Technical Ducato excessive MAF reading / black smoke

[Ducato Nat]

Good evening. As a newcomer to this forum, I was wondering if somebody could give me some guidance / advice. I have a 2014 Ducato X250, 2.3 jtd engine code F1AE3481D, 160K Miles.
I bought the van a year ago for conversion to a camper. When I got it, it was reluctant to start, eventually narrowed that down to excessive leak off from the injectors preventing sufficient fuel rail pressure on crank. I changed injectors on cyl 1 & 2, starts no problem now.
The issue I have is black smoke under acceleration. When it's ticking over it's fine, passed MOT no problems a couple of weeks ago. I thought maybe smoke issue was due to new injectors not being coded to the ecu so I asked the garage to do that when they had it for MoT. Their tool was unable to communicate with the injector module in the ECU so I took it to an injector specialist, same problem. He did say that the fuel trims were well within spec so he didn't think injectors were a problem.
So further probing showed me that the MAF was reading 17g/s at idle, which seems excessive. Also the MAP was reading 99kpa, also excessive. I pulled the MAP and it was choked with gunk so I thought I'd better have a look in the intake manifold but found it to be clean when I got it off. So the two sensor readings indicated to me that the issue may lie with the throttle body. If I understand correctly the MAF gives a reading of airflow into the engine which the ECU then modulates using the throttle body. The MAF reading also determines how much fuel to inject, so if there's too much airflow everything else gets thrown out. The problem I have is that the throttle body isn't modulating. It remains in the wide open position except on engine shut down when it closes for about 5 seconds. Is this normal or have I got hold of completely the wrong end of the stick regarding how the ECU controls airflow and fuel flow? I've fitted new throttle body also MAF sensor but it seems both of the old ones were functioning normally and the issue lies elsewhere. So if anybody can perhaps shed some light on what's going on with the airflow and also why the injector coding seems to be off limits I would be grateful, since I'm out of ideas now. Also my apologies for the length of this question.
Regards, Nat.

 

[camden]

I have similar issue with X250. It's smoking under acceleration (sometimes even under light accel.) and my MAF reading is even higher - around 21g/s at idle. Also lambda value is constantly around 2 (if I'm calculating it correctly) which suggests there is way too much oxygen in exhaust and ECU compensates with extra fuel (I think?).
Other than smoke, van idles and drives absolutely fine.

Found cracked air intake pipe (after MAF) but replacing it didn't make any difference. Still looking for cause.
I will be blanking off EGR soon. Will let you know if that makes any difference.

Btw, throttle body is normally fully open if I'm not mistaking. See screenshot.

 

[Ducato Nat]

Yeah the injector specialist I talked to recommended to delete the EGR. I stripped the system to inspect it, found the valve itself to be pretty clean but the cooler (internal fin type) quite badly fouled with hard carbon. I replaced the cooler with an internal tube type, has made no difference. Not that I expected it to since EGR is not active during a acceleration. Like you I suspect excessive O2 in the exhaust is causing ECU to overload fuel but I can't get to grips with why it's happening. The only code I get is PO130 (O2 sensor) but I'm pretty sure it's not the sensor faulty rather it is reading values out of the expected range.

 

[theoneandonly]

Their tool was unable to communicate with the injector module in the ECU so I took it to an injector specialist, same problem. He did say that the fuel trims were well within spec so he didn't think injectors were a problem.
So further probing showed me that the MAF was reading 17g/s at idle, which seems excessive. Also the MAP was reading 99kpa, also excessive. I pulled the MAP and it was choked with gunk so I thought I'd better have a look in the intake manifold but found it to be clean when I got it off. So the two sensor readings indicated to me that the issue may lie with the throttle body. also why the injector coding seems to be off limits I would be grateful, since I'm out of ideas now. Also my apologies for the length of this question.
Regards, Nat.

Nat
injectors first
a member here had a DPF problem due due to overfuelling . His injectors were removed and deemed perfect dpf cleaned but coking problem returned. Using multiecuscan he was able to show 2 and 4 were new injectors not read to the ecu and 1,3 were replaced in wrong order.
Injector tollerances have been tightening up with time.
MAF reading 17g/s And MAP reading how were they obtained. multiecuscan measures air flow as mg/l

 

[Ducato Nat]

Nat
injectors first
a member here had a DPF problem due due to overfuelling . His injectors were removed and deemed perfect dpf cleaned but coking problem returned. Using multiecuscan he was able to show 2 and 4 were new injectors not read to the ecu and 1,3 were replaced in wrong order.
Injector tollerances have been tightening up with time.
MAF reading 17g/s And MAP reading how were they obtained. multiecuscan measures air flow as mg/l

MAP and MAF readings are obtained from ECU using Foxwell 650 elite tool. That's also bothering me that 3 different tools are unable to establish comms with the injector module, including the injector specialist who had never seen that error message before. He said possibly it's locked and only accessible to a fiat dealer tool. He can read the trim data but can't alter injector codes. So I know for sure cyls 1 + 2 aren't coded to the ECU but have not been able to rectify that.

 

[theoneandonly]

MAP and MAF readings are obtained from ECU using Foxwell 650 elite tool. That's also bothering me that 3 different tools are unable to establish comms with the injector module, including the injector specialist who had never seen that error message before. He said possibly it's locked and only accessible to a fiat dealer tool. He can read the trim data but can't alter injector codes. So I know for sure cyls 1 + 2 aren't coded to the ECU but have not been able to rectify that.

Multiecuscan can be used to code injectors to the ecu. Have you access to a copy?

 

[Ducato Nat]

Multiecuscan can be used to code injectors to the ecu. Have you access to a copy?

No. I'll find out what it is and see if I can get one. I've always used the foxwell unit (which is supposed to have that function) and an icarsoft unit for my Renault cars.

 

[Ducato Nat]

Multiecuscan can be used to code injectors to the ecu. Have you access to a copy?

Do I need the registered version or the multiplex version in order to access the injector coding?

 

[theoneandonly]

Do I need the registered version or the multiplex version in order to access the injector coding?

I dont know i have a registered version and checked the codes no problem.
I have a second PC with the free version, tommorrow i will try to see if it reads the code but wont try to change it.

 

[Ducato Nat]

I dont know i have a registered version and checked the codes no problem.
I have a second PC with the free version, tommorrow i will try to see if it reads the code but wont try to change it.

Thank you for your time

 

[theoneandonly]

The Free version does not give you access to the modules you need. The simulate mode shows what it can do and is usful in playing around without actualy doing anything to the engine. It at least proves you can comunicate with the van.
The easiest way of getting a working solution is via Gendan using the OBDLINK SX usb connector and a yellow adaptor. For using with a Ducato you do not need the expense of the multiplex version but it does have the advantage of using on more than 1 PC (analagous to the way VCDS is used for Vag cars).

 

[Ducato Nat]

Yeah thanks for that. I ordered the cables this morning including the yellow one. I'll probably download direct from multiecuscan website hopefully I'll have the kit and give it a close looking at this weekend.

 

[Anthony489]

Hi Nat

I don't know the exact cause of your problem, but here are some basics you might find useful.

The F1AE3481D engine is Euro 5, and has a single EGR system plus Catalyst and DPF

A back of envelope calculation for a 2287 cc engine at 800 RPM idle suggests 13 grams per second mass air flow. Given the number of variables and the assumptions I had to make, 17 grams per second doesn't sound too far off.

A manifold pressure of 99 kilopascals is effectively atmospheric pressure, which is what you would expect as the turbo isn't producing any appreciable boost at idle so you are just getting almost unrestricted airflow from the atmosphere. I don't know the maximum boost for this engine but let's say 200 kilopascals so that's about what you should see when accelerating. If the reading doesn't change it indicates a blocked sensor that isn't seeing the pressure presented to it.

The injector coding is "icing on the cake", and helps with fuel economy, emissions and smooth running. I'm doubtful if incorrect coding would throw basic running off very much, but clearly it's best to get coding done. MultiECUscan should allow this.

Historically, the throttle body was added as part of the EGR system to manage manifold pressure and thereby encourage EGR flow, traditional diesels had no need for throttling. The flap will only partially shut during specific running conditions e.g. light load cruising and after idling for a minute or two. It also shuts during switch off to prevent judder, but that's just a bonus. The rest of the time it should be wide open.

It's a characteristic of diesels that power is controlled by varying the amount of fuel, whereas the airflow is prety much unrestricted. As long as there is enough air (i.e. 20% oxygen with no EGR) to burn the fuel, then the exhaust is clean. The exhaust will contain some unused oxygen. However, even a tiny bit of extra fuel beyond the level that the oxygen can support will result in black smoke, i.e. carbon particles from partially burnt fuel. In this situation there will be little or no oxygen content in the exhaust.

If the airflow is less than the ECU "thinks", or the fuel injected is more than the ECU "thinks", black smoke can result when significant power is demanded. Since the ECU relies on accurate sensor data to know what's going on, it's essential that they are all working. As an example, in addition to the MAF and MAP readings, it also needs to know air temperature to correct for varying air density, fuel temperature, exhaust oxygen etc.

I don't know how the exhaust oxygen sensor data is used, but I imagine the ECU expects to see a little bit of remaining oxygen as an indication that operation is below the smoke limit, and may tweak the fuelling down if it isn't. The DPF won't take kindly to being presented with excessive carbon particles.

If sensor readings deemed to be outside their expected range, you may be lucky and get an error code logged but that's not guaranteed

 

[Ducato Nat]

Hi Nat

I don't know the exact cause of your problem, but here are some basics you might find useful.

The F1AE3481D engine is Euro 5, and has a single EGR system plus Catalyst and DPF

A back of envelope calculation for a 2287 cc engine at 800 RPM idle suggests 13 grams per second mass air flow. Given the number of variables and the assumptions I had to make, 17 grams per second doesn't sound too far off.

A manifold pressure of 99 kilopascals is effectively atmospheric pressure, which is what you would expect as the turbo isn't producing any appreciable boost at idle so you are just getting almost unrestricted airflow from the atmosphere. I don't know the maximum boost for this engine but let's say 200 kilopascals so that's about what you should see when accelerating. If the reading doesn't change it indicates a blocked sensor that isn't seeing the pressure presented to it.

The injector coding is "icing on the cake", and helps with fuel economy, emissions and smooth running. I'm doubtful if incorrect coding would throw basic running off very much, but clearly it's best to get coding done. MultiECUscan should allow this.

Historically, the throttle body was added as part of the EGR system to manage manifold pressure and thereby encourage EGR flow, traditional diesels had no need for throttling. The flap will only partially shut during specific running conditions e.g. light load cruising and after idling for a minute or two. It also shuts during switch off to prevent judder, but that's just a bonus. The rest of the time it should be wide open.

It's a characteristic of diesels that power is controlled by varying the amount of fuel, whereas the airflow is prety much unrestricted. As long as there is enough air (i.e. 20% oxygen with no EGR) to burn the fuel, then the exhaust is clean. The exhaust will contain some unused oxygen. However, even a tiny bit of extra fuel beyond the level that the oxygen can support will result in black smoke, i.e. carbon particles from partially burnt fuel. In this situation there will be little or no oxygen content in the exhaust.

If the airflow is less than the ECU "thinks", or the fuel injected is more than the ECU "thinks", black smoke can result when significant power is demanded. Since the ECU relies on accurate sensor data to know what's going on, it's essential that they are all working. As an example, in addition to the MAF and MAP readings, it also needs to know air temperature to correct for varying air density, fuel temperature, exhaust oxygen etc.

I don't know how the exhaust oxygen sensor data is used, but I imagine the ECU expects to see a little bit of remaining oxygen as an indication that operation is below the smoke limit, and may tweak the fuelling down if it isn't. The DPF won't take kindly to being presented with excessive carbon particles.

If sensor readings deemed to be outside their expected range, you may be lucky and get an error code logged but that's not guaranteed

Thanks Anthony for clearing a few things up for me. I'm pretty confident now that everything upstream of the engine as far as airflow is concerned is working as it should be. So next on my to do list is code the two new injectors to the ECU. If that fails to show improvement, which I suspect it may, I'll probably pull all four injectors and get them bench tested.

 

[theoneandonly]

Thanks Anthony for clearing a few things up for me. I'm pretty confident now that everything upstream of the engine as far as airflow is concerned is working as it should be. So next on my to do list is code the two new injectors to the ECU. If that fails to show improvement, which I suspect it may, I'll probably pull all four injectors and get them bench tested.

Did you sucessfully code the new injectors?

 

[Ducato Nat]

Did you sucessfully code the new injectors?

I've been meaning to post an update on that. Oddly, the answer is yes but also no. So injector 2 is coded using multiecuscan, no problem. Injector 1 however, using the same process looks like it's doing it but when checked in Parameters, the new code hasn't updated. So I swapped the injectors 1 + 2, same result. I can code either injector into cyl 2 but neither into cyl 1. Weird.
Anyway the better data collected by multiecuscan as opposed to my previous tool has given me something else to think about. The P0130 lambda sensor code which I thought was simply from the sensor reading being out of expected range turned out to be in fact the sensor not returning any data at all. So further probing, it looks like a wiring issue to pre cat o2 sensor. I've been trying to identify plug d004 where the wires from o2 sensor terminate so I can do a continuity check and hopefully find which is broken. Anyway that's about where I'm up to with it.

 

[theoneandonly]

Have you got any MES data i could look at especialy dodgy O2 sensors

 

[Ducato Nat]

Have you got any MES data i could look at especialy dodgy O2 sensors

I'm otherwise engaged at the moment, I'll screenshot it tomorrow but it boils down to 0v reading from above cat o2 sensor leading to a p0130 01 error logged.

 

[Ducato Nat]

I'm otherwise engaged at the moment, I'll screenshot it tomorrow but it boils down to 0v reading from above cat o2 sensor leading to a p0130 01 error logged.

 

[Ducato Nat]

Have you got any MES data i could look at especialy dodgy O2 sensors

See above