None that I am aware of. @lycopersicum The various regens I have analysed have all occurred at the trigger point of 100% - interesting that you have observed 130%. The vehicles I have observed have been 2.3L dual EGR's, which vehicle is triggering at 130%?
Technical Twin EGR issues - 'EGR Cooler Syndrome' Symptoms but different Fault Codes - HELP!
- Thread starter Dancingbear
- Start date
Currently reading:
Technical Twin EGR issues - 'EGR Cooler Syndrome' Symptoms but different Fault Codes - HELP!
lycopersicum
Member
Mine is a 2.3 Euro 5. I suspect Car Scanner might be forgetting to convert to a percentage as 130 is shortly after the holy number 128 (power of two). But maybe the Euro 5 is truly different.
Thanks
Dancingbear
Member
Ok, I've captured 2 x DPF auto regen cycles, using Template 0 as suggested (thanks @deejays). The first regen stopped after 30 seconds - not sure why?
Attached are cvs files for both, Abbreviated data as follows:
Regen 1
Regen 2
So, assuming these auto regens are indicative of a 'good' DPF performance, i am trying to summarise my position:
1) This engine appears to be running well with no obvious issues. It drives well, has no EML, no other fault codes and ALL parameters across ALL observable systems seem to be functioning within tolerance.
HOWEVER,
2) there is still soot in the exhaust (I cleaned the tailpipe vigorously with brake cleaner and it was sooty again within 150km - residual or new? - how would i tell?). Several EGR related parts have already been replaced, along with the Throttle body (which failed due to water ingress), although i have NOT had the P0401 or LPEGR Cooler related codes.
3) The captured DPF data suggests a correctly functioning unit (please advise if this is wrong?? .csv data attached) MES tells me the DPF has never been replaced, so the soot is unlikely to be from an old 'resolved' problem, and I have to assume the DPF is compromised, Right??? Everything I know to be true tells me that Clean tailpipe = Intact DPF and conversely Sooty tailpipe = Cracked DPF
4) Having studied the schematic provided by @theoneandonly (https://www.fiatforum.com/guides/how-does-a-twin-egr-engine-work.964/) it seems that (although DPF parameters are functioning correctly) sooty air will likely be corrupting the entire engine via the LPEGR cooler, which may still be blocking (P0401 inbound in the near future???)
I feel like I've exhausted the diagnostics, and looking forward I guess I have 3 options:
A) Mr Comprehensive: Replace the DPF (and also get the LPEGR cooler examined / cleaned) at great expense £££?
B) Mr Cautious: Monitor DPF performance for the foreseeable future accepting that other components may fail as a result of the DPF compromise?
C) Mr Chilled: forget about it and drive it care free, fingers crossed that the throttle body replacement has cured an underlying cause?
What are your thoughts everyone? Have i summarised fairly? What would you do?
I really appreciate the knowledge you bring here and the pragmatic approach to problem solving, thank you
Attached are cvs files for both, Abbreviated data as follows:
Regen 1
| duration | 30 seconds |
| % soot | 13.8% loss |
| max temp | 768 degC |
| Av. Temp | 664 degC |
Regen 2
| duration | 360 seconds |
| % soot | 92% loss |
| max temp | 726 deg |
| Av temp | 654 degC |
So, assuming these auto regens are indicative of a 'good' DPF performance, i am trying to summarise my position:
1) This engine appears to be running well with no obvious issues. It drives well, has no EML, no other fault codes and ALL parameters across ALL observable systems seem to be functioning within tolerance.
HOWEVER,
2) there is still soot in the exhaust (I cleaned the tailpipe vigorously with brake cleaner and it was sooty again within 150km - residual or new? - how would i tell?). Several EGR related parts have already been replaced, along with the Throttle body (which failed due to water ingress), although i have NOT had the P0401 or LPEGR Cooler related codes.
3) The captured DPF data suggests a correctly functioning unit (please advise if this is wrong?? .csv data attached) MES tells me the DPF has never been replaced, so the soot is unlikely to be from an old 'resolved' problem, and I have to assume the DPF is compromised, Right??? Everything I know to be true tells me that Clean tailpipe = Intact DPF and conversely Sooty tailpipe = Cracked DPF
4) Having studied the schematic provided by @theoneandonly (https://www.fiatforum.com/guides/how-does-a-twin-egr-engine-work.964/) it seems that (although DPF parameters are functioning correctly) sooty air will likely be corrupting the entire engine via the LPEGR cooler, which may still be blocking (P0401 inbound in the near future???)
I feel like I've exhausted the diagnostics, and looking forward I guess I have 3 options:
A) Mr Comprehensive: Replace the DPF (and also get the LPEGR cooler examined / cleaned) at great expense £££?
B) Mr Cautious: Monitor DPF performance for the foreseeable future accepting that other components may fail as a result of the DPF compromise?
C) Mr Chilled: forget about it and drive it care free, fingers crossed that the throttle body replacement has cured an underlying cause?
What are your thoughts everyone? Have i summarised fairly? What would you do?
I really appreciate the knowledge you bring here and the pragmatic approach to problem solving, thank you
Last edited:
Hi there, getting past my bedtime, so I will look at the files in detail tomorrow. But, just from your tabulated data, the peak DPF temperatures are higher than anything I have ever seen, same goes for the average temp, but not as marked. The duration is in the ballpark for the second burn - perhaps a bit shorter than normal. What time gap between ending the first file and starting the second file? And was there anything unusual occurring that may have stopped the first regen that you were aware of?Ok, I've recorded 2 x DPF auto regen cycles. The first stopped after 30 seconds - not sure why? Attached are cvs files for both, Abbreviated data as follows:
1)
duration 30 seconds % soot 13.8% loss max temp 768 degC Av. Temp 664 degC
2)
duration 360 seconds % soot 92% loss max temp 726 deg Av temp 654 degC
So, i am trying to summarise:
What now stands out to me now, based on everything you guys have helped me learn, is this: This engine appears to be running well with no obvious issues. It drives well, has no EML, no other fault codes and ALL parameters across ALL observable systems seem to be functioning within tolerance.
HOWEVER,
1) there is soot in the exhaust (I cleaned the tailpipe vigorously with brake cleaner and it was sooty again within 150km). Several EGR related parts have already been replaced, along with the Throttle body (water ingress), although i have NOT had the P0401 or related codes.
2) The DPF numbers support a correctly functioning unit (please advise if this is wrong??), but because of the soot in the exhaust i have to assume the DPF has been compromised, Right??? MES tells me the DPF has never been replaced, so the soot is unlikely to be from an old 'resolved' problem, and everything i know to be true tells me that intact DPF = clean tailpipe.
3) Having studied the schematic provided by @theoneandonly (https://www.fiatforum.com/guides/how-does-a-twin-egr-engine-work.964/) it seems that (although DPF parameters are functioning correctly) sooty air will likely be corrupting the entire engine via the LPEGR cooler, which may still be blocking (P0401 inbound in the near future???)
I feel like I've exhausted the diagnostics, and looking forward I guess I have 3 options:
A) Mr Comprehensive: Replace the DPF (and also get the LPEGR cooler examined / cleaned) at great expense £££?
B) Mr Cautious: Monitor DPF performance for the foreseeable future accepting that other components may fail as a result of the DPF compromise?
C) Mr Chilled: forget about it and drive it care free, fingers crossed that the throttle body replacement has cured an underlying cause?
What are your thoughts everyone? Have i summarised fairly? What would you do?
I really appreciate the knowledge you bring here and the pragmatic approach to problem solving, thank you
Dancingbear
Member
@deejays they were on separate drives about 3.5 hours apart. The first (failed) was at about 60 - 70 mph cruising on the motorway - speed changes but no aggressive acceleration etc??. The second (complete) was slower, 50 - 60mph on country roads with roundabouts etc.
For background info, both initiated themselves within .1 over 100% soot load.
Thankyou
For background info, both initiated themselves within .1 over 100% soot load.
Thankyou
Thanks and goodnight, will check tomorrow.@deejays they were on separate drives about 3.5 hours apart. The first (failed) was at about 60 - 70 mph cruising on the motorway - speed changes but no aggressive acceleration etc??. The second (complete) was slower, 50 - 60mph on country roads with roundabouts etc.
For background info, both initiated themselves within .1 over 100% soot load.
Thankyou
Below is a typical regen of my vehicle:
Below is same as above, but zoomed in to observe detail at beginning of regen:
Below is your first file zoomed in on the beginning of the failed regen:
Below is your second file for the entire regen:
The file from my engine shows relatively tight regulation of the exhaust temperatures around 630C during the regeneration period. Your engine, for some reason is not regulating those temperatures correctly and is most likely the reason the regeneration was prematurely terminated in the first file due to the excessive overshoot of the DPF temperature to 768C. This would be to (amongst other things) protect the DPF from damage due to excessive temperature. The second file shows it getting to 754C, but it continued and completed the regen (perhaps due to it occurring at lower engine demands than the previous – so less power/heat generated from normal operation), but the temperature regulation is not tightly controlled, varying from 576C to a high 754C. The regen duration is OK and completed satisfactorily, but the temperature regulation is not controlled correctly. Is this due to the temperature sensor response speed, the injection quantity, the ECU or the DPF?
The sensors all track each other well as expected during normal operation so maybe it is the response speed of the DPF temperature sensor. There is something causing a lag in the temperature regulation feedback loop, either that or there is an excess or over-fuelling in the post injection causing a faster rise in temperature than the feedback loop is designed to control. The injectors were well within the FIAT idle test specification from previous test however. Apart from the temperature regulation and overshoot, everything else was behaving as expected – so the rest of the parameters are good.
That is, except for the Throttle Position(!) For all engines I have observed, the throttle position sits around 8% to 9% closed with spikes both positive and negative. Your original (faulty?) throttle was also displaying that same response – it appeared OK.
Immediately after its replacement with the new one, the CSV file then showed it to be hovering around 6% closed – more open than “normal” by 2% to 3%.
However, in the two latest files it is being shown as fully open, only traversing from 0.05% to 0.36% - so something is amiss there.
So, perhaps the throttle body and connections need to be revisited. One question- does the engine shut down normally with ignition off, or does it run on a bit and not stop cleanly – as that is an indication of a throttle body not closing completely with engine off?
Keep an eye on regen parameters, and the throttle.
Apart from that (and the sooty exhaust and likely blocking cooler) all else looks fine.
My engine is running ECU SW Version 32. Which SW Version is yours?
Below is same as above, but zoomed in to observe detail at beginning of regen:
Below is your first file zoomed in on the beginning of the failed regen:
Below is your second file for the entire regen:
The file from my engine shows relatively tight regulation of the exhaust temperatures around 630C during the regeneration period. Your engine, for some reason is not regulating those temperatures correctly and is most likely the reason the regeneration was prematurely terminated in the first file due to the excessive overshoot of the DPF temperature to 768C. This would be to (amongst other things) protect the DPF from damage due to excessive temperature. The second file shows it getting to 754C, but it continued and completed the regen (perhaps due to it occurring at lower engine demands than the previous – so less power/heat generated from normal operation), but the temperature regulation is not tightly controlled, varying from 576C to a high 754C. The regen duration is OK and completed satisfactorily, but the temperature regulation is not controlled correctly. Is this due to the temperature sensor response speed, the injection quantity, the ECU or the DPF?
The sensors all track each other well as expected during normal operation so maybe it is the response speed of the DPF temperature sensor. There is something causing a lag in the temperature regulation feedback loop, either that or there is an excess or over-fuelling in the post injection causing a faster rise in temperature than the feedback loop is designed to control. The injectors were well within the FIAT idle test specification from previous test however. Apart from the temperature regulation and overshoot, everything else was behaving as expected – so the rest of the parameters are good.
That is, except for the Throttle Position(!) For all engines I have observed, the throttle position sits around 8% to 9% closed with spikes both positive and negative. Your original (faulty?) throttle was also displaying that same response – it appeared OK.
Immediately after its replacement with the new one, the CSV file then showed it to be hovering around 6% closed – more open than “normal” by 2% to 3%.
However, in the two latest files it is being shown as fully open, only traversing from 0.05% to 0.36% - so something is amiss there.
So, perhaps the throttle body and connections need to be revisited. One question- does the engine shut down normally with ignition off, or does it run on a bit and not stop cleanly – as that is an indication of a throttle body not closing completely with engine off?
Keep an eye on regen parameters, and the throttle.
Apart from that (and the sooty exhaust and likely blocking cooler) all else looks fine.
My engine is running ECU SW Version 32. Which SW Version is yours?
Last edited:
Dancingbear
Member
@deejays thanks again for a comprehensive and clear analysis. To answer your questions:
1) the van shuts down clean and fast, no lag or spluttering.
2) My van is software version 28 and a manual gearbox - I understood V28 to be the latest for manual, and V32 for auto?? So is yours automatic? And if so, could the auto / manual difference account for any of the observed anomalies above??
It is perhaps noteworthy that the full regen analysed above was the first genuine auto regen done in ages, as the engine management light was on for a sustained period, then it failed a regen with 170% clogging, the DPF cleaner guy did his thing and then i had to build up enough soot for the auto regens analysed above - perhaps it will 'settle down' or provide more acceptable figures moving forwards with no cleaning residue etc????
You raised the question: "Is this due to the temperature sensor response speed, the injection quantity, the ECU or the DPF?" I don't really know where to start with answering this but if I am on the correct SW version 28 (?) I could get the injectors professionally sent away for testing (would this provide any info I don't already have?) Could I possibly replace the DPF temperature sensor (but this is probably integral to or co-located with the DPF making it expensive to access?) or just suck it up and get the DPF replaced and monitor from there???
I guess moving forwards in the short term, I will aim to continue capturing all regens and also monitor the throttle performance / position. I'll aim to come back with another 2 or 3 full regens and see if there are any differences to what we've already observed..
As always, thanks for your time, knowledge and thoughts!
1) the van shuts down clean and fast, no lag or spluttering.
2) My van is software version 28 and a manual gearbox - I understood V28 to be the latest for manual, and V32 for auto?? So is yours automatic? And if so, could the auto / manual difference account for any of the observed anomalies above??
It is perhaps noteworthy that the full regen analysed above was the first genuine auto regen done in ages, as the engine management light was on for a sustained period, then it failed a regen with 170% clogging, the DPF cleaner guy did his thing and then i had to build up enough soot for the auto regens analysed above - perhaps it will 'settle down' or provide more acceptable figures moving forwards with no cleaning residue etc????
You raised the question: "Is this due to the temperature sensor response speed, the injection quantity, the ECU or the DPF?" I don't really know where to start with answering this but if I am on the correct SW version 28 (?) I could get the injectors professionally sent away for testing (would this provide any info I don't already have?) Could I possibly replace the DPF temperature sensor (but this is probably integral to or co-located with the DPF making it expensive to access?) or just suck it up and get the DPF replaced and monitor from there???
I guess moving forwards in the short term, I will aim to continue capturing all regens and also monitor the throttle performance / position. I'll aim to come back with another 2 or 3 full regens and see if there are any differences to what we've already observed..
As always, thanks for your time, knowledge and thoughts!
According to page 7 of FIAT Service News 10-023.19 the last update was V32 for the Manual Transmission (as well). Yes, mine it the Comfortmatic, but V32 applies to the manual box engine as well.
I have no idea what the difference is between the versions, but mine was originally on V24 and was updated direct to V32, but on both versions the regen temp was the same. I did not have MES back then, but another scan tool which showed a similar graph on V24 – so perhaps not a big deal there.
But there is something wrong with the regen – and it won’t be related to the SW version – there is a fault somewhere. I am not an expert by a long shot – just have some experience in the engine and analysis as I have me head around it, but perhaps a bleed down test of the injectors which can be done in situ (again no expert in this area).
You might be right about the unusual soot loading in the DPF, or even internal structure damage causing this issue – sort of an overhang from the past. Time will tell – keep observing regens closely. Feedback loops are tricky to analyse – trying to figure out which came first – very much like the chicken and the egg. Is it injectors supplying too much fuel, to fast, or is it a slow temp sensor response, or is it something odd with the DPF structure. It is unlikely to be the ECU. The temp sensors track reasonably well – the DPF and post DPF sensors just get out of kilter during the regen. I know that if either of the three temp sensors drift outside the expected operational envelope with respect to each other that will generate a fault code – and so far, that has not happened.
The throttle is a bit weird – keep an eye on that – it must be closing off OK on shutdown.
So plan B sounds the way to go – keep observing.
I have no idea what the difference is between the versions, but mine was originally on V24 and was updated direct to V32, but on both versions the regen temp was the same. I did not have MES back then, but another scan tool which showed a similar graph on V24 – so perhaps not a big deal there.
But there is something wrong with the regen – and it won’t be related to the SW version – there is a fault somewhere. I am not an expert by a long shot – just have some experience in the engine and analysis as I have me head around it, but perhaps a bleed down test of the injectors which can be done in situ (again no expert in this area).
You might be right about the unusual soot loading in the DPF, or even internal structure damage causing this issue – sort of an overhang from the past. Time will tell – keep observing regens closely. Feedback loops are tricky to analyse – trying to figure out which came first – very much like the chicken and the egg. Is it injectors supplying too much fuel, to fast, or is it a slow temp sensor response, or is it something odd with the DPF structure. It is unlikely to be the ECU. The temp sensors track reasonably well – the DPF and post DPF sensors just get out of kilter during the regen. I know that if either of the three temp sensors drift outside the expected operational envelope with respect to each other that will generate a fault code – and so far, that has not happened.
The throttle is a bit weird – keep an eye on that – it must be closing off OK on shutdown.
So plan B sounds the way to go – keep observing.
@Dancingbear @theoneandonly
The weather here today in Southern Australia is bitterly cold with a biting wind, so what better thing to do than spend a bit more time analysing your data!
I decided to look at things from a slightly different perspective, now that we are zooming in on the granular details of difference. The issue is overshoot of the DPF temp (dangerously so - enough to trigger premature cessation of regen), coupled with the Post DPF temp lagging and both never reaching an equilibrium point of around 620C - 630C.
The main possible suspects are injector over-fuelling, the ECU, the Temp sensors, or the DPF itself.
Let us assume that the ECU is most likely not the issue. The same can be said for the injectors because we know they are within the FIAT specification for the Injector Correction test, and during numerous test-drives the Total Fuel Quantity matches precisely the Desired Fuel Quantity. So that leaves two temp sensors and/or the DPF.
Now, this is where we need to do some specific comparisons between the good and the bad Regen data and consider applying some reverse engineering.
I suspect that the ECU uses both the DPF and Post DPF temp sensors generally in combination, but it appears to be primarily using the Post DPF sensor for regulation of the approximate 625C set point and the DPF sensor for the max limit protective shut-down point. At least this reverse engineering (correct or not) can explain what is happening with your engine.
If we look at the various time durations for each engine, one point really stands out. For the bad engine the time taken for the DPF sensor to register 600C from initial Post Injection is significantly shorter than the good engine, however conversely, it takes significantly longer for the Post DPF sensor to reach 600C. It also takes significantly longer for the bad engine’s Post DPF sensor to register 620C.
This indicates that either the Post DPF sensor itself has an issue, or there is something happening inside the DPF canister that is causing the Post DPF sensor to not able to accurately measure the true or bulk gas flow temperatures. But which one is correct (if either assumption is correct!)?
The faster rise time of the DPF sensor in the bad engine can be explained by the larger initial application of Post Injection, but the much longer rise time of the Post DPF sensor cannot be explained by this – in fact it should be the other way around – and that is the bad engine’s Post DPF sensor should have a faster rise time because of that.
One way to find out is to replace the Post DPF Temp sensor and that should be relatively easy to get at. Prior to that it would be worthwhile disconnecting the exhaust output of the DPF to have a look inside with a borescope before anything further is done.
Good engine: From initial Post Injection, 60S for DPF to 600C, 112S for Post DPF to 600C
Bad engine: From initial Post Injection, 45S for DPF to 600C, 185S for Post DPF to 600C
Good engine: From main Post spike Inj, 40S for DPF to 600C, 92S for Post DPF to 600C
Bad engine: From main Post spike Inj, 26S for DPF to 600C, 166S for Post DPF to 600C
Good engine: Time for Post DPF to reach 620C: 120S/100S
Bad engine: Time for Post DPF to reach 620C: 194S/175S
Please check my measurements to be sure, but the above data is what I derive from the two graphs below.
Good engine:
Bad engine:
The weather here today in Southern Australia is bitterly cold with a biting wind, so what better thing to do than spend a bit more time analysing your data!
I decided to look at things from a slightly different perspective, now that we are zooming in on the granular details of difference. The issue is overshoot of the DPF temp (dangerously so - enough to trigger premature cessation of regen), coupled with the Post DPF temp lagging and both never reaching an equilibrium point of around 620C - 630C.
The main possible suspects are injector over-fuelling, the ECU, the Temp sensors, or the DPF itself.
Let us assume that the ECU is most likely not the issue. The same can be said for the injectors because we know they are within the FIAT specification for the Injector Correction test, and during numerous test-drives the Total Fuel Quantity matches precisely the Desired Fuel Quantity. So that leaves two temp sensors and/or the DPF.
Now, this is where we need to do some specific comparisons between the good and the bad Regen data and consider applying some reverse engineering.
I suspect that the ECU uses both the DPF and Post DPF temp sensors generally in combination, but it appears to be primarily using the Post DPF sensor for regulation of the approximate 625C set point and the DPF sensor for the max limit protective shut-down point. At least this reverse engineering (correct or not) can explain what is happening with your engine.
If we look at the various time durations for each engine, one point really stands out. For the bad engine the time taken for the DPF sensor to register 600C from initial Post Injection is significantly shorter than the good engine, however conversely, it takes significantly longer for the Post DPF sensor to reach 600C. It also takes significantly longer for the bad engine’s Post DPF sensor to register 620C.
This indicates that either the Post DPF sensor itself has an issue, or there is something happening inside the DPF canister that is causing the Post DPF sensor to not able to accurately measure the true or bulk gas flow temperatures. But which one is correct (if either assumption is correct!)?
The faster rise time of the DPF sensor in the bad engine can be explained by the larger initial application of Post Injection, but the much longer rise time of the Post DPF sensor cannot be explained by this – in fact it should be the other way around – and that is the bad engine’s Post DPF sensor should have a faster rise time because of that.
One way to find out is to replace the Post DPF Temp sensor and that should be relatively easy to get at. Prior to that it would be worthwhile disconnecting the exhaust output of the DPF to have a look inside with a borescope before anything further is done.
Good engine: From initial Post Injection, 60S for DPF to 600C, 112S for Post DPF to 600C
Bad engine: From initial Post Injection, 45S for DPF to 600C, 185S for Post DPF to 600C
Good engine: From main Post spike Inj, 40S for DPF to 600C, 92S for Post DPF to 600C
Bad engine: From main Post spike Inj, 26S for DPF to 600C, 166S for Post DPF to 600C
Good engine: Time for Post DPF to reach 620C: 120S/100S
Bad engine: Time for Post DPF to reach 620C: 194S/175S
Please check my measurements to be sure, but the above data is what I derive from the two graphs below.
Good engine:
Bad engine:
theoneandonly
Prominent member
I had hoped to spend today sorting out a proper reply . but just seen @deejays reply and think he needs more data. The regen data is not as tight as he thinks. (one engine maybe)
www.fiatforum.com
This at least shows different engines.
mine can be from about 605C max to 670 max. I need to look at some historic data and check the temperatures and initial temp rate versus qty of fuel injected for regen. @Dancingbear It may be injector or injector demand (ecu) issue? There is no doubt its on the high side and the Ramprate 500-600 C @14C/s is much greater than my examples about 4C/s.
Page 2 | DPF Regeneration
Passive regeneration is the third process in which the raised temperature attained durring normal running is high enough to burn off the excess soot in the DPF. The following examples (2 different engines) show this can occur in Ducatos but the...
www.fiatforum.com
mine can be from about 605C max to 670 max. I need to look at some historic data and check the temperatures and initial temp rate versus qty of fuel injected for regen. @Dancingbear It may be injector or injector demand (ecu) issue? There is no doubt its on the high side and the Ramprate 500-600 C @14C/s is much greater than my examples about 4C/s.
Dancingbear
Member
thanks both! frustratingly i've just missed a regen - bizarrely the soot dropped to 10% (without me recording data) but the 'km since last regen' did not reset - don't know if this is significant!? I'll keep collecting data and go from there. I possibly have access to a borescope and imagine i'm just checking for structural integrity of whatever is the first thing i find within the DPF?
I'm also awaiting a response from my local fiat dealer with regard to updating the software version from 28 to 32 and a full injector check - taking them out and sending them away - presumably this would be a gold standard rather than the bleed out test? I've also asked them about regen temperature ranges etc but i expect to get nothing back from them about this - pretty sure they won't know
I will wait for the fiat dealer reply and collect some more data. I've also considered getting AI involved for analysis - any thoughts or experience of this?
I'm also awaiting a response from my local fiat dealer with regard to updating the software version from 28 to 32 and a full injector check - taking them out and sending them away - presumably this would be a gold standard rather than the bleed out test? I've also asked them about regen temperature ranges etc but i expect to get nothing back from them about this - pretty sure they won't know
I will wait for the fiat dealer reply and collect some more data. I've also considered getting AI involved for analysis - any thoughts or experience of this?
theoneandonly
Prominent member
What does a regen look like? well heres some more examples and with a little bit of playing with numbers (nothing fancy)
The max and mean are calculated are based on dpf temperature being above 600C limmit. The Rate of increase is thrate based on the climb between 500-600C. The last is total time of fuel being injected durring the regen cycle. The previous 2 are the QTY injected for 5 measurement cycles and total injected. The observant will know the QTY is mm3/ injection so not a clean comparisson as RPM is involved, may have to think on that. I thought it more important you see it sooner with warts.
Dancing Bear results are outside this grouping
The max and mean are calculated are based on dpf temperature being above 600C limmit. The Rate of increase is thrate based on the climb between 500-600C. The last is total time of fuel being injected durring the regen cycle. The previous 2 are the QTY injected for 5 measurement cycles and total injected. The observant will know the QTY is mm3/ injection so not a clean comparisson as RPM is involved, may have to think on that. I thought it more important you see it sooner with warts.
Dancing Bear results are outside this grouping
lycopersicum
Member
I agree that the difference in rate of change of the two temperature sensors is significant and could explain the overshoot.
It looks like the ECU regulates the slope of the slower curve (temp after DPF)*, so the other one overshoots and potentially triggers an abort of the procedure.
* On the basis of the fact that the upstream temperature stops rising as soon as the downstream temperature starts rising.
Now, why is that sensor so much slower?
(a) it's faulty / not OEM specs: I find that unlikely. It would have to have a much larger thermal mass; I can't imagine that the thermal mass of the sensor is so significant given the large heat flux around it.
(b) it's not seeing a normal exhaust volume? This could be:
- a blockage;
- an exhaust leak;
- a crack in the DPF that diverts the exhaust away from the sensor
- a layer of carbon deposits on the sensor that insulates it
(c) this is all wrong. It's not slower; the upstream temperature rises faster because of that large early injection spike. The downstream temperature rises at a normal-looking speed because of the thermal mass of the DPF in-between.
It looks like the ECU regulates the slope of the slower curve (temp after DPF)*, so the other one overshoots and potentially triggers an abort of the procedure.
* On the basis of the fact that the upstream temperature stops rising as soon as the downstream temperature starts rising.
Now, why is that sensor so much slower?
(a) it's faulty / not OEM specs: I find that unlikely. It would have to have a much larger thermal mass; I can't imagine that the thermal mass of the sensor is so significant given the large heat flux around it.
(b) it's not seeing a normal exhaust volume? This could be:
- a blockage;
- an exhaust leak;
- a crack in the DPF that diverts the exhaust away from the sensor
- a layer of carbon deposits on the sensor that insulates it
(c) this is all wrong. It's not slower; the upstream temperature rises faster because of that large early injection spike. The downstream temperature rises at a normal-looking speed because of the thermal mass of the DPF in-between.
Last edited:
Dancingbear
Member
Right then,
I am not experienced in either mechanics or AI, but ti thought i'd consult chat gpt for some interest. after a bit of 'to and fro' this is what it generated. To be honest i'm a little overwhelmed at A) how clever it is, and B) how well informed you guys have been to reach pretty much the same conclusions. AI has not really identified / assessed or addressed the unstable temp data discussed here by you guys (yet) as this is obviously a comparative excercise... but otherwise it appears that you / it are agreeing entirely.
Interesting what it says about advertised FIAT regen temps - thoughts???
It doesn't change anything and irrespective of its output my direction remains the same :
1) collect more regen info;
2) borescope the downhill end of the dpf;
3) consider further injector diagnostics.
Happy to be told this is all complete bobbins, but based on my limited knowledge of what you've kindly given me, its condensed gold!!
............................................
Files interrogated:
- DPF regen 1a.csv.xlsx (failed regen)
- DPF regen 2a success.xlsx (successful regen)
- Template 5 30 secs at idle with a few mins driving.csv (idle baseline)
Regen 1 (FAILED)
- Post-injection on-time: 47.64 s (from t=1685.17 to 1732.81 s).
- Vehicle speed during active phase: 104.6–108.8 km/h (avg ~106.7).
- Engine speed: ~2115 rpm avg.
- DPF core temp (mean / max): 491°C / 740°C across those 47.6 s.
- Post-DPF temp (mean / max): 388°C / 404°C.
- DPF ΔP (mean): ~46 mbar during the active window.
- Soot load change during active window: 100.0% → 99.0% (-1.02%).
Regen 2 (SUCCESSFUL) – three contiguous “post-injection” segments
- Segment A: 101.8 s (t≈945.65–1047.42)
speed 48–66.8 km/h (avg 59.9), rpm ~1994, DPF 573°C (mean), max 733°C, post-DPF 332°C (mean), ΔP 25.7 mbar, soot 100.0%→95.5% (-4.6%) - Segment B: 204.7 s (t≈1063.55–1268.21)
speed 17.7–72.9 (avg 47.4), rpm ~1631, DPF 634°C (mean), max 740°C, post-DPF 579°C (mean), ΔP 13.7 mbar, soot 91.1%→35.8% (-55.3%) - Segment C: 69.5 s (t≈1274.42–1343.96)
speed 57.1–69.3 (avg 63.1), rpm ~2134, DPF 671°C (mean), max 733°C, post-DPF 611°C (mean), ΔP 30.6 mbar, soot 34.3%→10.5% (-23.9%)
Total soot reduction across Regen 2: 100.0% → 10.5% (~-89.6%).
Temperature stability: std-dev of DPF core temp falls across segments (≈160°C → 49°C → 34°C), i.e., it stabilises as soot burns down (not “running away”).
Idle baseline (your “Template 5” file, ~800 rpm, 0–1 km/h)
- DPF ΔP at hot idle: mean ~5.3 mbar (min 3, 75th percentile 6, single high outlier 17).
- Injector corrections @ warm idle (mm³/stroke):
- Cyl 1: -0.18, Cyl 2: +0.88, Cyl 3: -0.07, Cyl 4: -0.61 (sum ~0).
These are modest trims and (based on public MES/Ducato discussion) well within typical acceptance (often quoted ≈ ±2 mm³/str; see caveat below).
- Cyl 1: -0.18, Cyl 2: +0.88, Cyl 3: -0.07, Cyl 4: -0.61 (sum ~0).
Were those regen numbers “normal” for a Ducato 2.3 twin-EGR?
- Regen temperature band: Fiat eLearn documentation for Fiat DPFs states typical DPF temperature during regeneration ~700–800 °C; risk of thermal shock >1000 °C. Your maxes of 733–740 °C in both regens are right in the normal band, with no evidence of >800–900°C excursions. aftersales.fiat.com
- DPF ΔP at idle: Community data points for Ducato/MES show ~5–10 mbar at hot idle and ~14 mbar at 2000 rpm on a healthy filter. Your idle ~5.3 mbar fits that picture. (These are user-reported values, not official test specs.) motorhomefun.co.uk
- Injector correction tolerance: I could not find an official Fiat Professional spec sheet with hard limits in mm³/str for this ECU. However, MES/owner discussions commonly quote rough guidance ~±2 mm³/str as “generally OK” (beyond that begins to raise eyebrows). Your trims are within that envelope. Treat this as indicative, not factory-authorised. Multiecuscan+2Facebook+2
- Twin-EGR architecture (no AdBlue): Fiat’s Euro 6 (pre-AdBlue) Ducato used HP-EGR + LP-EGR; widespread cooler/flow issues are documented by owners/techs, and post-2019 Fiat moved to AdBlue. This is architecture context, not a tolerance. The FIAT Forum
Conclusion on Regen 2: By numbers, Regen 2 looks technically normal: correct temperature band, substantial soot mass reduction, and no abnormal ΔP or injector behaviour during the burn.
Can soot reach the tailpipe if the DPF is intact?
Short answer: In normal operation, no. The whole point of the wall-flow DPF is to physically trap particulate. Visible black soot depositing on the tailpipe strongly implies soot is getting downstream of the filter, which practically requires a bypass path (e.g., substrate crack, melted passage, failed/detached internal seal) or a downstream component producing soot (not applicable here).
- A truly intact wall-flow DPF will not coat the tailpipe black. Fine particle number can never be literally zero, but visible soot deposit is not normal for an intact filter.
- Therefore, tailpipe soot is strong evidence the DPF is compromised (cracked, melted channeling, seal failure) or there is an external bypass (e.g., flex-joint or flange leak that injects upstream soot after the DPF into the tailpipe—rare on this layout).
(Owner/tech guidance on Ducato specifically also uses tailpipe wipe as a field indicator of a cracked DPF—again, community guidance rather than official spec.) The FIAT Forum
So why is your DPF likely compromised? (Most probable root-cause pathways)
Below are hypotheses ranked by plausibility, referencing your data and Fiat-specific context; I’ll flag what is fact vs inference and how to confirm.
1) Thermal/mechanical damage from prior off-map running (turbo unplugged) → excessive soot → over-hot/uneven regeneration → substrate crack
- Facts from your case: Turbo actuator was found electrically unplugged for an unknown period. You observed tailpipe soot for ~6 months.
- Mechanism (inference): With VGT control offline, air/fuel ratio control deteriorates, soot output spikes, regens become hotter and more frequent. Repeated, aggressive, or unevenly heated regens are a known pathway to substrate cracking or melting (Fiat eLearn warns DPF can be damaged >1000 °C, and thermal shock is a risk). Even if your recent regens stayed at ~740 °C, historical events may have been harsher. aftersales.fiat.com
- How to confirm now:
- Borescope the DPF from the outlet (post-DPF) to look for radial cracks / channeling.
- Compare ΔP vs flow at fixed rpm/MAF to a known-good Ducato (or published reference). A too-low ΔP for flow can indicate bypassed brick (your post-clean numbers were very low per the mobile specialist).
- Thermography during a regen: watch for asymmetric heating across the brick (advanced, but revealing).
2) Historical sensor/actuator faults → “unstable temperature profile” → localised thermal stress
- Facts: You had HP EGR replaced, LP EGR replaced, DPF temp sensor replaced, throttle valve failure (limp), and P2002 after a failed regen.
- Mechanism (inference): A drifting temp sensor or throttle/EGR control error can cause the ECU to over- or under-fuel the exotherm, oscillating temps. Forum feedback to you about post-DPF temps wavering is consistent with control loop “hunting.” Over time, these oscillations can thermally fatigue the brick.
- Your current data: In Regen 2 the temp variation actually settles nicely (std-dev falls segment-to-segment), i.e., present behaviour looks stable. That doesn’t rule out past instability.
- How to confirm now:
- Correlate DPF inlet vs post-DPF temps during a fresh regen: you’re looking for smooth rise and less oscillation post-DPF, not spiky seesaw.
- Validate the DPF temp sensor against a calibrated probe (if feasible) to exclude drift.
3) Chronic over-fuelling from injectors (less likely on today’s evidence)
- Facts now: Idle injector corrections are modest and balanced (-0.18, +0.88, -0.07, -0.61 mm³/str; sum ~0). That does not scream over-fuelling.
- Caveats: Idle trims aren’t the whole story; under-load quantity deviations or nozzle dribble can still soot a DPF.
- How to probe: Conduct a correction map under steady load (e.g., 2000–2500 rpm, 30–50% load) and/or return flow tests. If any cylinder shows persistent positive correction under load, that nudges this up the list. (I don’t yet have your load-correction data.)
- Bottom line today: Based on your idle file only, injectors do not look like the prime suspect.
4) Pressure-sense plumbing error (past) producing misleading ΔP (edge case)
- Mechanism: Reversed/blocked ΔP hoses can mislead the ECU, but that typically causes regeneration mis-timing and P2002 rather than months of visible tailpipe soot unless the brick is already compromised.
- Your current ΔP behaviour at idle and during Regen 2 looks coherent, making this unlikely now.
Why Regen 1 failed but Regen 2 succeeded (without guessing about speeds)
- Your figures are right: Regen 1’s active burn lasted ~48 s at ~106–109 km/h and removed ~1% soot. Regen 2 removed ~90% in ~376 s across three segments at 17–73 km/h.
- Fact-based interpretation: Regen 1 was too short to matter; duration, not speed, was the difference. Your success at lower speeds in Regen 2 is consistent with the ECU being able to maintain the exotherm and finish the job once it committed to a full, multi-segment burn.
Answers you asked for, plainly
- “Is there ANY way soot reaches the tailpipe if the DPF remains intact?”
Practically: no. Visible soot deposit indicates bypass of the filter media (crack/melt/failed seal) or a very unusual downstream soot source (not applicable here). Occasional nanoparticle slip is not the same as black tailpipe soot. - “Was the successful regen within normal parameters?”
Yes, by temperature and outcome. Your ~570–670°C mean, ~733–740°C max core temps and the ~90% soot reduction are compatible with Fiat’s stated ~700–800°C regen regime (and well below damage thresholds). ΔP and injector trims during that regen do not look abnormal. aftersales.fiat.com+1 - “What most likely caused the DPF to become compromised?”
Based on your history and the data, the leading candidate is historical thermal/mechanical damage from running with the turbo unplugged and earlier EGR/throttle faults → excess soot → aggressive/uneven regens → substrate cracking/channeling. Present-day data doesn’t indict injectors or control instability, but your history leaves a clear high-risk window.
What I recommend next (to avoid replacing a DPF only to crack it again)
Confirm the DPF’s physical integrity now
- Borescope the brick from the outlet side. You’re looking for radial cracks, melted “holes,” or missing washcoat. A shop familiar with Ducato can do this in situ.
- Flow/ΔP plausibility check: At a fixed warm idle and at 2000 rpm no-load, compare your mbar to a known-good Ducato of the same engine (or use published owner references as a sanity check). Your idle ~5 mbar is good; if high-flow ΔP is unexpectedly low for airflow, that supports bypass.
Rule out upstream causes that could re-damage a new DPF
- Under-load injector balance: log injector corrections under steady load, not just idle; add rail pressure actual vs target, MAF, MAP. If all look tight under load, injectors move further down the suspect list. (I can template a MES logging set if you’d like.)
- Temperature sensor validation: Use the same drive to log DPF inlet, core, and post-DPF temps; look for smooth, monotonic heating during an active regen, not oscillation. If you can, cross-check the DPF temp sensor with an external probe to rule out drift.
- Exhaust leaks survey: Although leaks after the DPF won’t create tailpipe soot (the soot would still be upstream), leaks before the DPF can distort ΔP and heating. Quick smoke test is cheap insurance.
If the DPF is cracked (as is likely)
- Replacing it without addressing any remaining control faults is risky. But your current logs don’t show ongoing control problems. So replacement after the above confirmations is reasonable.
Fiat-specific references used (and limits)
- DPF regeneration temps & damage risk (Fiat eLearn / aftersales portal excerpt: “average DPF temp during regeneration 700–800 °C; damage above ~1000 °C/thermal shock”). aftersales.fiat.com
- Twin-EGR (Euro 6, non-AdBlue) architecture context and common failure prevalence (FiatForum threads summarising the platform). The FIAT Forum+1
- Typical MES ΔP readings from Ducato owners (idle ~8 mbar; 2000 rpm ~14 mbar as a community datapoint, not a spec). Your idle numbers match this order of magnitude. motorhomefun.co.uk
Want me to set up a tight MES logging template?
I can give you a single-run channel list (DPF temps in/out, ΔP, post-injection, MAF/MAP, rail pressure target/actual, HP/LP EGR cmd/pos, throttle, injector corrections) with sampling at 5–10 Hz and brief instructions to capture (a) warm idle baseline, (b) 2000 rpm no-load, (c) one complete active regen. That will let us definitively confirm sensor stability and under-load injector behaviour.
If you prefer, I can also produce quick plots from any new CSVs you upload and benchmark them against the numbers above.
lycopersicum
Member
A dP of 30 mbar and less at the start of the regen is suspiciously low IMHO. On my Euro 5 I get something closer to 100 mbar. This is indicative of a cracked DPF.
That could explain the abnormally slow heating of the DPF, with the hot exhaust just passing through, venting most of the heat down the pipe.
I do not trust the AI to grasp fine context like a certain dP being normal at idle but not at speed. I've seen it make mistakes of that sort quite often.
That could explain the abnormally slow heating of the DPF, with the hot exhaust just passing through, venting most of the heat down the pipe.
I do not trust the AI to grasp fine context like a certain dP being normal at idle but not at speed. I've seen it make mistakes of that sort quite often.
Dancingbear
Member
100% agree with those kinds of nuances, but it has also raised some potentially very useful data such as the 'normal' regen temp - which is outside of the temperature ranges of all who've reported their own observations here - of course i have not fact checked that yet... watch this space...
Last edited:
I think one needs to be very careful with AI assistants. For example, the references quoted from your ChatGPT post include this forum, as well as FB and other forums – so it is using references that may possibly be just opinion, hearsay or misquotes etc and could possibly lead to self-confirmation. So, next time it searches for answers, it may well come across your post of its response and self-consume to produce further corruption. Who knows?
It does however reference some quotes from the manufacturer FIAT, but even that has exposed some real issues. For example, it has stated “typical DPF temperature during regeneration ~700–800 °C; risk of thermal shock >1000 °C”. And then states that your maximums are well within the normal operating band! Further, it concludes with “Conclusion on Regen 2: By numbers, Regen 2 looks technically normal: correct temperature band, …” What absolute BS!
I know exactly which copyrighted FIAT document that was scooped from as I have read it. But those figures were taken from a general preamble discussion about the DPF materials and material specifications during a comparison discussion of the then current FAP system used by other manufacturers. Within that same very old document and several later references can be found that the DPF temperature must reach a minimum of 580C for an active regeneration, with typical operating range of 600C – 650C. And those figures are confirmed by our practical measurements of good engines.
Another issue is that it has stated it could not find any official FIAT documentation that referenced the injector correction tolerances and could only state it found some unsubstantiated information from MES, 2 FB + 2 other unnamed references. Well, it couldn’t have looked too bloody hard is all I can say! (I hope you were using the free version and not paying for such advice). Perhaps next time, after reading and scooping this it might reference FCA Service News 10-023.19 of 20/11/2019 and a much earlier FCA Service News 10-17.09 of 14/04/2019. I prefer to use my own intelligence.
But, after all that, it has basically agreed with my diagnosis – I suspect that the DPF is damaged and causing the issue which is why I suggested the first thing you do is take a borescope and investigate the downstream side of the DPF and also check the Post DPF temp sensor.
It has even suggested a parameter list for testing – which you have already done (my suggested template) to rule out any other possibilities. Be aware that DPF damage may not be obvious downstream as the filter medium may be damaged internally, or even on the upstream surface area. The upstream Catalytic Converter may also be damaged.
The other issue you need to keep an eye on is the new throttle body for correct operation as previously stated as it was completely open for the entire recent test run.
Just to confirm something. I have recently read a trusted document that validates my earlier reverse engineering musings, and that is the DPF Temp sensor is the “Policeman” and is the sensor responsible for triggering MAX TEMP cut-off to terminate the regen if the maximum allowable temperature is exceeded; and the Post DPF Temp sensor is used for the desired regulated regen temperature (in coordination with the DPF temp sensor). This adds to the weight of what is causing the regen termination – as previously shown by the diagnosis there is something malfunctioning to cause the Post DPF temp sensor from reporting the correct Post DPF temperature fast enough – it is either under-reading (sensor malfunction) or the bulk of the exhaust gas flow is bypassing it due to some form of internal damage to the DPF. I suspect the sensor is probably OK, as it eventually reports near correctly after some time lag, and it also follows closely the other temp sensors as expected prior to the regen occurring. It is the rate of the Post DPF temperature sensing that is the issue, and that can be explained by exhaust gas bypassing the sensor due to internal damage to the DPF.
It does however reference some quotes from the manufacturer FIAT, but even that has exposed some real issues. For example, it has stated “typical DPF temperature during regeneration ~700–800 °C; risk of thermal shock >1000 °C”. And then states that your maximums are well within the normal operating band! Further, it concludes with “Conclusion on Regen 2: By numbers, Regen 2 looks technically normal: correct temperature band, …” What absolute BS!
I know exactly which copyrighted FIAT document that was scooped from as I have read it. But those figures were taken from a general preamble discussion about the DPF materials and material specifications during a comparison discussion of the then current FAP system used by other manufacturers. Within that same very old document and several later references can be found that the DPF temperature must reach a minimum of 580C for an active regeneration, with typical operating range of 600C – 650C. And those figures are confirmed by our practical measurements of good engines.
Another issue is that it has stated it could not find any official FIAT documentation that referenced the injector correction tolerances and could only state it found some unsubstantiated information from MES, 2 FB + 2 other unnamed references. Well, it couldn’t have looked too bloody hard is all I can say! (I hope you were using the free version and not paying for such advice). Perhaps next time, after reading and scooping this it might reference FCA Service News 10-023.19 of 20/11/2019 and a much earlier FCA Service News 10-17.09 of 14/04/2019. I prefer to use my own intelligence.
But, after all that, it has basically agreed with my diagnosis – I suspect that the DPF is damaged and causing the issue which is why I suggested the first thing you do is take a borescope and investigate the downstream side of the DPF and also check the Post DPF temp sensor.
It has even suggested a parameter list for testing – which you have already done (my suggested template) to rule out any other possibilities. Be aware that DPF damage may not be obvious downstream as the filter medium may be damaged internally, or even on the upstream surface area. The upstream Catalytic Converter may also be damaged.
The other issue you need to keep an eye on is the new throttle body for correct operation as previously stated as it was completely open for the entire recent test run.
Just to confirm something. I have recently read a trusted document that validates my earlier reverse engineering musings, and that is the DPF Temp sensor is the “Policeman” and is the sensor responsible for triggering MAX TEMP cut-off to terminate the regen if the maximum allowable temperature is exceeded; and the Post DPF Temp sensor is used for the desired regulated regen temperature (in coordination with the DPF temp sensor). This adds to the weight of what is causing the regen termination – as previously shown by the diagnosis there is something malfunctioning to cause the Post DPF temp sensor from reporting the correct Post DPF temperature fast enough – it is either under-reading (sensor malfunction) or the bulk of the exhaust gas flow is bypassing it due to some form of internal damage to the DPF. I suspect the sensor is probably OK, as it eventually reports near correctly after some time lag, and it also follows closely the other temp sensors as expected prior to the regen occurring. It is the rate of the Post DPF temperature sensing that is the issue, and that can be explained by exhaust gas bypassing the sensor due to internal damage to the DPF.
Last edited:
I am suggesting option (b) is the most likely one, not just because I have suggested it as well, but if you compare the response of both the Temp sensors to each other and in particular with reference to the amount and duration of the Post Injection amounts it is clear that the upstream temp sensor (DPF Temp) does not rise super rapidly in comparison to the downstream sensor (Post DPF Temp) - specially telling is the response at the 1102S mark where there is a long duration and amplitude pulse of Post Injection.
