Mk4 Mondeo 2.0 TDCI Limp Mode

[Dee_82]

Following on From Toms mk3 thread, it seems ive ran in to a similar problem on the Mk4 (Euro IV) ! 

Coming back up the road from Suffolk, got about 100 miles up the A1(M) when she suddenly dropped in to limp mode. car was going absolutely fine up to that point, was cruising at 70 in 6th up a slight incline.

got to a service station, reset, another 100 miles an bam, off it goes again followed by another instance about another 150 miles later. in each instance these were the codes, no noticeable smoke.

 

(WARN) [14:14:38.661] DTCs in PCM: U0405:00-6C, P0691:00-6C, P003A:16-28, P2263:73-64

U0405 - is a cruise control issue, Its always been there and I dont think its related. theres a corresponding DTC on the BCM as well, I have a feeling its to do with that thermal fuse in the steering column.

P0691 - Cooling Fan 1 Circuit low, Not sure whats up with this, the fan appears to work fine, car isnt overheating...

P003A -  Boost Controller has exceeded learned limits

P2263 -  Seems to be a catch all code, from EGR, actuator and MAP
 

So ive pulled off the MAP and ICT, they were pretty dirty, so gave them a good clean. I was going to check the vacuum hose on the boost solenoid tomorrow, Turbo is completely inaccessible from up top so will need to wait before I can check for movement. ill log some stuff tomorrow, VNT position, MAP and see if I can get forscan to relearn the learned values.

That catch all code and the boost controller code is making me lean towards the actuator / vanes but I dont suppose you guys have seen similar issues? or have anything other areas to check out? 

If its the Actuator and if I can get to it, I can swap that out, but the vanes... if I have shoddy movement of the arm, do you have any suggestions on whats the best course of action? @stef123 @iantt @TomsFocus

[Dee_82]

just drove 30 miles to work, no errors.

so what ever is going on is either intermittent or only happens after 100 miles or so.... or the map sensor clean did the job.

 

[TomsFocus]

Well that's odd...but a good result hopefully lol!

[Dee_82]

Lets hope so! Stupidly I forgot to charge my laptop the night before so couldn't run forscan whilst driving as I needed to clear the codes so no snap shot of it going wrong.

The turbo vanes seem to operate from -150 to 150, at idle it sits around -55 or there about.  At the start of todays trip it seemed to sit at -55 as normal and would spike to 150 under acceleration.

however once on the motorway she would just operate from 150 to -150, quite erratic perhaps?

Im pretty sure -150 to 150 is the normal range for the vanes

EGR is ranging from 1.14v to 3.89v

MAP is 1 to 2.5 BAR 

MAF is 5 g/s to 146 g/s

The above two match engine load without any spikes.

 

[Dee_82]

righty, so im talking crap, fan isnt working at all, no issues with overheating until im sitting still! where it will happily climb up to 100c and the P0691 code appears

I have 12v across the 2 heavy wires entering the wee speed controlling box thing attached to the radiator. bypassing that little control box allows the fan to spin freely.

Im having a bit of a bad week on this one so any suggestions would be really helpful

my first thought is that little pink / purple 3rd wire going to the control box, from what I gather that can snap in half?

My second through was, whats the chances my limp mode problem was actually related to overheating (albeit, I received no temp warning) and the distance wasn't important, it was simply the length of time it took to heat up too much?

 

 

 

[Dee_82]

 

So the limp problem hasn't  happened again but then I haven't driven for more than 100 miles yet so that doesn't really mean anything.

Yesterday I was able to put a vacuum pump on the actuator and it moved, I didn't have time to pop the IC pipe off so couldn't feel it move but there was movement of about 1cm or maybe slightly more.

So today I recorded another run and noticed several instances like this..... 

 

 

 

you can see that initially the EGR, VNT MAP etc more or less sing in tune when I came off the throttle at 1565.

at 1573 I applied a small amount of throttle to maintain speed, again, that all matches.

at 1575 the Actuator sensor goes ape with no appreciable changes of RPM or speed, MAP and EGR more or less remain constant. 

Would I be right in thinking if the actuator was actually doing that, moving from fully closed to fully open there would be some changes to the MAP? or anything!

Below is it zoomed right in, I cant see any changes in anything. the EGR slightly changes but not much.

 

 

Would a faulty actuator sensor be suspect here? or solenoid or even the actuator?

Is that sensor on the same loom at the fan PWM cable? if one cable is corroding are the others go as well? 

😞 

 

 

[pragmatix]

Not 100% on the read outs but i would think it could just be a sensor thats giving you problems.

[TomsFocus]

Should've stuck with the DV6 reliability...  

As you say, the vanes can't actually be doing that without some change in MAP at the very least.

Just FYI, it's a different engine to the one in the Mk3, that was Ford and yours is PSA, turbo regulation is also different between them.  1cm vane movement is about right for these so it sounds like the vanes are probably ok as long as it moved smoothly.  I would agree that it looks like an issue with the position sensor or the wiring to it.  If you're still taking live diags, load it up in the wrong gear uphill, if it's going to overboost due to sticky vanes or the actuator not working those are the conditions that will do it, and it takes a few seconds of overboost to cause to limp.  I don't think it will go limp though, I think it's more a 'random' electrical signal issue from what you've said so far.

No idea on the rad fan wiring though...but that may explain your dodgy air con!

 

[pragmatix]

Dodgy air con will be down to the broken pink wire you mentioned earlier, i didnt pick up,on that its a common fault.

[Dee_82]

Cheers, Guys, 

Found the broken pink wire and I feel like ive aged 10 years trying to fix, she gets pretty toasty before it turns on but it spins up so happy days! no longer need to cook myself to regulate the temps!

AC gas has entirely vented but I suspect that wire killed the fan which then disabled the AC which then destroyed some seals some place and killed a few polar bears. was tempted to buy a vacuum pump and gauge set to try and find it but thats a major job and whilst AC would be a god send just now, im thinking theres bigger things to take care of.

I think the next thing is to rule out the EGR and  that Actuator sensor, ill keep doing some live runs and see what happens, interestingly, now you mention it, I have a feeling the times it did cut out was in 6th up a slight incline. it didn't sound like it was struggling but it was about 1500 RPM, so on the low side.

Appreciate the support guys! Thanks again!

 

 

[Dee_82]

Limp mode.

Less than 12 miles, I kept it in a high gear to put some strain on the system. its not generating over boost, its just not moving at all. 

interestingly, I noticed today that the MAP dropped to 0.6BAR, or ~8PSI seems a little ow given that it was absolutely roasting, if anything it should have been slightly higher than atmospheric. maybe unrelated.

So below,

• 1180, Limp mode, Actuator position constant, for approx. 70 seconds. 
• 1182, I decide to kill the engine and restart, actuator still frozen, 
• 1184, restart engine, actuator moves a bit. limp mode removed.
• 1189, accelerate a bit
• 1195, Limp mode (that last spike)
• 1196-, Actuator remains at 1.5V (which is where it sits with ignition on but engine off before starting) at idle it usually its around -56v

Cant get a new Actuator on its own according to TMS and ECP, needs to be the whole turbo. can probably get a new sensor from eBay for about £40 for something with a name im familiar with (Melett)

New Turbo gonna cost about 500 and if I cant do it, labour for good measure.

I dont want to pee away £40 quid on a sensor if it doesn't fix it then have to blow about 600-700 on the turbo. At a complete loss what to do next...

I suppose I should probably find out if this actuator arm moves the vanes freely 

FFS, At this rate it would have been cheaper refitting a certain part to my old car 😞

on the plus side, fan works!

 

 

[TomsFocus]

The vane setup isn't easy to DIY which is why Garrett don't sell new actuators separately from the turbo.  Ford buy the complete turbo units from Garrett so you can't buy individual parts from Ford themselves either.

With the turbo on the mk3 though, breakers sell the used electronic actuators separately...so I wonder if the same is possible with the vac actuator on your mk4...  May not be perfect, but it'd be cheaper than a new turbo.

Did it throw a fresh code when it went limp this time?

 

 

[Dee_82]

yeah, same as before, this time it was a hard fault, on the second turn off and on whilst driving, I got the whole engine warning light as well as management warning. but that might have been because I was trying to be sly and reset it on the move, probably not the safest thing to do but anyway. no other faults other than the generic boost / actuator / EGR / MAP fault. garage cleared the code but im sure it was the same as before. garage confirmed my own thoughts, "something is sticking"

im looking at this actuator and I dont think I can get it off in situ, I can probably take off the sensor without bother but there are 3 bolts plus the one on the arm itself which look like a complete bugger to get to. id need to remove the cross member, CAT and intercooler pipe for proper access. 

[TomsFocus]

Hmm, for the sake of £40 and the least hassle I'd probably worth trying the sensor first in that case.  You can always sell it on again if it doesn't make a difference to reduce some of the losses.

Tbh you could just remove the sensor and check that it's rod moves smoothly, maybe clean and lightly lube it if not?

 

[Dee_82]

Argh, ill take a punt on the sensor then, should be able to change it with minimal fuss.  Ill need to try test the boost solenoid somehow too.

I have some spray lithium lube that might do the job, o you know the deal with Lubing these things up, must get pretty toasty down there.

[Tdci-Peter]

10 hours ago, Dee_82 said:

I noticed today that the MAP dropped to 0.6BAR, or ~8PS

I saw on the Forscan trace that the MAP dropped below 1 Bar just as the engine rpm dropped towards zero. This must be the intake throttle valve closing, in its anti-judder mode. That (apart from totally blocked filter, which you don't have!, or a MAP fault), is the only way I can see to get MAP readings appreciably below atmospheric.

The varying signal on the vane looks just like an open circuit to me. If the sensor is just a standard mechanical pot (variable resistor), then a fault in the wiper could do it. Or a broken wire or duff connector. I have had similar wild variations in my EGR signal, due to faults in the pot in the position sensor.

The graph is a bit fuzzy on my system, I can barely read the numbers on the time axis, even after downloading it and zooming in. Windows does muck about with picture resolutions, even though it is a .PNG.

 

[Dee_82]

Hi Peter, Thank for the reply 

Ive attached a ZIP file which contains two forscan traces from today, you will need to download the file, rename it to something.zip, then extract the two files.

Forscan should open it up nicely

Didn't have any bother on the way home, I kept it in 5th and pulsed the accelerator, my thoughts being, just keep that actuator arm moving about might help free up any problems. No idea if it helped but it didn't go in to limp, so that has to go in the win column right!? 

something else you might see on the return leg is that the actuator arm egr and MAP all seem to be more or less in sync, where as on the way to work, it was ok then completely lost it.

That MAP point makes sense, drops below zero as there is still an element of each cylinder sucking air, valve closes, drop in pressure. Ill need to see if I can find a PID for the throttle valve, would be useful to know what thats doing as well.

 

FORScan.txt

[Tdci-Peter]

14 hours ago, Dee_82 said:

Didn't have any bother on the way home, I kept it in 5th and pulsed the accelerator, my thoughts being, just keep that actuator arm moving about might help free up any problems.

A lot of info there! I have only been able to have a quick look. Keeping the APP moving may have been good for the car & driving, but it is not so good for diagnosis. Small APP changes are hardly visible on the compressed graph, but have a major effect on the other traces.

But on the home trip, Monty8, I did see a few unexplained big swings in the vane position. See pic below. APP was completely constant.

I do not like the look of that vane position trace at all. It seldom hovers in mid travel, and the voltage values (+/-150v) from Forscan are physically very improbable. It could be +/-150mV, as Forscan seems a bit prone to this sort of error, the Time axis claims to be ms, when it is clearly seconds (on all my runs too). If it was +/- 150mV, that could indicate a broken sensor or wire.

However I do not know for sure what it should be like. I have not managed to locate a vane position PID on my 1.8 engine, despite looking. I must check with the latest Forscan version, as my earlier version did not accept some of your PIDs. But I know that if it was swinging about like that, the poor turbine would have low efficiency due to badly unstable air flow. (Air flow & pressure has been a major part of my work for the last 30 years!)

For anyone else foolish enough to be interested (!), the list of PIDs is:

FANDC    0.00 %    Variable Fan Duty Cycle
ECT    79 °C    Engine coolant temperature
EGTAC    303 °C    Exhaust Gas Temperature After Catalyst
VSS    59.7 mph    Vehicle Speed
RPM    2140 rev/min    Engine Revolutions Per Minute
APP    37.26 %    Accelerator Pedal Position
MAP    1.80 Bar    Manifold absolute pressure sensor
MAF    65.81 g/sec    Mass Air Flow
EGRVP_V    1.19 Volt    Exhaust Gas Recirculation Valve Position
Vane Pos    155.75 Volt    Turbocharger Variable Vane Actuator Position
DP_DPF    7.1 kPa    Differential pressure on particulate filter

 

 

 

[Dee_82]

Thanks peter, those random moments of unexplained  fluctuations happen periodically. no pattern that I can see, as you may have gathered ive got 8 trips worth of data and whilst the limp mode only happened on #7, the others all exhibit the same random fluctuations at various point. 

I googled the range and the +- 159(m)V looks right, fully open to fully closed, it should idle at around -50(m)V 

I found an old test run I did on my 1.6TDCI which monitored the vanes, its not using voltage as a measurement but instead uses % 

Theres much less fluctuation on that and as you said, its much more "regulated" in that it maintains values between 0-100% as well

The question is, what would cause it to fluctuate?

Either its not and the PCM is just "seeing" or it is and...what the boost solenoid is going mental? ...

 

 

 

fuel railproblem[862] - Copy.txt

[Tdci-Peter]

3 hours ago, Dee_82 said:

The question is, what would cause it to fluctuate?

Either its not and the PCM is just "seeing" or it is and...what the boost solenoid is going mental? ...

I really do not think the vanes can actually be moving like that. No sign of it on MAP, MAF or EGR. The EGR is probably controlled by measuring the difference between MAF flowrate, and a flowrate calculated from RPM and MAP and intake air temp. The latter is the engine air flow, including EGR flow, the MAF flow is net air in, not including EGR flow, so the difference is EGR flow, which will be controlled to a target value. Hence EGR position is an amplified, and rather noisy, combination of MAP & MAF.

If it is caused by mad variations in the boost solenoid, then for some reason it does not seem to be having much effect on the vanes. Or maybe the vanes are flicking off-on so quickly the turbine has no chance to respond and spool up, so no effect on MAF.

Repeating the vacuum pump test on the actuator while looking at the vane position reading with Forscan might narrow it down a bit. The reading should work in Key on / Engine off situation,

 

[Dee_82]

Righty, been at her all morning. 

I've done as you suggested, I'll post the recordings when I work out how to get them off my phone (laptop battery died) 

I was able to connect a long hose to the actuator vacuum pipe at the point it would normally connect to boost solenoid. Under the car with my hand held vacuum pump and my phone I clearly observed the arm moving, it appears to move freely when being sucked. 

When I release the vacuum it smoothly returns back to the starting point.

The sensor is reporting 1.8mV with no vacuum.

Each time I pump it rises in steps of about 7mV each step holds steady for the few seconds I left it. Once it hits +150mV, the next pump then shoots to -150mv, successive pumps then rise from -150mv  up to -50mv. I was able to bleed of the vacuum in a semi controlled way and saw the value drop to -150 then shoot up to +150 before slowly dropping down.

So in my mind that's either really weird or it's designed that way. In either case it seems to be moving smoothly.

 

LiveData_2018-06-30_11.27.25.txt

[Tdci-Peter]

2 hours ago, Dee_82 said:

Each time I pump it rises in steps of about 7mV each step holds steady for the few seconds I left it. Once it hits +150mV, the next pump then shoots to -150mv, successive pumps then rise from -150mv  up to -50mv. I was able to bleed of the vacuum in a semi controlled way and saw the value drop to -150 then shoot up to +150 before slowly dropping down.

So in my mind that's either really weird or it's designed that way. In either case it seems to be moving smoothly.

I have re-processed the data to remove the +159 to -159 jump, by adding 318 to all negative readings. Then the graph looks a lot more sensible. I can't output the modified graph or modified data at present.

There are two possibilities I have thought about:

A Forscan programme error: PIDs are stored in a binary form in the car, and in some conventions, the MSB is used as a sign bit (2s complement arithmetic), in other conventions the MSB is normal. So in 8 bit binary, 127 is hex 7F, 128 is Hex 80. Hex 80 can be interpreted as 128 or as -128, depending on the convention used. Scaling factors can change this switch point to any value.

A sensor error: Some change in the operation of the sensor, like if it was a pot, a shift of the wiper arm to go through the track ends.

In favour of the sensor error, at each +/- transition, there is a ramp of about 10 readings. I can not see why Forscan would introduce this. If the error was in the sensor, there is very likely to be some noise filtering where the signal goes into the PCM, before it is digitised, which will convert any sudden change into a ramp.

[Dee_82]

Cheers Peter, 

The sensor looks to be in pretty poor state, I should be able to whip the old one off and put a new one in without much bother though.

Id be over the moon if it is sensor related and to be honest, after seeing it move today it appeared ok, again I couldn't get my hands on it to move it directly without taking a bunch of stuff off but I was more than happy with the way it glided back after removing the vacuum. obviously thats not to say that it doesn't jam up but its a good indication its not totally choked with carbon !

Whats your thoughts on the boost solenoid not playing ball, if the PCM is telling the solenoid what to do and its not doing it, id imagine it would have something to say about that as well?

[Tdci-Peter]

On 6/30/2018 at 5:35 PM, Dee_82 said:

Id be over the moon if it is sensor related and to be honest, after seeing it move today it appeared ok, again I couldn't get my hands on it to move it directly without taking a bunch of stuff off but I was more than happy with the way it glided back after removing the vacuum.

I can see from the traces that the arm is moving smoothly. The obvious error is the big jump. It could be a red herring, just a software error, but the ramps and the fact that the PCM is complaining about something related to that signal, does suggest it is a real error in the sensor.

I can not see how the solenoid could cause such a consistent sensor shift, always at the same values. And the rest of the trace looks reasonable.

I have managed to output a couple of pictures of the re-processed vane trace. One zoomed out a bit to show some changes, one zoomed in to highlight the ramps that happen exactly the same at each polarity change.

[Dee_82]

Thats really interesting and as you say that trace looks much more sensible with the corrected values. May I ask what software your using to open those CVS files like this? Excel goes completely mental with the quantity of data and its all but unusable.

Ive not got this sensor replacement yet but it appears to be just 3 TX screws. ill need to pop this off in situ so I just hope to buggery that it doesn't pop apart like some spring loaded clock. if anyone knows how this comes apart it might be really useful.  The sensor looks like https://www.ebay.co.uk/itm/Turbocharger-Actuator-Position-Sensor-Ford-C-Max-Focus-Galaxy-Kuga-2-0TDCI/272882384851?hash=item3f890ebfd3:g:MBgAAOSw44BYggjW

I also managed to get some of that "Revive" stuff which seems to get a good write up by most folk who use it, it just sprays in to the inlet (post MAF) works its way in to the compressor and eventually the turbine, its probably glorified water but its worth a punt since its clearly not a massive issue, just an intermittent annoying one!!

Really appreciate your input!

 

Many thanks