Technical Ducato P0401 P0402 P0236 P0238 and limp mode its 2017 Euro 6, EGR changed, What do I check next

[Fredastaire]

@Dexron there's not much point in opening up the LP EGR to remove just the visible mesh. I've just cut the infeed pipe of my removed 'nearly blocked' cooler to look at using the pipe. Then I looked inside the canister. There is a complex interior with lots of tightly packed fins. All full of soot.
.
You need to start with either a brand new cooler or a chemically cleaned one.
.
Then look at undoing the Flexi to redirect to fresh air and an air filter.
.
Don't forget you need to also check and clean the intercooler (I've done this), then perhaps remove and clean the inlet manifold, (I haven't done this and would welcome the 'how to' so I can do my own).

 

[phaphanou]

concernant
j'avais retiré les 2 boulons par le bas et boucher le fap avec une plaque de tôle ronde épaisse plus le collier
j'avais aussi penser aune entretoise mais la sonde de température est avant le haut de la jonction et je ne voulais pas qu'elle soit perturbée
j'avais également pensé faire une entretoise avec un tiroir et un tamis fin interchangeable dans ce tiroir ,j'ai abandonner car je ne voulais plus être emmerder avec ce système
bonne journée
J'ai retiré les deux boulons du bas et j'ai bouché le FAP avec une épaisse tôle ronde et le collier de serrage. J'ai aussi envisagé une entretoise, mais la sonde de température se trouve avant le haut de la jonction et je ne voulais pas la dérégler. J'ai aussi pensé à fabriquer une entretoise avec un tiroir et une crépine fine interchangeable dans ce tiroir. J'ai abandonné, car je ne voulais plus m'encombrer de ce système. Bonne journée.
View attachment 466057View attachment 466056

@Cariou ceux que tu as nettoyé avaient beaucoup de km ? Il sont plus vers les années 2017 ou plus récentes ?
C'est intéressant de voir que tu as réussi a nettoyer mais tu as été obligé de tout démonté alors ?

 

[deejays]

@Cariou ceux que tu as nettoyé avaient beaucoup de km ? Il sont plus vers les années 2017 ou plus récentes ?
C'est intéressant de voir que tu as réussi a nettoyer mais tu as été obligé de tout démonté alors ?

@Cariou, did the ones you cleaned have a lot of mileage? Are they more from 2017 or newer?
It's interesting to see that you managed to clean it, but you had to take everything apart then?

 

[theoneandonly]

@jansla
here is the missing reply
Have you now got a P0401 error?
Have you got a clean or sooty exhaust?


These values are a bit on high side

The most relevent thing about these results is we realy need more recent ones for a longer period these are only just warm.
The obvious statement is there is a vast turbo differential

 

[Cariou]

HELLO, the vehicles were built in April 2017 and have around 30,000 to 45,000 km.Yes, you must disassemble from underneath and clean the entire intake and cooler under the radiator.Otherwise, without disassembly, use a heat gun through the lower inlet.Here is the complete intake with the throttle valve, EGR valve, and pressure sensor. Everything should be clean.

 

[deejays]

HELLO, the vehicles were built in April 2017 and have around 30,000 to 45,000 km.Yes, you must disassemble from underneath and clean the entire intake and cooler under the radiator.Otherwise, without disassembly, use a heat gun through the lower inlet.Here is the complete intake with the throttle valve, EGR valve, and pressure sensor. Everything should be clean.
View attachment 466511

Thank you @Cariou , If I want to clean the LPEGR cooler with a hot air gun whilst in vehicle, how do I gain access to undo the two screws that hold the cooler to the 3-way LPEGR valve?


Merci @Cariou, si je veux nettoyer le refroidisseur LPEGR avec un pistolet à air chaud pendant que je suis dans le véhicule, comment puis-je accéder pour dévisser les deux vis qui maintiennent le refroidisseur à la vanne LPEGR à 3 voies ?

 

[jansla]

@theoneandonly Last year, november, Fiat-doubtfully-"Professional" changed MAF sensor on my van (previously MAP, solenoid, software update) explaining that it is for sure where the problem was. But it was NOT. P0236 and 238 came back under exactly same circumstances the first day I started the van after the winter. In my case there have been and still are only two codes P0236 and 238 but not the 401 thus not sure if it is really the same problem as discussed in this thread. Last summer I've done 6k km without any limp modes until autumn when all was back again. Recently I was driving with Multiecuscan connected and recorded some parameters when the fault has happened. 2 times.

Would you like to look at it? On the 1st of May It happened at row 553, odometer 65602.9km, and on the 8th of may, second file row 307. There is still significant difference between desired and actual boost at times. There were some rides (the same distance same route), when there was no more than 250-300mbar in difference during the ride, but always higher, up to 400 when fault has occurred. May it be an intermittent boost leakage? How the leak can be intermittent?

My differential pressure sensor is now showing much better values (nothing was done to that besides software update) exhaust pipe is more or less the same sooty as it was all the time and I'm more or less same sad not being able to solve the problem

 

[deejays]

@theoneandonly Last year, november, Fiat-doubtfully-"Professional" changed MAF sensor on my van (previously MAP, solenoid, software update) explaining that it is for sure where the problem was. But it was NOT. P0236 and 238 came back under exactly same circumstances the first day I started the van after the winter. In my case there have been and still are only two codes P0236 and 238 but not the 401 thus not sure if it is really the same problem as discussed in this thread. Last summer I've done 6k km without any limp modes until autumn when all was back again. Recently I was driving with Multiecuscan connected and recorded some parameters when the fault has happened. 2 times.

Would you like to look at it? On the 1st of May It happened at row 553, odometer 65602.9km, and on the 8th of may, second file row 307. There is still significant difference between desired and actual boost at times. There were some rides (the same distance same route), when there was no more than 250-300mbar in difference during the ride, but always higher, up to 400 when fault has occurred. May it be an intermittent boost leakage? How the leak can be intermittent?

My differential pressure sensor is now showing much better values (nothing was dane to that besides software update) exhaust pipe is more or less the same sooty as it was all the time and I'm more or less same sad not being able to solve the problem

Hi @jansla I have downloaded your files but I cannot open them in MES - it responds with a file format error, so you must have changed something that upsets MES. I can open them and view as a CSV file, but it would be much better to view them as a graph within MES. Have you a copy of the file that has not been changed? Or perhaps another file that has not been changed - as a check?

 

[jansla]

before posting I've tried to open in Numbers and it worked ok. Excel should do as well as Google sheets. A .csv format is interchangeable between most spreadsheet programmes which MES is not. I'm not using MES for graphing nor for analysis as spreadsheets give much wider possibilities. Yes they are modified as I added a column with calculated difference between desired and actual boost, to be able to see the differences easily as well as the distance from start, but no problem I can attach originals from MES if you prefer. Here you are
the file from 1st of May time stamp is 643.95s, and in the file from 8th of May the time stamp is 410.24sec when the fault and limp mode happened

 

[deejays]

before posting I've tried to open in Numbers and it worked ok. Excel should do as well as Google sheets. A .csv format is interchangeable between most spreadsheet programmes which MES is not. I'm not using MES for graphing nor for analysis as spreadsheets give much wider possibilities. Yes they are modified as I added a column with calculated difference between desired and actual boost, to be able to see the differences easily as well as the distance from start, but no problem I can attach originals from MES if you prefer. Here you are
the file from 1st of May time stamp is 643.95s, and in the file from 8th of May the time stamp is 410.24sec when the fault and limp mode happened

Thanks @jansla I can view them now in MES. I will leave the main examination and report with @theoneandonly as expected. Cheers. If I find anything that may be important I will respond.

 

[deejays]

before posting I've tried to open in Numbers and it worked ok. Excel should do as well as Google sheets. A .csv format is interchangeable between most spreadsheet programmes which MES is not. I'm not using MES for graphing nor for analysis as spreadsheets give much wider possibilities. Yes they are modified as I added a column with calculated difference between desired and actual boost, to be able to see the differences easily as well as the distance from start, but no problem I can attach originals from MES if you prefer. Here you are
the file from 1st of May time stamp is 643.95s, and in the file from 8th of May the time stamp is 410.24sec when the fault and limp mode happened

@jansla I have only taken a quick look at the data from 01/05/25 using MES graphs, and I would say you have a blocked or nearly fully blocked LPEGR. cooler

 

[jansla]

@deejays sounds not so great.... which values are telling that?

 

[deejays]

sounds not so great.... which values are telling that?

Well, I would rather leave the diagnosis up to @theoneandonly as that is what has been set in train, and I do not want to but in where I may not be wanted, but I have now also looked at the second file and it indicates a similar probability. The LPEGR temp is really low, even though it is being supplied with high temp gas as witnessed by the particle filter temperature. To back that up, the LPEGR valve at no stage modulates, but oscillates between fully open and closed and spends a lot of time fully open - which should raise the LPEGR temp. Corresponding with this is a low manifold temperature. As pointed out by @Fredastaire this is a hallmark of the blocked cooler symptom and results in the air supply being cut off. What is interesting and maybe points to something else occurring is the HPEGR valve appears to be behaving itself and not going fully open in sympathy - so there may be something else going on, either by itself or in tandem. To get a better idea I think you need to check with a set of parameters that include for one, desired air intake quantity and a few others as well. I am still learning myself and there are a few gaps in my knowledge. For example, I understand what Low Pressure EGR Position refers to, but what does Low Pressure EGR Valve Opening refer to? I thought they were likely one and the same - one expressed in actual mm of opening, the other expressed as a percentage of maximum opening, but looking at the actual measurements that does not make sense. Both EGR valves are tracking where the ECU would want them - so they look OK. The throttle valve looks OK. The DPF pressure looks good. As I said - my guess is a blocked or nearly blocked LPEGR cooler.

 

[jansla]

Thank you @deejays I do agree it is not easy to understand all MES readings and we are still learning... and as I am not able to change LPEGR myself, quite sadly and possibly I need to continue feeding "Fiat-not-so-Proffesional" with my money... if it only could give some result

 

[deejays]

Thank you @deejays I do agree it is not easy to understand all MES readings and we are still learning... and as I am not able to change LPEGR myself, quite sadly and possibly I need to continue feeding "Fiat-not-so-Proffesional" with my money... if it only could give some result.

Thank you @deejays I do agree it is not easy to understand all MES readings and we are still learning... and as I am not able to change LPEGR myself, quite sadly and possibly I need to continue feeding "Fiat-not-so-Proffesional" with my money... if it only could give some result

Best to wait and see the outcome from @theoneandonly I reckon. I would not be able to change the LPEGR myself either - which is one reason (like you) I want to learn and understand what we are dealing with here. But you are quite correct - quite often the "experts" have no idea and are quite happy spending your money to feed the parts canon. Which is why we need to be able to at least try and diagnose what is wrong.

 

[theoneandonly]

@theoneandonly Last year, november, Fiat-doubtfully-"Professional" changed MAF sensor on my van (previously MAP, solenoid, software update) explaining that it is for sure where the problem was. But it was NOT. P0236 and 238 came back under exactly same circumstances the first day I started the van after the winter. In my case there have been and still are only two codes P0236 and 238 but not the 401 thus not sure if it is really the same problem as discussed in this thread. Last summer I've done 6k km without any limp modes until autumn when all was back again. Recently I was driving with Multiecuscan connected and recorded some parameters when the fault has happened. 2 times.

Would you like to look at it? On the 1st of May It happened at row 553, odometer 65602.9km, and on the 8th of may, second file row 307. There is still significant difference between desired and actual boost at times. There were some rides (the same distance same route), when there was no more than 250-300mbar in difference during the ride, but always higher, up to 400 when fault has occurred. May it be an intermittent boost leakage? How the leak can be intermittent?

My differential pressure sensor is now showing much better values (nothing was done to that besides software update) exhaust pipe is more or less the same sooty as it was all the time and I'm more or less same sad not being able to solve the problem

I'm sorry but these are simular to previous runs in that you only just get upto temperature. minimum of a 20 min run 10 min to get upto temperature. Preferably at least 10mins (at temp) at 60mph. It is not prudent to make a diagnosis wth this data. When i make a diagnosis i want to be certain (as i can be) thus the need for more data. I agree with @deejays that it is probably a blocked cooler BUT prefer to see more data, the behaviour when warm is conclusive this is not. Recent experiance has shown it does not need a P0401 to have a blocked cooler but recently I have seen another with the same large discrepancy between actual and desired boost with even more time with O2 sensor less than 10% and this one definitely has not got a blocked cooler (not resolved that problem yet)

@deejays I hope i have written the guides with as much information as is needed for almost anyone to diagnose these problems. I am not precious about my knowledge its for all to share, so be prepared to have a go and I will not slap your hands if you dont get it perfect first time out.
I did have a real daft thought in the bath earlier, Should I write to the Ceo of Fiat and describe my methods and ask how that complies with Fiats knowledge and how we could best deseminate it to the Fiat professional organisation. At least I wasn't singing in the bath.

 

[Fredastaire]

For newer members, I'm the guy who started this now 85 page thread. Early on I had my cooler changed , subsequently and infrequently I've sent CSV files to @theoneandonly .late last year his analysis spotted the discrepancy for the turbo boost requested versus actual, I had no codes whatsoever. Then I 'knew' performance wasn't quite there so as my Ducato was going to the garage for MOT (needing up on his H ramp, I decided to have the cooler replaced again, at this point MES showed a stored 0401 yet it hadn't occurred on screen or brought the MIL light on.
.
So with another cooler change done, I've no idea how long before it took blocks with soot.
.
Watching the current posts I think persons need to be aware that if your exhaust shows any soot then the cooler WILL block shortly.
.
To get out of the problem you need to buy a new DPF AND either a new cooler or get the original chemically cleaned out.
.
My alternative choice do far has been to replace the cooler knowing it will only last a while, then refit the original that I've caustic cleaned.
.
On the question of dealers in ability to sort, their engineer/ mechanic won't have the knowledge as maybe they've never seen the twin EGR version before.
.
The overall problem is we don't what failed first and Fiat will never say. I suspect that the 2017 DPF manufacturing year probably had a low quality level which was easily damaged by software program instructions leading to porosity of the element and soot passing through.
.
I guess there will be even more victims joining us.......

 

[deejays]

Hi @jansla , I have been giving some thought to the issue with your vehicle and your empty pockets. Considering that MES can log a maximum of 26 parameters, I have put together a list of what I consider to be the most critical 26 for you to examine. This morning, I took my 2017, 150HP, 2.3L, Comfortmatic Ducato for a drive whilst collecting the data for those 26 parameters so you could have a known reference for what a good engine with no issues or faults, and a very clean exhaust pipe looks like. I have owned the vehicle from new, service it myself, and has no parts replaced at all. It has travelled 36,000km.

I suggest you load the same parameters and start with a cold motor and go for as long a drive as possible to obtain some meaningful data. I deliberately kept the speed, and throttle demands low for a very significant part of the journey to try and match what I have seen in your past records. You will note that even very early on, without any significant higher speeds that the LPEGR cooler rises to much higher temperatures than what was achieved with your vehicle in its current state.

To analyse the data, I only use MES graphs – I see no need whatsoever to export the data and view/manipulate in spreadsheets such as excel etc. (As a side note, within MES graphs you can select any parameter to place on the X-axis to directly plot the relationship for a related parameter on the Y-axis).

My approach is to start with comparing “desired” and “actual” for each parameter individually and selecting the x-axis scale to view the entire drive in one shot – if there appears to be any non-tracking, then zoom in to see the detail – and then if it is a significant non-tracking start overlaying other parameters one at a time to try and find a causal relationship. The reasonable tracking of “actual” to “desired” proves that at least the device is accurately following the ECU instructions – so that we can at least assume the device is behaving correctly.

Then I move through selective parameters overlaying what I consider having likely relationships to each other and look for anything that may be out of place.

To try and keep the number of graphs low in this post, I have not included all the graphs I look at, just some that may be important for you.

For your reference, I provide the MES file below so you can look at the data in detail. It may well be a good reference for anyone else with similar issues.

I will also post some of the pertinent graphs that I select to examine.

I would be happy to have a detailed look at your MES file of these 26 parameters once you upload it. I hope the motor will allow you to do so.

 

[jansla]

@deejays Thank you very much for your effort and time. You have no idea how long I have dreamed of having real life MES records from a good engine for further comparisons. Your examples confirm that MES is really good for analysing these records if done as you suggest and one don't have to transfer anything to a spreadsheet. Naturally, I will make the registration according to your proposed set and post it, I am away from home now, coming back on Tuesday, work Wednesday-Friday, so realistically hopefully on Friday after work, so it will take a while but it will be done Once again great thank you!

 

[theoneandonly]

@deejays @jansla at last an alternative gold sample. I have published simular in guides and all over this thread as engine A. If you could add the csv files i would appreciate it. I use semi auto graphing for comparisons and using real numbers the means and other statistics can be applied more accurately than trying to read a max value from a screen. A typical example is reading the lpegr cooler mean value but only if engine has exceeded 80C. The additional advantage of not using autoranging as MES ie a constant Y axis for a particular application is ease of comparison and less confussion for those not regularly reading graphs.

tHE number od pids able to be measured ie 22 to 28 varies on the PIDs and Dongle BT able to do less than USB. this is basicaly why Alfaobd resitricts you to 16 pids (quite restrictive). The Template I use has been tweeked over the years to hone in this specific problem and to see what systems are working and what not including regen. eg i only check the actual values of the egr valves because this indicates they are working, if outside normal parameters then you can drill down and look at more specific issues.