would appreciate information on catalyst emission control function.
It looks like this is the board to get that information.

The particular car, '98 Buick 3.8 liter.
It ran ok up to about 70 K miles where upon the check engine light came on.
Code check says low catalyst activity.
There is no after-market catalyst available. GM 's price $500 and about $1000 installed.
I bought an after-market catalyst which is the same size and looks similar.
Cut the flanges off the original catalyst and welded them on the after-market unit.
Well that ran about 15 K miles and we get the low catalyst activity code again.

Now for the questions:
1. I understand the engine control modulates the air/fuel ratio each side of stoichiometric.
What is the frequency and approximate swing of this modulation?

2. How do the O2 sensors work? I understand that the voltage swings from about 0.9 to 4 volt.
Do they self generate this voltage or is it sinking externally supplied current?
What is a typical internal impedance of a sensor?

3. I read about various "washes" to activate catalysts.
Are these available? Would it do any good to do this myself?

Thanks, Km

 

[1Bluegrass]

To start you off..
1. the ox sensors modulate the fuel system in closed loop configuration. The sensor is an input for control.
2. the ox sensors must operate at over 600 degrees (as well as the cats) and generate a voltage output related to oxygen content. The output swing is normally .1 to about 1 volt.
The impedence can't be measures by the standard meter without risking damage. The actual impedence may be quite different at operating temp.
3. there are some chemicals that will tend to clean off deposits from the catayst substrate.
...
My sense is that you have some other fault causing the problem.
Air injection, if the car uses that system may not be operating the cat at high enough temp.
An air leak ahead of the cats.
Fuel system running to rich.
Leaking fuel injectors.
Something in the fuel may be contaminating the cats.
Since the problem has returned, I would assume the cats are a secondary result problem and is the only way the engine control system has to alert you by lighting the MIL lamp and setting a code as the (end) result.
Forgive me a bit if I am inaccurate about GM, as I am totally familuar with Ford systems in relating to these functions but they all work much the same.

 

[btrueblood]

Or, you are blowing enough oil through the piston rings to foul and/or poison the catalyst.

 

[patprimmer]

EP additives in oils can poison cats if the engine is burning oil.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[magnograil]

There are two types of O2 sensor, narrow band (which is probably what you have) and wide band. Not electrically interchangeable. Narrow band jump between lean and rich at approximately 0.45 volts. Wide band are nearly linear with F/A ratio. If you bought the K&L F/A meter ($200) you wasted your money. The narrow band sensor cannot give meaningful output on how rich or how lean, only rich or lean. A dot mode driver, bargraph LED and sensor is all it is.
There are several different types catalytic converters available. The first only converted hydrocarbons, the later CO and then NO2. Find out which yours is supposed to have. Some also use air injection.
You can check your sensor with a propane torch, digital volt meter and a pencil. Heat the sensor with the torch, then burn the pencil between the torch and sensor. Check fo a lean/rich voltage change. Look up O2 sensors on a search and you will find lots of information. For $35 or so for a new sensor it may not be worth it.

 

[TurboTR]

I didn't see you mention- how is the health of your downstream O2(s)? Is it working correctly? OBD2 mandates that the catalyst function, and anything else that can affect emissions must be monitored. If the ds O2 is bad it might be false coding.

TurboTR

 

[CCycle]

Great info. Thanks a lot.

More on the car:
1. The car runs great. I change oil when ever it tells me to which is about 6K miles.
The oil is not down much and we have never added any oil.
It gets low to mid 20s mileage.

2. I understand the catalyst is a 3 way.
There is no build up in the exhaust system .
The catalyst brick is clean.

3. I did change both O2 sensors.
The first one did go bad according to the code reader.

4. I have not purchased any analyzer equipment other than a OBDII code reader.
But I was considering building a virtual O2 sensor for use in leu of the second sensor.
The signal would be derived from the first sensor, delayed and sent to the control system.

 

[TurboTR]

Hmm. So you say the primary O2 went bad? When did you replace the sensors? Does the car lose any coolant at all? Are the exh headers tubular low grade stainless? If so they can crack, and that might throw things off severely. The turbo Buicks were notorious for cracking the ps side.

TurboTR

 

[CCycle]

TurboTR, It is not turbochanged. The exhaust system is tight as far as any inspection can show. There is no noise like a leak makes.
We have never had to add coolant. But one time about 80 K miles I noticed some coolant on the floor under the car. I could not define where it was coming from other than it was high up on the engine. I was hard pressed for time right then so I took it into the dealership. They changed the intake manifold which I understand is plastic and all go bad. Some coolant could have gotten in the engine but it wasn't much.

As for the O2 sensors, The car started showing codes around 70K miles. The codes were P0131, O2 sensor Low Voltage Bank 1 Sensor 1 and P0420, Catalyst System Efficiency Below Threshold Bank 1. I changed the first O2 sensor. It then showed P0420 code only. So I changed the second O2 sensor. It still gave the P0420 code. So I changed the catalyst. No more codes for about 15K miles where upon the P0420 code is back.

 

[EricFahlgren]

Could you clarify which O2 sensors you changed? I was under the impression that these cars had three sensors, two pre-cat and one post-cat. This sort of problem, as TurboTR says above, is usually indicative of the post-cat sensor going bad and this would be consistent with the mileage on the vehicle.

 

[CCycle]

Eric, This car has two O2 sensors. One before and one after the catalyst.
Both of the sensors got changed but not at the same time. See sequence previous post.

Thanks for your interest and info.

 

[TurboTR]

As far as the frequency and swing of the a/f modulation, roughly, at low speeds it could be in the range of 10-30+ or so per second say. At higher speeds maybe 40-60+ or so. Again, roughly; your mileage may vary The older GM scheme used a variable called O2 crosscounts that maintained a running, rollover count of the number of crosscounts. The time period we were used to looking at per counter update ('87 Grand National 3.8L) was 1.4 sec, the udpate rate of the ALDL scantool data output from the ecu.

The a/f swing, I would guess probably say +/- 1 a/f ratio or less might be a reasonable goal. Just a guess though in this case...

TurboTR

 

[TurboTR]

Re: a/f swing, FWIW, I do SCT tuning for late model Fords, and we generally have a wideband hooked up to the car as well. I generally see an a/f swing that fits what I posted above. Hope that helps some.

TurboTR

 

[CCycle]

TurboTR, I believe you are telling me that the a/f ratio crosses stoichiometric 10 to 50 times a second! What then is the criteria for showing a P0420 code?

Some GM literature I saw someplace said the low catalyst activity code (P0420) was determined by the time difference in transitions between the first and second sensor. But if the a/f swings are fast enough a rich or lean edge may never break through the catalyst.

 

[TurboTR]

Yes. Who knows, maybe there is even a special diagnostic routine the ecu runs occasionally to check the catalyst health Unfortunately I don't know the details of how that part is monitored.

TurboTR

 

[Rob45]

The ecu looks for the upstream O2 sensor and downstream O2 sensor output to be "in phase" so to speak, rather than looking purely at lean vs rich.
If they're out of phase, it assumes the cat's gone bad.
Something you could expect in 15K miles from an aftermarket cat - why do you suppose it was so much cheaper?

- R

 

[Aquaduct]

Unfortuneately, you've made a mistake. It isn't your fault, but you're down the road already.

You've got an OBD2 vehicle. The fore and aft O2 sensors monitor oxygen storage in the catalyst. As the catalyst deteriorates, O2 storage also decreases. The light illuminates when the control system determines that catalyst function has deteriorated to the point that the car is emitting more than 1.5 times the regulated amount of some pollutant. The cat efficiency is probably 85-95% at that point. By hooking up an aftermarket catalyst that isn't designed for the engine, you're burning it up in the same manner only faster.

The mistake you made is that the catalysts (as well as the O2 sensors and the ECU) are emmission related components that, by law, are warranted to 120,000 miles or 8 years, regardless of the warranty on the rest of the vehicle. If you would have taken it to the dealer, it'd be free.

Sorry.

 

[patprimmer]

Aquaduct

CCycle did not say where he is. Are you sure that law applies to everywhere a 1998 Buick 3.8 litre might live.

If the law does apply to CCycle's location, it might be a bit naughty, but fair seeing as the original cat failed, if he was to use a second hand cat and get that replaced under warranty if it fails.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Aquaduct]

Pat,

I was in Powertrain Engineering at GM's crosstown rival for most of the '90s. I've been in heavy trucks for the last 5 or 6 years, so my memory may be a bit fuzzy on some details, but I'll try to help.

OBD2 is a mandated part of the emission system both Federally and in California (there are some minor non-technical differences). Canada adopts US standards so it's good there too.

Different states and smaller jurisdictions have different ways of enforcing it, however. California, for instance, requires that the Check Engine light be functioning but not illuminated (comes on for a couple seconds at start and then goes off) before they'll renew your plates. If CCycle lives in one of these jurisdictions, there's a paragraph or 2 on the emissions warranty in the Owner's manual (small and tough to find, but the Feds say it's gotta be there) and if he went to the dealer and told the sad story of what he's done, he may be able to talk them into doing the warranty work (heck, they'll just bill it back to the General). Some proof like receipts may help and I'd bet the old parts would really do the trick but that's up to the warranty system.

Most jurisdictions don't enforce it at all. Several years ago my light came on in my '96 Aerostar (I live in Virginia) for a similar Evaporative Emission system problem. The EVAP stuff isn't under that warranty. I ignored it and 2 years ago, the light burnt out. No more problem. These failures don't cause operating problems, they just annoy the crap out of you. If he's too impatient to wait for the "burnt bulb" fix, some black electrical tape will do.

 

[patprimmer]

Aquaduct

Are Buicks sold or privately imported into areas outside USA and Canada?

We haven't seen them here since the 60's, But GM here uses their V6 in a locally designed and made car, and we require OBD2 standards to be met and maintained, and checked via yearly roadworthy inspections after the car is 3 years old, and occasional impromptu or random inspections by the EPA

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.