\\ Oxygen Sensors and Engine Damage //

[Radraddad]

I posted last year that I had issues getting my 84' 911 to start. After diagnosing and replacing many potential culprits, I took it to the only shop in the area that works on these cars. He got it running, and his description of how was cryptic. One of the things he did was to unplug the O2 sensor, and he said it was a "nice to have"

In other posts about cat bypasses, people mentioned that bypassing the O2 sensor could lead to fuel mixtures that might harm the engine over time. I recently removed the Cat and couldn't get the O2 sensor off the fitment, so I removed it with the part completely.

Question: Should I buy a new sensor for the cat bypass and re-attach it to the engine? If my car won't start with an O2 sensor, how should I approach repairing this mixture issue? Does it matter?

The car has been running great for the past six months, other than excessive white smoke on start-up now that I removed the Catalytic converter.

[GH85Carrera]

If it is adjusted properly, the O2 sensor should be plugged in, and doing its job. If it does not run right, or not at all when plugged in something is wrong and should be addressed.

[Dan J]

And.... our problem is?

[911obgyn]

From what i understand O2 sensors on motronic engines dont do anything until started and warmed up.

[pmax]

Originally Posted by Radraddad

I posted last year that I had issues getting my 84' 911 to start. After diagnosing and replacing many potential culprits, I took it to the only shop in the area that works on these cars. He got it running, and his description of how was cryptic. One of the things he did was to unplug the O2 sensor, and he said it was a "nice to have"

Find another shop or find the good mechanic at that shop.

In other posts about cat bypasses, people mentioned that bypassing the O2 sensor could lead to fuel mixtures that might harm the engine over time. I recently removed the Cat and couldn't get the O2 sensor off the fitment, so I removed it with the part completely.

Question: Should I buy a new sensor for the cat bypass and re-attach it to the engine? If my car won't start with an O2 sensor, how should I approach repairing this mixture issue? Does it matter?

The car has been running great for the past six months, other than excessive white smoke on start-up now that I removed the Catalytic converter.

The O2 sensor plays no part in whether the car starts during which the system's running open loop.

[EC900]

“(the O2 sensor), and he said it was a "nice to have"
Nice to have; as in Emissions Testing perhaps… nice for converting back to stock. ? ?

But a O2 sensor with/without won’t affect exhaust appearance.
As white smoke on start up, if it goes away after few minutes, it’s indicative of worn valve guide(s) seeping after prolonged sitting.
If white smoke remains persistent through out running, it’s worn rings.

Post a pic of the set up you had and what plan to install.

[AndrewCologne]

Im totally with Pmax.

@EC900
White smoke means evaporation of condensed "water", no burned Oil or overfueled mixture.
Blue smoke is caused by the culprits you mentioned.
Black smoke is cause by significant overfueled air/fuel combustion.

@Radraddat
Keep the sensor connected, it's not only a "nice to have" but if working properly it's a garant for proper fueling after engine has warmed up, means when the ECU/system switches from open to closed loop.
And here is where the problem lies with a disconnected sensor: If the "initial CO" has been set up at factory recommendation/default, means in a way where a working sensor is assumed/expected, then after the engine has been warmed up (and therefore with sensor disconnected a forced open loop state of the ECU) it could result in lean running engine with possible consequences, means in a "worst case" with overheating.

[yelcab1]

Need a better mechanic.

[Schulisco]

@Radraddad
That mechanic is a jerk. Go for one whose really experienced with the 3.2 Carreras ...

Running lean an engine can cause serious headaches...no matter if it's a caburetted, CIS or Motronic equipped engine, not so much when cruising, but going high load and revs on hot ambient temps evan can push an aircooled engine to the limits and above... So this topic has been discussed here several times:
https://forums.pelicanparts.com/porsche-911-technical-forum/555916-effects-car-running-lean.html#
https://forums.pelicanparts.com/porsche-911-technical-forum/652336-cis-running-lean-high-rpm.html

Thomas

[spuggy]

As Andrew says.

White == steam.
Black == fuel (rich)
Blue/grey == oil.

Water vapor is a by-product of combustion in hydrocarbon fuels. Because oxidization of hydrogen.

"White smoke on startup" is steam. It can sometimes also be a coolant leak. Pretty unlikely on an air-cooled motor...

Water vapor condenses out of the exhaust gas when it hits the cold headers, and the hot gases turn it back into steam. While the exhaust system is still warming up, actual water will often be visible trickling out of the muffler or blowing out in droplets.

Especially noticeable if its cold enough that moisture from your breath is condensing out in air. It'll do it until the exhaust system is hot enough the water vapor isn't condensing out anymore.

Which simply means you can't see it.

[kyngfish]

I’ve seen this debate so many times on these forums, it would be cool to see it resolved. There are people on here with a lot of knowledge that say the O2 sensor doesn’t really matter, and the Euro versions of the 3.2 don’t actually have O2 sensors and still run Motronic supposedly fine. There are at least 4 or 5 threads on Pelican over the years with people insisting it’s fine and others saying those people don’t know what they are talking about.

It would be good to have a clear thread about what the O2 sensor does or doesn’t do.

https://www.mye28.com/viewtopic.php?t=91137

According to that page, the fuel mapping is mainly managed by the Air Flow Sensor and Engine Speed, so I’m assuming Crank Sensor. With a modifier from a few other sensors possible - but it looks like in the 3.2 it would mainly be the CHT.

Where the O2 Sensor comes into play is in making adjustments to the mapping, but it doesn’t look like this is something it does on the fly, rather there’s an adjustment that happens on a delay. From other threads on here it sounds like with the O2 sensor off, the mix runs a bit richer, which is bad for your Cat.

Edit: So then the question is - if the O2 sensor is in the car, and can make make adjustments, then why would you disconnect it? Well, in the threads I read, the idle just seems smoother when the O2 sensor is disconnected, and if the O2 sensor is only present in the cars with a Cat, and a rich condition hurts the cat, then the question in my mind is, what if what it’s doing is making the mixture mostly leaner to protect the cat and for fuel economy or emissions, and because the adjustments present themselves on a DELAY, what happens is that an environmental factor makes the mix look rich, but since this presents itself as an adjustment on a delay, the real impact is crappy idle?


I’m not a motronic expert, but it sounds like the O2 sensor doesn’t play a huge role, and that without the O2 signal, Motronic corrects by running richer, which is bad for your cat, but not necessarily bad for the engine. And again, the Euro 3.2s without a cat, just don’t have an O2 sensor. So, if you’re running a cat, probably better to run the O2, and if not, maybe it doesn’t matter?

So really, which is it?

[spuggy]

Originally Posted by kyngfish

It would be good to have a clear thread about what the O2 sensor does or doesn’t do.

Steve Wong's web site describes the changes they make to the fuel/ignition maps - and why - in a fairly detailed manner: 911 Chips Motronic Fuel Maps

[mysocal911]

Originally Posted by kyngfish

I’ve seen this debate so many times on these forums, it would be cool to see it resolved. There are people on here with a lot of knowledge that say the O2 sensor doesn’t really matter, and the Euro versions of the 3.2 don’t actually have O2 sensors and still run Motronic supposedly fine. There are at least 4 or 5 threads on Pelican over the years with people insisting it’s fine and others saying those people don’t know what they are talking about.

It would be good to have a clear thread about what the O2 sensor does or doesn’t do.

https://www.mye28.com/viewtopic.php?t=91137

According to that page, the fuel mapping is mainly managed by the Air Flow Sensor and Engine Speed, so I’m assuming Crank Sensor. With a modifier from a few other sensors possible - but it looks like in the 3.2 it would mainly be the CHT.

Where the O2 Sensor comes into play is in making adjustments to the mapping, but it doesn’t look like this is something it does on the fly, rather there’s an adjustment that happens on a delay. From other threads on here it sounds like with the O2 sensor off, the mix runs a bit richer, which is bad for your CAT.

Edit: So then the question is - if the O2 sensor is in the car, and can make make adjustments, then why would you disconnect it? Well, in the threads I read, the idle just seems smoother when the O2 sensor is disconnected, and if the O2 sensor is only present in the cars with a Cat, and a rich condition hurts the cat, then the question in my mind is, what if what it’s doing is making the mixture mostly leaner to protect the cat and for fuel economy or emissions, and because the adjustments present themselves on a DELAY, what happens is that an environmental factor makes the mix look rich, but since this presents itself as an adjustment on a delay, the real impact is crappy idle?


I’m not a motronic expert, but it sounds like the O2 sensor doesn’t play a huge role, and that without the O2 signal, Motronic corrects by running richer, which is bad for your cat, but not necessarily bad for the engine. And again, the Euro 3.2s without a cat, just don’t have an O2 sensor. So, if you’re running a cat, probably better to run the O2, and if not, maybe it doesn’t matter?

So really, which is it?

The O2 system does not affect the fuel maps, but tweaks the fuel mixture such that Lambda (AFR air/fuel ratio) is optimized around 14.7
for proper catalytic converter functionality and fuel economy.

The 911 3.2 Lambda system with an AFM used a narrow band O2 sensor which just switched from a rich mixture (>.70 volts)
to a lean mixture (<.40 volts) to optimize the mixture for the CAT. This system resulted in a small idle "hunting" (~ +/- 50 RPM).
The later 993 ('96 & later) with a MAF sensor used wide band O2 sensors, a more precise mixture tweaking with negligible "hunting".
Additionally, those 993s had rear O2 sensors to measure the CAT efficiencies.

So if a 911 3.2 has a CAT installed, the O2 needs to function properly and be installed or the CAT will be damaged.

The graphs here should be helpful understanding Lambda and the O2 system;https://www.systemsc.com/graphs.htm

[proporsche]

the 1st day i got my 911 back in 1996 i took off the CAT and disconnect the O2 sensor.On the ECu i bypass the lambda circuit by a switch.So,if i had to go for smog in CA i put back the Cat and sensor plus flip the switch ......always passed with flying colors.The CAT is off now for many many miles..since then let`s say 200K ,911 runs like a champ.....

Ivan

[pmax]

Originally Posted by mysocal911

The O2 system does not affect the fuel maps, but tweaks the fuel mixture such that Lambda (AFR air/fuel ratio) is optimized around 14.7
for proper catalytic converter functionality and fuel economy.

Running cleaner can't be worse for an engine.

The 911 3.2 Lambda system with an AFM used a narrow band O2 sensor which just switched from a rich mixture (>.70 volts)
to a lean mixture (<.40 volts) to optimize the mixture for the CAT. This system resulted in a small idle "hunting" (~ +/- 50 RPM).

Can't speak for anyone else but the idle in my 87 is rock solid.

So if a 911 3.2 has a CAT installed, the O2 needs to function properly and be installed or the CAT will be damaged.

An "out of tune" engine will reduce the CAT's and the engine's lifetimes. So will too many short trips.

Without the closed loop feedback control, the ECU is just making its best guess as to how well the engine is running at the moment.

[mysocal911]

Originally Posted by pmax

Running cleaner can't be worse for an engine.



Can't speak for anyone else but the idle in my 87 is rock solid.



An "out of tune" engine will reduce the CAT's and the engine's lifetimes. So will too many short trips.

Without the closed loop feedback control, the ECU is just making its best guess as to how well the engine is running at the moment.

Actually, the DME ECM does not make "its best guess". It's like any other system operating in open-loop, e.g. the engine could be over-heated or knocking and DME ECM would have no idea.
There's no "branch" in the 911 3.2 DME ECM EPROM firmware to "guessing"! As an example, the temp sensor has no effect on the fuel mixture once the temp sensor exceeds a value.

[AndrewCologne]

Originally Posted by mysocal911

It's like any other system operating in open-loop, e.g. the engine could be over-heated or knocking and DME ECM would have no idea.

That brings it to the point.


@pMax
Air/Fuel mixture generated by map values are more or less "static" as the values in the map on the chip are fixed and so AFR results do mainly depend on rpm and load. So no "guess" but a simple Air/Fuel setting based on the given map value at XXXX rpm and XX kpa load.
This is enough if performance is the target, but beside the aspect that an additional O2-Sensor based mixture adjustment does provide ideal combustion and so avoid lean or overfueling running, it's also the only way which makes the AFR before combustion resulting in a wanted lambda 1 after the combustion --> as this is exactly needed to end up in emissions which do match with the official requirements, no matter if in California or Germany or whatever country with exhaust emission limits. Without a Lambda 1 hitting fuel mixture control you won't pass emission tests.

[proporsche]

Originally Posted by AndrewCologne

That brings it to the point.


@pMax
Air/Fuel mixture generated by map values are more or less "static" as the values in the map on the chip are fixed and so AFR results do mainly depend on rpm and load. So no "guess" but a simple Air/Fuel setting based on the given map value at XXXX rpm and XX kpa load.
This is enough if performance is the target, but beside the aspect that an additional O2-Sensor based mixture adjustment does provide ideal combustion and so avoid lean or overfueling running, it's also the only way which makes the AFR before combustion resulting in a wanted lambda 1 after the combustion --> as this is exactly needed to end up in emissions which do match with the official requirements, no matter if in California or Germany or whatever country with exhaust emission limits. Without a Lambda 1 hitting fuel mixture control you won't pass emission tests.

Most of States in USA do not require smog test..i believe

Ivan

[AndrewCologne]

A lot States and there will be more and more.

https://www.findlaw.com/traffic/drivers-license-vehicle-info/state-emissions-standards-and-testing.html
But I guess each state comes with different requirements i.E. regarding the year the car was manufactured. That's why (IIRC) in California it's not permitted to change cat based exhaust systems on SCs to i.E. SSIs or similar.

I read that in Washington the environment air is that clean that their emission check program has ended

[proporsche]

Originally Posted by AndrewCologne

A lot States and there will be more and more.

https://www.findlaw.com/traffic/drivers-license-vehicle-info/state-emissions-standards-and-testing.html
But I guess each state comes with different requirements i.E. regarding the year the car was manufactured. That's why (IIRC) in California it's not permitted to change cat based exhaust systems on SCs to i.E. SSIs or similar.

I read that in Washington the environment air is that clean that their emission check program has ended

To be quite honest the emission laws since the 1980`s in CA worked pretty good for the environment..You should have seen the yellow mushrooms clouds over LA back in those days..
Also, back in 1998 i even passed the smog test without the CAT and OXY sensor...Of course the limits have changed a lot since then....basically, if your 911 is set up correctly they run very clean

Ivan