Mk4 Mondeo 2.0 TDCI Limp Mode

[Tdci-Peter]

3 hours ago, Dee_82 said:

May I ask what software your using to open those CVS files like this?

Its software "wot I writ", it does not run on Windows, and only on certain (older!) hardware, so is not very portable!

I wrote earlier this year as I wanted to compare the MAF readings with flow calculated from MAP & RPM on my car. And also I find the Forscan graph scrolling feature is very clumsy and clunky on my system. Especially on bigger runs.

A CSV file is just a text file. I have tested the Microsoft Notepad utility, and it will chop it up. It is very easy to shorten it, just a bit of deft mouse work. In fact any portion(s) can be cut out, as long as the header line stays in place at the beginning. Each line has a time value as well as the data values, so any section will read ok.

To delete big chunks of data, highlight a line by dragging the mouse cursor over it, then use the scroll bar to move through to the end or 2nd cut point, then press the shift key while dragging the mouse over the end line to be removed. All lines from the 1st highlight to the 2nd should turn blue (or whatever is your highlight colour), then press the delete key.

That way, though a bit fiddly, should result in a .CSV file small enough for Excell to process.

Looking at the sensor, it does not look like a load of gubbins will pop out of the actuator when removed, but access may be a problem as there is quite a lot of the sensor inside the actuator, so space is needed to withdraw it.

[Dee_82]

haha, love, it. Need  a tool? then make one 🙂

Im recording these to and from work since it can happen at any time so the files are huge, Excel and I have a long relationship so fiddling with it is fine its just clearly not kitted out for dealing with such long graphs.

I suppose if I can find an interesting time segment I can just chop around it but having a sensible scrolling function on forscan would be much better, id quite like them to be able to remove traces and even import CVS files but I doubt thats going to happen any time soon either.

I think space wise its ok, it sits behind the engine above the CAT and theres a fair bit of room to the firewall, what might be more of a challenge is getting them off, quite a few stories on the HDi forums that suggest replacement bolts will be required once I strip these ones getting them off. 

 

[pragmatix]

Ive been following this thread with great interest, facinating and educational, thanks guys.

[Dee_82]

So this was from today, 

No Limp Mode displayed.

Consistent loading, no apparent changes to MAP or EGR.

The second picture shows you the same (more or less) period of time with the negative values adjusted in excel.

Its pretty quick, perhaps too quick for it to make a difference to anything else, but if you look else where on the trace (third picture) you can clearly see other spikes that are matched with EGR changes, albeit these seem to tie in with APP increase / decrease. 

 

 

 

 

10_monty_test_-_Copy.txt

[Tdci-Peter]

2 hours ago, Dee_82 said:

Consistent loading, no apparent changes to MAP or EGR.

The second picture shows you the same (more or less) period of time with the negative values adjusted in excel.

It looks like mad movement of the vane, but as you see on the Excell, it is just sitting on the +/- changeover point, and when corrected, the signal is quite stable apart from those little ramps at each sign change. So the vane is just quivering a bit, not surprising, and the sensor signal reading is veering between about +150 to -150. So it is all looking quite consistent at the moment.

Have you considered back probing the sensor signal voltage when doing the vac. pump test? Probably easier said than done, due to limited access, and it would only serve to double check that it is a real fault, not some silly software glitch. Changing the sensor might be the easier option. It should then be quite easy to test the old sensor on the bench if any questions still remain.

[Dee_82]

Laptop now packed in, typical! Not my week at all.

So I am no further forward but I recorded something interesting on my phone today.

Came out of a roundabout in second, sped up, changed to third, (first big spike at the end) then mid acceleration I let go of the throttle and dropped it hard again. At which point she sped up again only this time I got some hesitation and a noticeable decrease in power. 

No warnings

If you look at the graphs you'll see the vanes going mental as normal but there's also a little drop in MAP which coincides with the hesitation.

Any thoughts? 

I'm going to swap out that sensor tomorrow, I'll work out the resistance across all the pins on the good one and see what the old one shows.

 

LiveData_2018-07-07_08.32.14.txt

[Tdci-Peter]

On 7/7/2018 at 8:34 AM, Dee_82 said:

If you look at the graphs you'll see the vanes going mental as normal but there's also a little drop in MAP which coincides with the hesitation.

Is the dip at time=242.0 sec?

I see no obvious reason for it, though the sensor is doing some odd things, so may make the actuator move. The RPM continues to rise at pretty much the same rate after, as far as I can see.

Looking at the sensor, it seems to have 3 pins, I would not be surprised if it was just a pot, which would have constant resistance between two pins, and a varying resistance from the third (pin 3 on the Focus 2L DW10 drawing) pin to either of the other two. It looks a bit as if the sticking out bit is a rod that moves axially rather then rotating, but it is hard to say from the photos.

If it was some fancy electronic (eg Hall effect) sensor, then resistance readings would not be very meaningful, though comparison might give some clue.

[Dee_82]

Hi Peter, yes, thats right, the RPM continues to rise but the MAP drops at the point it hesitates.  I ran some Revive turbo cleaner through the intake manifold and it smoothed out a lot of the hesitation but its still drops in to limp mode.

Im pretty sure I can cause it to limp.

6th gear, 65-70MPH, approx. 1200-1500 RPM 

Maintain that for 10 min or so and it may drop in to limp. If I pulse the accelerator and make sure that vane keeps moving, it doesn't go in to limp.

Ive got good vacuum on the actuator, good idle vacuum on the solenoid. 

Couldn't get good access to the sensor to get it off, the bolts were welded on, didn't want to destroy the head!  

Right now im pretty sure the solenoid is working fine, if the vanes actuator or sensor are screwed then replacing the turbo would fix it. That pretty much leaves wiring issues.

another problem has recently developed, unrelated. Getting tired of this!

 

 

[Dee_82]

Hi Guys, planned to get the turbo replaced on the 6th so fingers crossed, but from an academic point of view I thought is share some more data for you to pick over!

All that data is attached, ive screen shot the two areas of interest. 

So the first picture is a good example, this was a relatively short period of hard acceleration, you can see the Throttle open up, speed increased, MAP maxed out at 249 and held under acceleration. The vanes opened and closed, no noticeable hesitation, as mentioned before the vanes voltage isnt as clear as high and low and the negative values need to be adjusted to make sense of it, but it more or less looks ok

 

Picture Two was a few moments before, same test. hard acceleration, MAP increases as expected, vanes are all over the place and then you can see the MAF readings go a little mental, this is very noticeable in the car, feels like a drop and surge of power and ties in exactly where the vanes are going nuts. So it would appear that the vanes are actually moving and is having an effect on the MAP

 

 

 

Time stamp 1205

 

Time stamp 948

 

12 monty test.txt

[pragmatix]

Intersesting to see, ive learned a lot due to your misfortune, so its not all been in vain 😀 (sorry no pun ntended)

[Dee_82]

2 hours ago, pragmatix said:

Intersesting to see, ive learned a lot due to your misfortune, so its not all been in vain 😀 (sorry no pun ntended)

Haha, 

Every day is a learning day. just wish I "knew" what needs fixed. 

Here is a corrected version of the same two graphs, B for bad and G for good. 

You can clearly see now that pressure builds fine until it hits peak, but just before it does the vanes slam shut, that noticeable drops the pressure down, it then returns to a mid position briefly and gains a little pressure and drops it again, as it returns to a mid point, it then completely swings back to fully closed before snapping fully opened up, we build a little pressure then it snaps closed, dumps some pressure returns to a mid point, gains a little pressure, closes once more and losses it again before snapping fully open and returning to a normal semi open and maintains its pressure until I ease off the throttle.

The good graph as that same snapping going on mid where through but if you look at the number of the dots you can see it moved very quickly, perhaps so quick there wasn't enough of a change to make a difference, there was no limp mode in this example.

The question is, why is it doing it.

The vanes are controlled by a actuator arm which is controlled by a vacuum thats regulated by a boost solenoid, that vacuum is generated by a vacuum pump which feeds the brakes, EGR and throttle valve.

The actuator has a sensor that tells the PCM what position its at, the PCM tells the solenoid what amount of vacuum to pass on to the actuator arm, Boost is generated and reported to the PCM via the MAP (and MAF id assume) as that builds the PCM sees the boost level, then instructs the solenoid to make changes based on load.

we have a valid MAP reading so the PCM knows what is going on with boost, we also have a signal from the actuator which appears to be recording the movement of the vanes but we dont know if its error free.

The DTCs reported that the reduced boost performance was down to a stuck vanes / actuator, the position of which is reported by the actuator sensor.

So the PCM sees they aren't moving as it expects them to move and throws a warning that the turbo isnt a happy bunny dropping it in to limp mode which effectively disables the turbo.

The actuator holds a vacuum ok, so we aren't loosing vacuum, the solenoid could be nackard but why would it work sometimes and not others, only putting the car in to limp mode when we maintain low RPM at speed for a prolonged period of time. 

So my thoughts are, the PCM instructs the solenoid to give boost which it does, the actuator tries to continue to opening the vanes, on the G graph you can see it comes down, hesitates, then continues to drop, on the B graph it doesn't drop further, I reckon the PCM sees it stuck then in a bid to correct itself slams it closed and reopens it, didn't work, so it does it again, slams it closed, then reopens, only this time it does open and we run in to max boost, so closes it abit then cycles a couple more times to make sure its free then opens up to get where it wants boost to be before levelling out.

Im thinking the turbo is full of crap, but im happy to hear any other versions you might think make sense!! 🙂

 

 

[Tdci-Peter]

7 hours ago, Dee_82 said:

we have a valid MAP reading

Hmmm. I have looked at 12 Monty a bit, and there are some rather odd MAP readings.

At t=942, APP=21.5, MAP=82kPa, RPM>2k, vane voltage low and dropping (high boost position ?).

At t=973, APP=14.5, MAP=77, RPM>2k, vane voltage at min.

At t=919, APP=0, MAP=122, RPM>2k, vane voltage high (low boost position ?)

MAP readings much below 100kPa are unlikely, unless there is a blockage in the system (throttle valve, air filter, etc). I am wondering if the MAP baseline is moving about, and this is driving the odd vane movements, rather than the other way round. 

The top limit for MAP is staying at 249kPa, but this could be the control action of the vanes, limiting the reading. The actual pressure may be varying. Or, perhaps more likely, 250kPa is the calibration top voltage, it can not read higher regardless of actual pressure.

Also I had noticed that on occasions the vane voltage changes seemed to be lagging behind the MAP signal. Normally if a MAP change is caused by the vane movement, the vane should lead. I am not at all sure about this, there may be odd delays in the CAN bus data signals.

I hate to add confusion to chaos , but maybe it is the MAP, or its wiring, that is the underlying problem.

[Dee_82]

The MAP is limited to a top value, my 1.6 tdci was the same but I installed a boost gauge and it would routinely rise above its maximum sensor value when using overboost, I suspect this is the same.

Ill need to double check the position of the throttle valve, pretty sure the MAP is before the valve so if the valve was shutting then the pressure wouldn't drop below ambient. 

Its interesting you mention CAN errors though. 

MAP is one of the suggestions albeit the sub code does suggest vanes sticking in the problem.

but more interestingly is the ABS errors. 

===ABS DTC U0100:00-08===
Code: U0100 - Lost Communication With ECM/PCM A

===ABS DTC U0126:00-08===
Code: U0126 - Lost Communication With SASM

===ABS DTC U0401:68-8A===
Code: U0401 - Invalid Data Received From ECM/PCM

 

Specifically that second one, SASM is, as far as I am aware, part of the module in the steering column, the same steering column that has the cruise control module problems and in fact the left side also seems to forget which is volume and which is next track. 

 

DTC_2018-07-21_17.54.41.txt

[Tdci-Peter]

22 hours ago, Dee_82 said:

but more interestingly is the ABS errors. 

Now who's adding confusion to chaos!

The ABS errors are potentially serious, if they put the ABS warning light on that may be an MoT (or equivalent) fail, and if they are real and frequent, they may affect operation of the ABS. Though the essential safety features of this unit are quite self contained and reliable. CAN bus errors are not very likely to affect it seriously, exactly because CAN bus errors are frequent and common.

But the MAP, MAF, Turbo position sensor, Turbo solenoid, and the primary APP sensor are all local to, and directly connected to, the PCM. On the Mk2 Focus, the secondary APP sensor goes via the CAN bus from the IC, I am not sure if this is the same on the Mondeo, however no APP correlation errors are appearing.

ABS communication errors might cause a reduced power mode, but I would expect a warning light and/or message if that was happening. It does look as if the reduced power mode is due to the turbo / MAP relationship going haywire somewhere.

So I find it difficult to relate the ABS errors to the turbo problem. I may be wrong, won't be the first time (!), but it looks like two separate problems to me.

 

[Dee_82]

lol, I know, computer systems designed to make cars easier to identify problems just end up making things more complicated to diagnose. 

the TC warning light, turns on when it enters limp mode but not at any other time. Im actually wondering if those ABS errors might be related to the vibration problem ive had on bumpy roads! lol or maybe the cruise control problem which I just fixed today.

I fixed that CC problem today and all DTCs have now gone and remained gone.

Im pretty sure I can replicate the limp mode by being steady on the throttle and cruise at 65-70 in 6th, tomorrow im going to try use cruise control in 6th at 65-70 and see if it limps, 

If it limps on CC then I know its nothing to do with the accelerator sensors as they read 0.

The Throttle valve opens cleanly, registers fine. im going to monitor the EGR side, detach the ground points and clean them up and swap out the MAP for a new one.

[Tdci-Peter]

19 hours ago, Dee_82 said:

I fixed that CC problem today and all DTCs have now gone and remained gone.

Out of interest, what was wrong with the CC switch / unit ?

Good news that the ABS errors seem to have gone, I hate multiple faults, I am never quite sure if there is a link or not, and it can waste a lot of time. Let's hope they stay gone.

Forscan should have a DTC count PID, this can be handy as a permanent record of when an error happened.

 

[Dee_82]

good shout, ill have a look at that tomorrow. I drove about 60 mile today on CC and no problems

Ill add this data for reference, to me im seeing a lot more graduated changes in the vane position, it still has spikes which mostly even out when the negative values are corrected.

The only thing I can see that is a bit strange is the EGR Error % 

No limp mode and I might be imagining it but some of the flat spots seem to have gone.  Im going to get that EGR deleted on Wednesday anyway so im not to bothered by it but its interesting.

Im expecting there to still be problems but so far, nothing, ill swap out that MAP for a brand new unit.

I "fixed" the cruise control by absolutely bathing it in contact cleaner and pushing the buttons whilst doing it., the volume controls are still a bit funky, vol down sometimes skips track but the CC buttons worked perfectly. monitoring BCM the buttons clearly turn on and off.

Its along shot, and it will likely still have issues but it would be crazy to think that the CC buttons were messing with the CC control module, which was messing with the PCM, SASM and ABS, which all seemed to be messing with the PCM as well.... could that really causes all these problems, hesitations and limp mode??

14 monty test - Copy.fsl.txt

[Dee_82]

well, new update, so no limp mode after I fixed the CC buttons, I replaced the MAP for good measure but I couldn't see any sub ambient pressure numbers, all above 100kpa.

I also got the EGR mapped out and for good measure I figured id go for broke and have the whole thing remapped. Since shes a little long in the tooth I got a 165hp 370nm, off the shelf map.

No flat spots, no hesitation, that chap was pretty sure shes been sucking exhaust gas when she shouldn't albeit, it wasn't serious enough to flag errors, just moments of hesitation. the vanes appear to be acting normal and gradually changing (once corrected)

Im not ready to call this resolved since I seem to be having  hard time with this car and I dont want to get my hopes up but, if its not fixed then its sure as hell better then it was.

And damn, she pulls hard. 6th gear overtake from 70 to.... something, was a breeze, couple of seconds, job done, 5th even easier.

Hopefully this will be the end of this sorry tale! 

Really appreciate your support Peter and everyone else that had some input in to this.  I know we didn't strictly speaking find the problem in the data but …..

On ‎7‎/‎21‎/‎2018 at 11:07 PM, Tdci-Peter said:

Also I had noticed that on occasions the vane voltage changes seemed to be lagging behind the MAP signal. Normally if a MAP change is caused by the vane movement, the vane should lead. I am not at all sure about this, there may be odd delays in the CAN bus data signals. 

….. That comment led me to try and fix the CC and resolve those comms problems before going any further...

🙂

 

 

[pragmatix]

Fingers crossed, hope all is well from now on.

[Dee_82]

10 minutes ago, pragmatix said:

Fingers crossed, hope all is well from now on.

aye, next problem......, the "knocking" also known as.....

 

 

[Dee_82]

Im calling it resolved. 

No more drama at all, when I run a BCM test I am getting a low voltage on accelerator position error but its not present unless I do the test, no limp, no hesitation, EGR has been "Fixed" and all appears good.

Fancy that, CC button screw - ing with the SAMS ABS and PCM, introducing erroneous MAP and actuator readings....

 

[Dee_82]

Not resolved. 

It started dropping in to limp mode again, I removed the CC switches  and gave them a good clean, completely dismantled them but it made no difference

Soo. what to do from here, If I test the ABS I get a lost coms with SASM, all the buttons and stalks work perfectly. I can read all ABS PIDs and everything appears normal, I am still getting this accelerator DTC when testing the BCM.

Perhaps one of those issues are related to the limp mode.

I found the 3 grounds on each door at the front and resistance from each of them to the battery was fine. I checked and cleaned the 6 under the battery. 

on the accelerator peddle I checked the Power ground and the resistance was fine. on the ground via the BCM it was 20 Ohms,  I think ill need to find the connectors and check each wire separately.

This is actually doing my head in now, it drives absolutely fine (and better than before I cleaned those CC buttons) but its still not resolved.

 

[Tdci-Peter]

5 hours ago, Dee_82 said:

If I test the ABS I get a lost coms with SASM, all the buttons and stalks work perfectly. I can read all ABS PIDs and everything appears normal, I am still getting this accelerator DTC when testing the BCM.

Are you still getting any of the PCM codes like P2263?

And if the warning light is on, does Forscan say which code it is on for?

On 8/10/2018 at 12:39 PM, Dee_82 said:

when I run a BCM test I am getting a low voltage on accelerator position error but its not present unless I do the test

Any accelerator position DTCs?

Accelerator errors are regarded as very serious, there will be two sensors that are constantly compared, and differences will flag up a DTC. The BCM test may be more severe, and test for mechanical offset or slack in the system, that affects both sensors and is not detectable in normal driving. If there are no DTCs, it is hard to see how this can cause "limp mode", though it could cause reduced power directly if a mechanical fault meant the two sensors were not being fully operated. But this seems very unlikely.

In general terms, it might be important to differentiate between reduced power mode deliberately caused by the ECU because it thinks there is an error that will damage the engine or be dangerous, and reduced power because of loss of some function like accelerator travel or turbo actuator that creates the power.

Limp mode (caused by the ECU) should always be signaled by a DTC, but direct loss of power can bypass or fool the ECU.

[Dee_82]

The Limp mode is always triggered with the following DTCs being visible, normally driving it a steady pace without changing of acceleration, the MAP and MAF have been replaced and contacts cleaned.

===PCM DTC P2263:73-27===

Code: P2263 - Turbo / Super Charge Boost System Performance

===OBDII DTC P2263-C===
Code: P2263 - Turbo / Super Charge Boost System Performance

===ABS DTC U0401:68-8A===  ***EML ON***

Code: U0401 - Invalid Data Received From ECM/PCM

 

If I use Forscans ability to test the ABS module I get:-

===ABS DTC U0126:00-8A===  ***EML ON***

Code: U0126 - Lost Communication With SASM

 

If I Use Forscans ability to test the BCM module I get:-

===PCM DTC P2127:00-2E===

Code: P2127 - Throttle/Pedal Position Sensor/Switch E Circuit Low Input

===PCM DTC P060D:00-2F===
Code: P060D - Internal Control Module Accelerator Pedal Position Performance

 

Sometimes I can get this one as well, thats because of a dodge solenoid on the rear drivers side door. It needs cleaned up.

===BCMii DTC B1108:77-6C===

Code: B1108 - Driver door central locking motor

 

In terms of Driving, the accelerator for the most part works fine, ive run countless traces on it and both E and D almost always match up perfectly, you will see on the picture below that there is a slight spike in APP E where there isnt one on APP D, that spike coincides with an increase in RPM and MAF and MAP so the spike happened, what didn't happen was any input on the accelerator. I have been able to recreate this when I come to a stop and edge forward with traffic, after I come off the accelerator and the clutch pushed down there is a momentary blip in RPM.  I purchased a newer second-hand accelerator the other day and connected it up, it made no difference to the DTCs being produced. Both returned after being cleared.

At no time do I get any of the APP DTCs whilst driving around, only when I do the BCM test, if it wasn't for the blips in acceleration id have just put this down to a forscan problem. 

 

[Dee_82]

Not sure how relevant this might be but ive been Doing some digging on 

===ABS DTC U0401:68-8A===  ***EML ON***

Code: U0401 - Invalid Data Received From ECM/PCM

Google suggests that it might be that Engine speed data doesn't match what the ABS sensors are telling it. 

Feels a little like Alice in wonderland this!