[Dormermike]

Originally Posted by jirka330i

I had mine less sooty but ultrasonic cleaning did not help...I was told that there is some rhodium and other strange parts (apparently kind of very sensitive layers...) in the probe and this mostly why the probe dies...but you can always try; if it does not help keep the nox unit ( I see that on the picture that schriss has Generation03...) and then you can, like myself, experiment with a connection to some other probes (the worst thing which can happen is that you kill the analogue part of nox unit...I hope...)

Interesting. I see 8 wire generic nox sensors are available. Sounds like you tried replacing just the head but that didn't work out?

[jirka330i]

Originally Posted by Dormermike

Interesting. I see 8 wire generic nox sensors are available. Sounds like you tried replacing just the head but that didn't work out?

I do not believe that there are any serious generic nox probes on the market...and that is why I have not spent some 100+ euro on what I consider fake...I rather prefer testing other (wideband..) probes with my old oem nox unit...my son likes plaing with arduino and I like playing with nox :-)
If you are aware about any...post a photo here...I will try to comment...

[schriss]

I think it would be easy to use Arduino or something similar as an emulator for the Probe itself, sending signals to the electronic unit. The only problem is the harsh conditions under the car. On a plus side, Arduino/Raspberry could send lower signals than actual probe, same as NOXEM, meaning reporting much lower nox ppm.

[jirka330i]

Originally Posted by schriss

I think it would be easy to use Arduino or something similar as an emulator for the Probe itself, sending signals to the electronic unit. The only problem is the harsh conditions under the car. On a plus side, Arduino/Raspberry could send lower signals than actual probe, same as NOXEM, meaning reporting much lower nox ppm.

I would not put arduino under the car...connector+cable would allow bit more flexibility for the location...for example in the boot :-)

[schriss]

Originally Posted by jirka330i

I would not put arduino under the car...connector+cable would allow bit more flexibility for the location...for example in the boot :-)

Yea, shielded cable to a safe place, but I think definitely it could be done. One thing is I don't know what kind of signals the Probe is sending, but if they are analog like from Lambda sensors (just some voltage range) then probably easy. If it's some data protocol then not easy. By I would think probe<>module is just analog, module<>car is some protocol (but we don't care).

[jirka330i]

Originally Posted by schriss

Yea, shielded cable to a safe place, but I think definitely it could be done. One thing is I don't know what kind of signals the Probe is sending, but if they are analog like from Lambda sensors (just some voltage range) then probably easy. If it's some data protocol then not easy. By I would think probe<>module is just analog, module<>car is some protocol (but we don't care).

I have the same view...probe delivers some analogue signal (say 5-500mV) to analogue part of the unit...where it is processed into set of data in canbus protocol

something like this could read the communication between ecu and nox sensor easily and if it has 2way function then even send back the emulated data... http://www.rs.canlab.cz/?q=en/node/68

[Dormermike]

The probe is 8 wire but those include o2 and heating. O2 should be easy to spoof. The nox value is probably a voltage range but would need to monitor a good one live.

The other potential issue is if the DME uses upstream sensor values to corroborate or sanity check the values coming from the nox ecu. I.e. WOT but the nox sensor is seeing nox associated with light throttle.

Would make a great arduino recipe!!

[schriss]

Yes would be very cool "fix" for nox probes. Basically a software emulator. Knowing the voltage range of outputs of real probe is main thing. Actually Inpa reports some voltage from real NOX sensor, but that is probably some different value not related to nox ppm.

[jirka330i]

Originally Posted by Dormermike

The probe is 8 wire but those include o2 and heating. O2 should be easy to spoof. The nox value is probably a voltage range but would need to monitor a good one live.

The other potential issue is if the DME uses upstream sensor values to corroborate or sanity check the values coming from the nox ecu. I.e. WOT but the nox sensor is seeing nox associated with light throttle.

Would make a great arduino recipe!!

I do not say it is easy...I just believe it is possible...and even at lower costs with no residue lambda probe (you would just need a m20x1,5 screw to blind out exhaust)

and yes...with oem nox as a theoretical core unit there would be a little challenge to manage signals of 2 aditional wires; with arduino and like no wires needed :-)

however I still search for a cleaner (and reversible) way of mapping the nox system out in ecu...

by the way... noxem has 6 wires between unit and probe..guess why?

and moreover I noticed today that they say at their main web page the following about noxem: " Bosch latest generation sensor controlled and measured by ourselves developed analogue control module provides highest accuracy of exhaust Lambda value. Now you can use NOx emulator Lambda data as reference"...and wonder whether the same statement was there in May when I purchased noxem...anyone remembers better???

[jirka330i]

Originally Posted by schriss

Yes would be very cool "fix" for nox probes. Basically a software emulator. Knowing the voltage range of outputs of real probe is main thing. Actually Inpa reports some voltage from real NOX sensor, but that is probably some different value not related to nox ppm.

I spoke to a guy today who would be able to "emulate" a canbus communication between ecu and nox unit...but he would need to know the detail of the communication protocol (not the content but rather the form/process)...between msd80 and the nox..
anyone aware where this could be found?

[schriss]

After installing NOXEM and coding sensor replacement and nox cat replacement and doing adaptations reset I did some driving around 60 km or so.
MPG for mixed driving is now 29. Strange enough, when idling, mpg seems to be increasing slowly and coming down a little when driving.
Also it runs uneven on idle, same as before I put NOXEM in it. I thought engine/transmission mounts were the problem, but no, replaced them and same thing still.
Can feel vibrations in the seat and on gear stick when idling with something like slightly stronger jolts every now and then.
Coils all replaced (most were arched) but spark plugs still old.

[Dormermike]

I'd change the plugs.

29 mixed is not great is it?

I used to get a solid 23-24mpg for a 2 mile each way run to the supermarket in my f10

[jirka330i]

Originally Posted by schriss

After installing NOXEM and coding sensor replacement and nox cat replacement and doing adaptations reset I did some driving around 60 km or so.
MPG for mixed driving is now 29. Strange enough, when idling, mpg seems to be increasing slowly and coming down a little when driving.
Also it runs uneven on idle, same as before I put NOXEM in it. I thought engine/transmission mounts were the problem, but no, replaced them and same thing still.
Can feel vibrations in the seat and on gear stick when idling with something like slightly stronger jolts every now and then.
Coils all replaced (most were arched) but spark plugs still old.

Did you delete all adaptations? Flywheel adaptations succeeded??
Encoding with Inpa or Ista? Inpa (some versions?) sometimes fails on some (few) operations( like proper new battery encoding)...Ista is more reliable in this respect

[schriss]

Originally Posted by jirka330i

Did you deleted all adaptations? Flywheel adaptations succeeded??
Encoding with Inpa or Ista? Inpa (some versions?) sometimes fails on some (few) operations( like proper new battery encoding)...Ista is more reliable in this respect

All done with Ista, yea. Did flywheel adaptations 3 times while driving, but tomorrow will look at INPA, will check Vanos (I cleaned them but they are oem old) and rough run menu, maybe injector leaking, and will check various statuses including adaptations. Also will check misfire counters. And I think I need to check in INPA if I can switch to homogen-lean mode (or stratified with rpm 2500).
Car seems to be worse on cold start with rpm dipping and gets better idle after driving.

[jirka330i]

Originally Posted by schriss

All done with Ista, yea. Did flywheel adaptations 3 times while driving, but tomorrow will look at INPA, will check Vanos (I cleaned them but they are oem old) and rough run menu, maybe injector leaking, and will check various statuses including adaptations. Also will check misfire counters. And I think I need to check in INPA if I can switch to homogen-lean mode (or stratified with rpm 2500).
Car seems to be worse on cold start with rpm dipping and gets better idle after driving.

a friend of mine had an issue with pedal adaptation after new nox encoding...rpm jumping etc...he found the "how to" with google and it helped
what you describe could be the new adaptations stil unfinished (but also old not deleted properly...i some cases the process is very sensitive when partial steps need to be stored in ecu...ignition off and key out until car get to sleep mode

also flywheel adaptation can be quite tricky...perhaps because of firmware in ecu...in my case i needed to go over 4000rpm at 4th geear and then let it fall down below 2000rpm...and 1 attepmt was enough

[Andystobbs]

Originally Posted by schriss

All done with Ista, yea. Did flywheel adaptations 3 times while driving, but tomorrow will look at INPA, will check Vanos (I cleaned them but they are oem old) and rough run menu, maybe injector leaking, and will check various statuses including adaptations. Also will check misfire counters. And I think I need to check in INPA if I can switch to homogen-lean mode (or stratified with rpm 2500).
Car seems to be worse on cold start with rpm dipping and gets better idle after driving.

Have you ever had timing chain replaced?

[schriss]

Originally Posted by Andystobbs

Have you ever had timing chain replaced?

Unlikely.

[Dormermike]

Odd question to ask - n53 timing chains are pretty solid aren't they?

[schriss]

One weird thing happened: in heavy start-stop traffic, I moved and went to stop before red light and a second after coming to full stop I felt like something switched and engine noise was different quieter, became super smooth vibrations gone almost as if it wasn't even running. Not sure if it ever happened again but that was like I guess how the engine supposed to work.

[jirka330i]

Originally Posted by schriss

One weird thing happened: in heavy start-stop traffic, I moved and went to stop before red light and a second after coming to full stop I felt like something switched and engine noise was different quieter, became super smooth vibrations gone almost as if it wasn't even running. Not sure if it ever happened again but that was like I guess how the engine supposed to work.

good sign I think...perhaps you start to get fuel adaptations completed...be patient

[schriss]

Hopefully. In the meantime I still want to change sparkplugs (looked at 4 of them, 2 were light grey good but 2 has some soot, but not too much and so far no fuel smell but need to recheck) and look at INPA into Rough Run menu (injectors), VANOS menu (https://bimmerprofs.com/rough-run-vanos/) and the one that shows which mode the engine runs in. And use phone to see ODB Mode 6 results for misfire codes.
All that rough running I only noticed after car bought, on cold start. At that time I didn't do any adaptations resetting yet, I did nothing to the car. But it had "Nox cat Aging" code so no strat mode for sure.

[Andystobbs]

I was wondering about balance shaft timing if a chain had been done.
As a rule of thumb I'd say there is no such thing as a good quality BMW timing chain on a modern engine.