[bluebee]

This quizzically precise suggest by Steve530 today got me wondering how the secondary air system (SAS) actually works:
- E39 (1997 - 2003) > 2001 530i question...

For the SAP, let the car sit overnight. Start it and let it idle for at least 2 minutes 13 seconds.

Doing a search for "SAS" in the bestlinks, I skimmed these threads to see if there was a good description of how the entire system works:
- How to maintain (1) (2) (3) & replace (1) & troubleshoot the BMW E39 SAS SAP valve secondary air pump system (1) (2) (3) (4) & an SAP valve group buy (1)

In particular, this cn90 DIY in that list above explained a good portion of the system:
- E39 (1997 - 2003) > DIY: Troubleshooting S.A.S. and How to Replace the Famous Fuse # 107!

This is the sequence of S.A.S. system when you start the car cold:
- ECU sends signal to the Electrical Valve (this Valve sits under the Intake Manifold), which in turn opens a small channel to allow vacuum from the Intake Manifold to be applied to the Air Valve (which sits on the Exhaust Manifold)
- At the same time, signal is sent to the Relay to close the 85-86 "primary" circuit, which in turn closes the 30-87 "secondary circuit". In general, the "secondary circuit" in most Bosch relay circuits controls the high current flow.
- In the case of the S.A.S. Air Pump, the 30-87 circuit is controlled by the Fuse # 107 ...

But, I don't see anything in there about 133 seconds (for example).

Q: Where is the best description of how the secondary air system works?

 

[bluebee]

Starting with cn90's listing (and generalizing for USA E39s) is this a complete list & photo of all SAS components?
- BMW-SAS-Failure.pdf

1. Air pump (just behind the passenger-side bumper)
2. Air valve (on the passenger side front of the engine)
3. Air valve gasket (between the air valve & the exhaust manifold)
4. Electrical valve (in the rear of the engine)
5. One-way check valve (next to the electrical valve)
6. Manifold vacuum port (in the rear of the engine)
7. Relay K6304 (under the passenger side cabin air filter housing)
8. 50A fuse #107 (under the passenger seat)

• 2002 M54 Engine => Vacuum control => AIR PUMP F VACUUM CONTROL

• 2002 M54 Engine => Exhaust manifold => EMISSION CONTROL-AIR PUMP

The pictures below & the wiring diagram above from the referenced cn90 thread.

REFERENCE:

- In order to understand the S.A.S. system, you need to read the attached pdf on S.A.S.:
---> In brief, when engine is cold, the Air Pump injects additional air into the exhaust to reduce pollutants.
---> The Air Pump is designed for high output but short run, so it injects air for anywhere between 2.5 seconds and 105 seconds or so, depending on engine temp.

- These are PNs for 1998 528i, listed only for reference, later years are slightly different. This is taken from www.realoem.com:
* Electrical Valve; PN 11747537612 (about $45)
* Air Valve: PN 11727540467 (about $110)
* Air Valve Gasket; PN 11727505259 (about $4)
* Pierburg Air Pump; PN 11721427911 (about $250)
* Air Pump Relay (K6304): schema is 85-86 and 30-87a-87, PN 12631742690 (about $8)
* Fuse #107: 50A Special Fuse: BMW PN 61138365901 ($4.00); Napa PN 782-1144 ($4.00).

 

[540iman]

Not 100% sure what you really want to know, but the ECU software is proprietary. The info BMW will share is that the SAP is turned off when a couple things happen. One is that the O2 sensor reaches a pre-determined temperature and the other is that the center point of the hydrocarbons emmissions voltage swing as seen by the O2 sensors starts to normalize leaner and the system believes that without the additional external air being injected, the emmissions will be within legal limits. The entire system is there because some cars can not meet the latest US emmissions standard during a cold start. A cold start does not have to mean that the O2 sensor is 20 degrees, it just means that the ignition has not been turned on in a sufficient amount of time that all the residual heat in the exhaust system has evaporated and the ECU assumes the
O2 sensor is at ambient temperature as is the block and heads. The part I'm not sure about is whether the ECU has a limit on how long it will allow the SAP to run regardless of what feedback it gets to protect the SAP motor. As we know, 4-5 minutes tops and the motor is toast. I hope this is kinda what you are seeking.

 

[Steve530]

It is my understanding that the purpose of the SAP is bring the oxygen sensors to operating temperature more rapidly, not dilute the high concentration of hydrocarbons in the exhaust during startup.

The SAP injects excess air into the exhaust to causes a lean condition. The ECU then injects more fuel which causes the catalyst to heat up faster. From what I've read, the sensors heat up enough to enter closed-loop operation within 2 minutes. What I don't know is how the ECU determines that the oxygen sensors are ready for closed loop operation.

 

[194449]

Pardon my ignorance, just want to ask something for my own edification....

The SAP on the i6 (and v8s?) only pumps air into the exhaust manifolds, not into the cylinders to create a lean condition there?

 

[Steve530]

Pardon my ignorance, just want to ask something for my own edification....

The SAP on the i6 (and v8s?) only pumps air into the exhaust manifolds, not into the cylinders to create a lean condition there?

Yes. The ECU controls the electric valve which supplies vacuum to open the SAP valve to route air from the air pump to the exhaust manifold. The one-way valve between the electric valve and intake manifold assures that no gases flow from the exhaust manifold through the electric valve to the intake manifold.

 

[Fudman]

It is my understanding that the purpose of the SAP is bring the oxygen sensors to operating temperature more rapidly, not dilute the high concentration of hydrocarbons in the exhaust during startup.

The SAP injects excess air into the exhaust to causes a lean condition. The ECU then injects more fuel which causes the catalyst to heat up faster. From what I've read, the sensors heat up enough to enter closed-loop operation within 2 minutes. What I don't know is how the ECU determines that the oxygen sensors are ready for closed loop operation.

Not exactly. Per the BMW TIS bulletin (which I can no longer access), the purpose of the SAP is to improve emissions performance. By adding air into the exhaust stream, before the cats, it improves complete combustion of unburned hydrocarbons in the catalytic converters. Actually, your theory reminds me of the "emissions system" in my old Lancia, which used an air pump to pump air into the exhaust stream. This car was pre-catalytic converter, in the days of leaded gas. That system was purely there to dilute the exhaust to pass emissions.

They bigger question is why is this SAP system needed since most other cars with cats don't have a SAP system? There are two theories for having a SAP:
1. During the first minute or so of engine operation after a cold statup, the engine's combustion process is less efficient, resulting in a larger % of unburned hydrocarbons.
2. Cold catalytic converters are less efficient than hot ones and require additional air to complete the catalytic process.
Either way explains why the system only operates during the first minute or so after cold startup. It believe it could be a combination of the two. I am guessing that the high performance BMW engine is prone to spew out more hydrocarbons at cold startup than a normal car and that the cats need an assist. Just my theory. Who knows... :dunno:

 

[JimLev]

This should clear up some of the mystery.
When you start the engine the DME is programmed to run the engine in a rich mode, cold engines won't run well at the proper A/F ratio.
So there is excess unburnt fuel in the exhaust system that the EPA frowns upon.
The SAP pumps air into the exhaust which causes the unburnt fuel to burn in the exhaust manifold.
This additional heat does 2 things.
It heats up the O2 sensors quicker so they will start working, cold sensors don't output a signal.
It also heats up the CATs so they function, cold CATs don't work either.
When the O2 sensors start functioning the DME goes into close loop mode, shortly after the SAP and it's valve close.

 

[Fudman]

This should clear up some of the mystery.
When you start the engine the DME is programmed to run the engine in a rich mode, cold engines won't run well at the proper A/F ratio.
So there is excess unburnt fuel in the exhaust system that the EPA frowns upon.
The SAP pumps air into the exhaust which causes the unburnt fuel to burn in the exhaust manifold.
This additional heat does 2 things.
It heats up the O2 sensors quicker so they will start working, cold sensors don't output a signal.
It also heats up the CATs so they function, cold CATs don't work either.
When the O2 sensors start functioning the DME goes into close loop mode, shortly after the SAP and it's valve close.

Now THAT is a logical and rationale explanation to the issue although the TIS bulletin indicated the actual combustion of the unburned hydrocarbons occurs in the cats and not in the exhaust manifold.

 

[JimLev]

Now THAT is a logical and rationale explanation to the issue although the TIS bulletin indicated the actual combustion of the unburned hydrocarbons occurs in the cats and not in the exhaust manifold.

Fudman, there could be some combustion in the pre CAT's too, however the exhaust coming out of the cylinders is very hot because it's closest to the cylinders. Now add some fresh air and unburnt fuel with a lot of heat and a little flame from the cylinders. I can't but help thinking there isn't burning fuel before the CAT.
I've had my exhaust system off (from pre CAT back) and looked at the exhaust out of the first CAT's, didn't see any flame when the engine was started and reved up when cold.

 

[540iman]

Guys, the BMW standard answer is that it gets the O2s warmer faster, but they give no explanation of how this happens. The truth is closer to what Fudman and JimLev said (and I) than what Steve says. The simple explanation without getting into a lot of detail is that many vehicles have SAP systems...Audis, Subarus, American cars as well. These cars are naturally being run rich at start-up and do not pass US emissions until the cars enters closed loop. There are unburned fuels being pumped into the air which the US does not like-these are hydrocarbons. In Europe, for example, if the entire SAP system fails, it does not even light a ses code! It is thought to be that insignificant. The reason BMW does not just state that the addition of fresh air dilutes the exhaust is that people can recognize that it is like diluting a drink. The same amount of booze is entering your system, but the "proof" is lowered. The fresh air dilutes the emmissions and whether it actually allows more hydrocarbons to burn is just academic. A "sniffer" in the tail pipe is "happy" with the extra air. Whether the O2s actually heat faster (remember they are heated already and get red hot quickly anyway) is moot. The standard is written such that emmissions are tested at the tail pipe and it is no different than if they stuck the sniffer into the tail pipe only 2 inches so it sucked in some outside air.

Gotta go with Fudman and JimLev on this one, but we all get the idea whichever way we see it.
This is what Power Chips offers when you buy a power chip. You have options like removing top speed, raising the rev limit, and switching to the Euro software for the SAP system. It will be in memory in the ECU, but will not set a SES thus allowing you to pass emmissions even if your heads have Excessive carbon build-up and you would normally with US software be setting the SES light.

 

[Fudman]

I stand corrected, Jim. I used combustion when I should have said oxidation. You are correct wrt location of any combustion. The combustion must occur in the manifold pre-cat. Any combustion in the cat would probably destroy or degrade the cat. What technically occurs in the cats is oxidation of the remaining hydrocarbons (converting them to non-toxic CO2 and H2O). By igniting the unburned hydrocarbons before the cat, you increase the temperature of the exhaust stream which then allows the cats to reach their operating temperature sooner.

BB: Here is the answer to your orginal question:

1. At cold startup, the DME uses a richer fuel/air mixture.
2. Simultaneously, the DME activates the SAP at cold startup.
3. The combustion in the chambers cannot burn all the hydrocarbons of the richer fuel mixture. In addition, the catalytic converters cannot oxidize all hydrocarbons as they are not yet at their operating temperature.
4. The addition of air into the exhaust manifold causes secondary combustion of the unburned fuel in the manifold.
5. The additional combustion of fuel in the exhaust stream increases the exhaust temperature.
6. This in turn causes the catalytic converters to reach their optimal operating temperature sooner.
7. As the car drives, the DME readjusts the fuel air mixture to normal parameters.
8. Once the catalytic converters reach their operating temp (60-90 seconds), the SAP shuts down.
9. At this point, the catalytic converters are operating normally.

Fundamentally, the SAP does two things:
1. It helps burn unburned hydrocarbons in the manifold.
2. It accelerates the process of bringing the cats up to standard operating temperature where they take over the process.

Given all that, I am less inclined to remove my SAP as this actually performs a useful function.

 

[540iman]

I bought you first answer, but disagree with this one. You ignore the dilution theory completely and I think if you talk to the tech at any Bimmer dealer, they would like everyone to believe what you say as it sounds neat and tidy, but the truth is that its not nearly as much what you just said as simply dilution of dirty air. Some of what you say happens, but that is an after-thought. Either way, I'm done with it. And you have no choice but to maintain your system. Even with PC software, all it does is open up the O2 sensor parameters of what is considered clean and what is considered dirty. The DME needs to see the O2 voltage swing in concert with the SAP air being there. If there is no swing of O2 voltage, it assumes there is no SAP air present whatsoever and will still trigger a SES light. Carry on guys!

 

[Steve530]

I think Fudman is fundamentally correct. I think the stopping of the SAP coincides with the injection system entering closed loop control. I'd like to know how the ECU decides to shut down the SAP. Is the temperature of the catalyst monitored?

Not to be contrary, but I'm not convinced that the purpose of the SAP is to dilute emissions. Here's my reasoning on this issue.

The emissions standards for vehicles are specified in g/mi. Obviously, not every car is tested. Instead, the manufacturer submits samples for certification. The certification process is to run the vehicle through the drive cycle (FTP) on a dynamometer while monitoring capturing all of the emissions. The results are reported in grams of pollutant. The mass of the pollutants are divided by the actual miles covered on the dynamometer. The results are compared to the standard. If that emission rate in g/mi meets the standard, the vehicle is certified. The manufacturer can then sell the vehicle as long as the emissions equipment is not changed.

Since all of the emissions are reduced to mass, diluting the emissions with excess air will not change the results. It's also important to realize that the drive cycle includes a cold start, a hot start, and different speeds. The total emissions are integrated over the entire drive cycle.

Here's a diagram of the emissions certification setup.

 

[Trebbia]

8. Once the catalytic converters reach their operating temp (60-90 seconds), the SAP shuts down.

How does it know the operating temp of the cats?

The bmw tis in the beginning of the 2nd post says: "After starting the engine, the actuation time of the secondary air pump is from 2.5 to 105 seconds depending on the engine temperature and engine speed."

I can only guess the dme gets engine temp from the dts and speed from the cps?

 

[cn90]

1+ to what JimLev said.

READ the attached pdf file by Pierburg and you will get all the info you need.

Happy Holidays to you all!

 

[Steve530]

Probably reads both coolant temperature and intake air temperature.

That excerpt from the TIS hints at an answer to the question as to how how the ECU decides to shut down the SAP. Based on that my guess is that the ECU reads the air and coolant temperatures and sets the SAP run time when the engine starts. Without further monitoring, the SAP shuts down after the run time is complete. Anyway, it's a theory.

 

[Steve530]

..READ the attached pdf file by Pierburg and you will get all the info you need....

Right, the purpose is to burn HC and CO which causes rapid catalyst heating as a side effect.

I've read that the oxygen senors will read a lean mixture at this time with voltages below 0.20 V.

 

[Trebbia]

READ the attached pdf file by Pierburg and you will get all the info you need

That pdf shows the dme turns off the sap based on four inputs

engine temp
engine speed
lambda sensor
throttle position

The temp of the cat isn't an input but maybe its inherent in the lamba sensor input?

 

[JimLev]

Good description in post #12 Fudman.
I'm not buying the dulition theory either (sorry Steve530) , the main purpose of the SAS supplied air is to complete the process of burning the rich mixture which gets the CAT's hot so they can oxidize what's left.
After the O2 sensors start producing a signal the DME knows that it's getting ready to shut down the SAP when the other inputs that Trebbia mentioned get to some pre-determined value.
Not exactly sure when the DME enters close loop mode, when my 540 is cold it idles at ~1500 RPM, as it warms up the idle begins to drop. As soon as the SAP shuts off the idle drops to 500 RPM.
Now I could connect my AutoEngunity and look at the DME and O2 sensors while reading the exhaust manifold temps (external digital meter) with the SAP running (higher temp) and then again with the SAP disconnected to see if SAP air is burning the rich mixture or just duliting it (lower temp).
However I don't think it's worth the time, this thread has been beaten to death. See you guys in some other post.