Secondary Air Injection (SAI) Bosch ECU strategies .. Help!!

Ask for and share advice on using the PicoScope kit to fix vehicles here.
User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 377
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK
Contact:

Secondary Air Injection (SAI) Bosch ECU strategies .. Help!!

Post by FioranoCars » Sat Nov 26, 2016 11:07 am

Hi All

Another odd-ball question, Petrol Engine stuff ... sorry but unable to find concrete answers elsewhere, and know someone here will have an answer ... please ... :?

I know that the Bosch MED9.1.1 uses the Lambda sensor to confirm SAI airflow is taking place and it's flow is indeed adequate. As I understand, it checks during activation of between 2sec-8secs (different numbers from different sources - hot/cold difference?) each start cycle (not just cold start) and confirms from Lambda sensors that it's seeing/registering the extra oxygen. The SAI is used in part to help mix with the extra fuel it cycles to heat the catalyst faster (during cold start only?) but also to charge the catalyst with oxygen (Hot/Warm/cold start?). The SAI pumps longer than the monitoring, again can find no information to say if either Water temp, Exhaust Themo sensors or Lambda's (via a thermo calc - Hey Steve, make any progress on this?) are used to control the switch off of the SAI, or it's just timed from an initial calculation or stored map ... Let me know if you can add/subtract/correct this?

But I've got a Bosch ME7.1.1 and can't find any data on the strategy for monitoring (other than pump relay and solenoid coil), and REALLY would like to find out if the Lambda sensor is used to confirm flow? Or some other technique?

Oddly this particular case is throwing on the MIL during driving, not at/just after starting, so will be replacing the relay and checked the fuse board generally / solenoid operation including added vibration during/after commanded operation (tapping it gently to see if the error can be repeated) ...

We have tested pump flow into the exhaust on both banks, by manual opening the valves and powering the pump, and there is good flow on both branches, and the vacuum lines/solenoid command system are all 100% functioning correctly during tests. So on the face of it there appears to be good operation, other than any ECU strategy (like lazy lambda sensors) ... :?

Being an anorak, I started searching to see if previous generation systems used the lambda for SAI "closed loop" monitoring, like the M5.2 which is an ECU we see a lot of but except BMW is quite rare (unlike the M5.2.1 which is used extensively, but is completely different) or M2.7 - both of which have the ability to control a SAI setup, but no data on their strategies - for my own records any help on how these 2 work (M5.2 or M2.7) would be welcome too. I'm confident that the M2.5 did not utilise SAI ... but let me know!

I tried to look at the ME9.1.1 (as this is the next Gen of the ME7.1.1 where as the MED9.1.1 is a different branch so might not be indicative of the ME family) but can't find much on that either ... again any links/facts/pointers welcome.

Why bother, well in looking at 2 cars with SAI issues this week (a MED9.1.1 with P0491) and then this ME7.1.1 I realised our normal approach for SAI was distinctly lacking, and worse each of us had a different understanding of the systems operation, and went about testing completely differently ... so I'm taking the opportunity to write up a simple diagnostic procedure and explanation to cover this element, as while we would each find the cause easily enough, we might not all have been as thorough or tested the full system at the same speed ... and it can be hard to do so, especially with our cars packaging.

I will try and grap some Pico captures of the SAI pump with Solenoid controls verse Lambda, to show prior to SAI, during and after, with any valve shuffles in case this used to calibrate the readings. Might be impossible to decipher as the extra fuel injected would be very hard to capture the metering levels (or tell me otherwise!) ...

Thanks and have a great weekend!
Richard Lukins

KimAndersen
TwoWaves
TwoWaves
Posts: 206
Joined: Fri Nov 30, 2012 2:53 pm
Location: Denmark

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by KimAndersen » Sat Nov 26, 2016 12:29 pm

Hi Richard

I can help you with Bosch ME7.1.1 though I have documentation of the OBD Bosch ME9 - but SAI is not used in this version.

Bosch ME 7.1.1 makes use of lambda values along with air mass values as seen in this chart. SAI system looks at two phases passive and active to make the right decisions, also the exhaust gas tenperature sensor used in the passive phase, but again this is for BOSCH ME7.1.1 - BOSCH MED9.1.1 can have a completely different strategy, but I doubt it !!!.

Following charts shows the phases of the AIR monitoring via lambda deviation for passive and active monitoring.
BOSCH_SECONDARY_AIR_SYSTEM_MONITORING.jpg
BOSCH_SECONDARY_AIR_SYSTEM_MONITORING
Regards
Kim :wink:

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 377
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK
Contact:

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by FioranoCars » Sat Nov 26, 2016 5:51 pm

Thanks Kim
Looks like the ME7.1.1 was capable of working either with temperature sensors or from Lambda calculations for exhaust temperature, as the MY2004 versions dropped the thermo probes but retained the same core ECU, but a different software flash no doubt.

From the diagram, I'm unsure of the reason to run the air pump with the valves closed, as the systems we have this would achieve nothing, the pipe work is simple and the valve is the only exit! Just increase the amps drawn as it creates pressure ... unless it then opens the valves in some sort of burst mode (on/off) ?

I'm not sure I correctly understood the bit about ME9.x, did you mean that you don't know it's strategy or that it 100% for sure does not use SAI monitoring ... we have cars with ME9.1.1 that have SAI ?

Thanks
Richard

KimAndersen
TwoWaves
TwoWaves
Posts: 206
Joined: Fri Nov 30, 2012 2:53 pm
Location: Denmark

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by KimAndersen » Sat Nov 26, 2016 7:23 pm

Let me rephrase what I rewrote.

I know the strategy of Bosch ME7 Secondary Air System from documentation that I have. What I was uncertain about was, if the Bosch ME7 and BOSCH MED 9.1.1 uses the same strategy to monitor the Secondary Air System.

On what car are you working !

Do you have any DTC !

Regards
Kim

Dave Hill
TwoWaves
TwoWaves
Posts: 63
Joined: Fri Jun 22, 2007 8:15 am

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by Dave Hill » Sat Nov 26, 2016 7:27 pm

FioranoCars wrote:
From the diagram, I'm unsure of the reason to run the air pump with the valves closed, as the systems we have this would achieve nothing,
Plausibility test (leak detection) ?

Perhaps?

:?

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 377
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK
Contact:

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by FioranoCars » Sat Nov 26, 2016 7:30 pm

Hi Kim
I know that the MED9.1.1 does use the Lambda to "close loop" monitor the SAI.
The ME9.1.1 I'm not sure about
And the ME7.1.1 you are sure about :D

your comment
I can help you with Bosch ME7.1.1 though I have documentation of the OBD Bosch ME9 - but SAI is not used in this version.
gave the impression that the ME9 is not using SAI control via lambda?

Sorry, ME9 verse MED9 might sound like small change, but very different systems!
Thanks
Richard

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 377
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK
Contact:

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by FioranoCars » Sat Nov 26, 2016 7:39 pm

Hi Dave
Not sure how it can, and that's not a flippant response, it's there to be corrected ...

but the Pump is relay controlled so the ECU only knows the coil ~80ohm/100-120ma load, and not the switched load, so it can't see the pump current load increase, or in fact if the pump is even connected! (same with most fuel pump monitoring!)

It does monitor the solenoid valve that controls the vacuum control, and that's direct, but back to pump, what use does pumping have when the systems output is closed? Maybe different cars have other routing of pipes, but for this car ... not

Sorry Kim
It's a Maserati 4200 Coupe MY 2003. Has Thermo probes :)
Don't have the P code in front of me, sorry paperwork is at the office :oops:

But it was a wider research than the one/two cars this week with similar issues, as it showed were we could improve our knowledge, and all use a common/best practice approach to diagnose in future ...

Thanks for continued input
Richard

KimAndersen
TwoWaves
TwoWaves
Posts: 206
Joined: Fri Nov 30, 2012 2:53 pm
Location: Denmark

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by KimAndersen » Sun Nov 27, 2016 11:14 am

Hi Richard

This document from Maserati may be of interest to you. It covers both Bosch Motronic ME 7.1.1 and Bosch Motronic ME 9.1.1 with regards to documentation of the secondary air system - see page 270 and 291 for more info.

I think there are some variables to the many different versions of Bosch Motronic ME system - for example when you look at page 270 it states that " The lambda sensors are disabled for the entire duration of the cycle " which does not correlate with function chart, that I previously had uploaded - well, it was also from another version.
MASERATI_TRAINING_DOCUMENTATION_2009.pdf
Maserati Training Documentation 2009
(25.33 MiB) Downloaded 145 times
Regards
Kim

User avatar
FioranoCars
TwoWaves
TwoWaves
Posts: 377
Joined: Thu Dec 13, 2012 11:26 am
Location: London, UK
Contact:

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by FioranoCars » Sun Nov 27, 2016 11:39 am

Thanks Kim
Found a version of this somewhere this morning, not read it all yet, but also note the last 2 pages (bar back cover) show the OBD pinouts!

Will read it today,
KimAndersen wrote:page 270 it states that " The lambda sensors are disabled for the entire duration of the cycle "
I think they mean the ECU ignores them for normal fuel adjustment purposes, as they are being used for the self test SAI monitoring ... and Catalyst Pre-heating etc ... So worth remembering not to try lambda monitoring in the first 90secs from startup as they will be doing strange things ... as will ECU parameters ... best leave the car for a good 2minutes post starting before reading anything in to the data you see!
Dave Hill wrote:Plausibility test (leak detection) ?
Ok, my first reply was made without due thought process ... thinking about air flowing, not system sealing/integrity!! Sorry, but yes of course the system checks that it can see the extra Oxygen, then closes the valves but pumps on, to see if the seals for the valves are indeed intact and the valves are neither stuck open or closed.

Too close to the coal face, and not thinking of the total system! Sorry. :oops: :oops: :oops:

And then P270 rubs my nose in it, but the penny did drop before I read it, honest !! :shock:

Thanks to you both, Dave and Kim

Best
Richard

User avatar
STC
Banned
Posts: 790
Joined: Sun Sep 13, 2015 2:10 am

Re: Secondary Air Injection (SAI) Bosch ECU strategies .. He

Post by STC » Mon Nov 28, 2016 1:41 pm

Richard
Any Chance of the Fault code, read through OBD and VM Specific Diagnosis.

Then Scope trace of SAI Pump Current, Valve Solenoid Current and both pre cat O2's from cold with 2 minutes across the screen ?

Post Reply