Overboost problem, 2005 Patrol

[Timbo_1975]

A bit of a move forward on this. I have now installed a boost gauge (stepper motor, electronic type) in the cab and connected it to power and have run the signal cable into the engine bay. I have yet to connect the gauge sensor into the intake airstream - a job for tomorrow.

As the gauge is a dual purpose vacuum/boost gauge, it occured to me that it might be interesting to see what the ECU & VNT control solenoid are asking the turbo to do by temporarily teeing into the hose between the solenoid and the VNT actuator and using the gauge to measure the control vacuum in the hose - and then have a drive around and see what's happening.

1) At tickover, the VNT lever is at the top of its travel (or near - I haven't checked with a feeler gauge yet) with the vacuum gauge reading 18-19" Hg. I *think* this is pretty much what it is supposed to do.

2) Over a gentle run of about 5 miles on mostly small throttle openings, the vacuum readings stayed at or near 18" Hg. It only fell below that reading if I leaned on the throttle. Even leaning hard on the throttle (at any engine speed & gear) rarely took the vacuum reading below 10" Hg (maybe to 7 or 8" Hg) and then only momentarily if I suddenly floored the throttle pedal. I suspect (but don't know) that this is normal behaviour).

3) I headed towards a junction onto the M62 heading east (motorway/highway heading up the Pennine Hills). Before going onto the M62, I stopped and cleared an 0905 ECU fault code. I tried to keep a fairly steady speed around 65mph (2600-2700 rpm in top) up the hill on maybe half throttle or so. Though I didn't notice any reduction in performance (though maybe I was daydreaming), on a couple of occasions without any change in throttle opening, the vacuum wandered back & forth between 14-15" Hg and 10"Hg. Lifting off the throttle and then reapplying sent the vac straight back up to 10"Hg. When I got back home, I discovered that the ECU had stored another 0905 code.

I can't help wondering why the vacuum never goes (much) below 10"Hg - and why when the 0905 overboost occurs it doesn't go back up to 18-19"Hg. Does anyone have any insights on this?

Andrew

Just say again where in the vacuum lines you have your gauge connected to the vnt unit??

At engine stop you will have no vacuum and the vnt unit should move the vanes to the fully open position. At engine start, full vacuum should be applied and pull the vanes back to the fully closed position against the stop.

During running, the ecu will bleed off vaccum to modulate the vanes position. This will depend ALOT on load, engine speed and amount of EGR required. At cruise on a light throttle you should see a constant reading which will suddenly rise and hold as the ecu applies more EGR.

Really you need a boost gauge fitted as well to see what is going on, but i would suspect you have the early signs of a sticking egr valve.

As I discussed at some length abit ago, the standard boost control / egr maps on the 3.0di are f**king dreadful which is why its very prone to overboosting , excess egt's and piston / HG issues.

Either you mod the vnt actuation to manual control and blank the egr, OR you have a remap done by someone who is familiar with the VW 1.9 TDI PD which has the same set up with an n45, so that it actually modulates correctly, disabling the egr again.

 

[andrewk]

Just say again where in the vacuum lines you have your gauge connected to the vnt unit??

It is (temporarily) teed into the hose between the VNT actuator and the VNT control solenoid.

At engine stop you will have no vacuum and the vnt unit should move the vanes to the fully open position. At engine start, full vacuum should be applied and pull the vanes back to the fully closed position against the stop.

Indeed so - and that is exactly what does happen. Didn't I explain that earlier in the thread?

Really you need a boost gauge fitted as well to see what is going on

That is exactly what I am in the process of installing - see my previous post. As the gauge is a dual vac/boost gauge, I've just pinched it temporarily to check what vac is being applied to the VNT actuator. Now that I know that, I'll remove that tee, reinstate the hose from the VNT actuator to the control solenoid and finish installing the boost gauge.

Andrew

 

[andrewk]

I've now replaced the vacuum hose between the VNT control solenoid and the VNT actuator "just in case". I've also finished the installation of the boost gauge and had a drive around to see what is happening. I need, though, to be a bit more scientific and plan it properly. I've certainly learned a few things though ..........

In normal footling around at low speeds & small throttle openings, I'm seeing boost pressures of only a couple of psi. Boost seems to start at about 1200rpm ish. Leaning on the throttle a bit, again at low speeds sees boost pressure rise to somewhere in the region of 10-15psi. Once the engine was good and hot, it occasionally flickered up towards 20psi but only for a fraction of a second before again settling somewhere between 10-15psi whilst accelerating and then dropping back below 5psi at a constant 50mph in 5th on the flat.

After a few miles of wandering about, I was feeling quite optimistic as I hadn't yet seen much amiss (I think) - so, I decided to take it for a blast on the M62 motorway (Highway, I think, in Oz speak). There is an uphill section about 5 or 6 miles long which I would normally drive at anything up to 70mph (110kPH) in 5th. I gradually increased speed to about 2500rpm in 4th. Boost pressure stayed somewhere between 10 and 15psi. I then floored the throttle and held it there. Boost pressure increased to 25-30psi & after 5 or 6 seconds at round about 2800-3000rpm, I got the sudden power loss that I associate with an overboost error. The boost pressure immediately & quickly dropped to a low level, just a few psi. Lifting off and reapplying throttle restored normal behaviour. Definitely an overboost error, I think. Just to be sure, I repeated the exercise a couple of times and got the overboost error each time.

Just to clear up a point or two about sensors.

* I did swap the MAF sensor about 12 months ago (for a cheapie off Ebay) with no evidence that the olriginal was faulty - just an attempt at a quick fix.

* The quick fix didn't work, so I swapped the MAP sensor (boost sensor) for a genuine new Nissan Part, as the next most likely culprit. For 8 or 9 months and about 9000 miles (including maybe 2000 miles towing a 1500kg caravan) I didn't see a single instance of an overboost - until Xmas 2010 when it happened again. Nissan kindly replaced the boost sensor under the part warranty. Replacing it again didn't fix the problem.

* After my drive around this afternoon, I decided that I'd check the voltages on the two boost sensors. The latest one produces a signal of about 2.7 volts at atmospheric - a tad high, but still within spec. The original boost sensor (i.e. fitted from new and replaced as faulty 12 months ago) produces 5.6 volts at atmospheric - so, totally goosed!!! The goosed sensor was, I think, almost certainly responsible for the original overboost symptom.

Ponder ............. is it possible that I replaced the original MAF with a doubtful one that was just about OK when I installed it - but which has subsequently failed? I still have the original OEM MAF sensor (40,000 miles from new when I replaced it) - so I'll test the voltages on both in the morning and slot the old MAF sensor back in (now cleaned properly) and have another drive around tomorrow. I don't really feel very confident - but who knows .......

 

[andrewk]

Ponder ............. is it possible that I replaced the original MAF with a doubtful one that was just about OK when I installed it - but which has subsequently failed?

I need to do some more testing, but I think I might know the answer to that one!!! It's a definite maybe.

I've swapped the MAF and reinstalled the original Nissan OEM MAF (that one that was in the motor from new and has done about 40,000 miles) and have been for another drive. Boost pressure seems to be a lot more predictable.

At constant speeds, on the flat or uphill, in any gear, boost pressure doesn't seem to go above 15 psi.

At medium engine speeds in 2nd, 3rd or 4th, flooring the throttle sends the boost pressure up to about 15 psi instantaneously. It then falls quickly back to 10 psi and gradually rises again as the engine revs rise, if you keep your foot to the floor. Even being brutal and keeping my foot to the metal up to 4500rpm only saw 18 psi.

I'd say it was fixed, apart from one incident on the motorway. I was going a steepish hill at maybe 2500 rpm in 4th. I leaned on the throttle a bit and just as boost pressure increased to 15 psi, floored it. It went well over 20 psi, which is not good news. It dropped back very quickly, but I'm not sure whether the ECU did that or whether I lifted off to avoid smashing into the back of a truck that was only doing 40 mph (not a great idea!!).

I think I need to do some more testing - and maybe give it an "Italian" tune up. Just maybe I'll buy a brand new (genuine) Nissan MAF sensor too. I can't be certain yet, but I think I'm quietly confident.

Andrew

 

[andrewk]

I can't be certain yet, but I think I'm quietly confident.

I take that back. I'm not in the least bit confident. :-(

I've been around the same route again and have managed to generate the overboost/limp on a couple of occasions. One was a bit odd, because it occured at a boost pressure UNDER 20psi - don't know what to make of that. In both cases, though, I was being a bit brutal - under normal driving everything was fine (with the other MAF it wasn't)

Well, I've disconnected the battery for a bit as I changed the MAF - and I keep reading that making the ECU re-learn it's sensors can produce positive results. I'll have another run around tomorrow - and maybe buy a new (genuine) MAF if there is no improvement.

Hmmm .......

Andrew

 

[Timbo_1975]

Have you checked the output from the MAP sensor. This is the ONLY reference that the ecu has for actual boost pressure. A spike or eroneous signal will set the code.

 

[andrewk]

Have you checked the output from the MAP sensor. This is the ONLY reference that the ecu has for actual boost pressure.

The current MAP sensor (new, genuine Nissan) is only a couple of months old. Output is 2.7V at atmospheric - so probably not faulty. The original (fitted when new) MAP sensor output at atmospheric is 5.6 V, so it was definitely faulty/out of spec. (see three posts back).

I don't have a means of measuring the output under pressure while driving.

Andrew

 

[Timbo_1975]

Hmmm inthat case you really need it on a Consult and do some live monitoring.

I bet your boost is spiking and the stepper motor gauge is too slow to react.

 

[andrewk]

Hmmm inthat case you really need it on a Consult and do some live monitoring.t.

I don't think that the ECU makes the MAP sensor voltage available to Consult (or equivalents), does it?

Andrew

 

[rayf3262]

I don't think that the ECU makes the MAP sensor voltage available to Consult (or equivalents), does it?

Andrew

NO it does not.
I would temporarily replace the MAP sensor with with a fixed voltage that the ECU is happy with. Then test drive it.

 

[andrewk]

There is a way of testing the MAP sensor in the dealer workshop manual. Basically - leave the sensor connected to the wiring harness, connect it to a hand pump, raise pressure in stages whilst measuring output volts on the MAP sensor's centre pin. I *have* verified that it is OK at atmospheric, but not at any other pressure. I don't have a hand-pump with a pressure gauge - I've been trying to think of Heath Robinson ways of applying a known pressure using an inflated tyre and a bit of vac hose. If I were really convinced that I could have a MAP sensor problem, then I would find a way of measuring it's compliance to the calibration curve in the workshop manual (Heath Robinson or otherwise) BUT - it does seem to be doing what it should.

I'm *fairly* (but not very) confident that the turbo is OK. There's no evidence that the vanes are sticking and quite a bit that they are not. The VNT actuator itself works. The vacuum system and VNT control solenoid seem to be doing what they should.

I want to be convinced that the MAF sensor is OK. There was a definite change (improvement) in behaviour after re-fitting the old (original) sensor. Tomorrow, I'm going to check its output volts at tickover and every 500 rpm to at least 3500 rpm. If I'm not convinced that it's spot on then I'll have to buy another. If the MAF is OK, then I'm running out of suspects and maybe I'll have to start listening to those who suggest that the EGR valve might be sticking. I don't know whether it is easy/hard or even possible to test and/or remove, clean and "unstick" an EGR valve - but I just might have to find out.

OTOH, I could always trade the 'Troll in against a Disco.

Andrew
(I didn't mean it - honest)

 

[rayf3262]

For what it's worth the MAP sensor output is not a mapped input, it is simply range checked and fairly infrequently compared to other sensors.
Hence the reason it is not available as live sensor data.

You can simply unplug it but that will result in an out of range condition straight away.
Eliminating this device with a fixed voltage that the ECU is happy with on on start would suffice as the data from this sensor is not live mapped only rather crudely limit checked. Only if the output is way out of range at a particular moment will the ECU drop to a safe (no boost) map. I believe this is why the overboost condition can prevail for some seconds before finally deciding enough is enough.
There is of course always the issue of signal integrity from any of the sensors to the ECU, which is why its a nuisance that this particular sensor is not available as live data.

 

[andrewk]

Eliminating this device with a fixed voltage that the ECU is happy with on on start would suffice as the data from this sensor is not live mapped only rather crudely limit checked.

I do understand what you are saying and can imagine how to implement it - BUT I cannot for the life of me understand why I would want to do that.

Surely, if I am hiding the true level of boost from the ECU then it doesn't know to generate an (inconvenient) overboost error if the boost pressure were to go to 25 psi and stay there. Isn't that a great (and extremely inconvenient & expensive) way of smoking the engine - or am I missing something?

Do explain ...........

Andrew

 

[rayf3262]

I do understand what you are saying and can imagine how to implement it - BUT I cannot for the life of me understand why I would want to do that.

Surely, if I am hiding the true level of boost from the ECU then it doesn't know to generate an (inconvenient) overboost error if the boost pressure were to go to 25 psi and stay there. Isn't that a great (and extremely inconvenient & expensive) way of smoking the engine - or am I missing something?

Do explain ...........

Andrew

Just a case of eliminating that particular sensors possible erroneous input, I am not suggesting it is a fix just a simple diagnostic procedure.
At the very least it separates MAP from MAF.

You have a boost gauge so it not as if you will be running on unknown/undesirable boost pressure.

 

[andrewk]

Just a case of eliminating that particular sensors possible erroneous input, I am not suggesting it is a fix just a simple diagnostic procedure.
At the very least it separates MAP from MAF.

You have a boost gauge so it not as if you will be running on unknown/undesirable boost pressure.

One thing I do know is that the boost is going well over 20 psi under some circumstances and the ECU then generates an overboost error - so the MAP sensor seems to be doing what it should. I'd be more concerned if the boost pressure was way too high and there was no overboost error.

I *think* the shortlist of potential culprits (for crime of generating an excessive boost pressure) is now down to two - the MAF sensor (still my personal favourite) or (as suggested by a number of folks) a sticking EGR valve, though I still don't really understand the means by which a sticking EGR can cause it.

As the MAF is very conveniently located and easy to test, I will have condemned it or pronounced it innocent of all charges before tomorrow lunchtime. If found innocent, I'm going to have to develop a strategy for sorting out the EGR valve. Has anyone managed to (successully) remove/clean/unstick an EGR valve on a ZD30? Any tips on the best/easiest way to do this?

I know that many folks block EGR valves to fix this kind of problem - but (apart from it being illegal in the UK) I would then be catapulted into having to install an exhaust gas temp gauge and re-calibrate the turbos VNT actuator - which in my book is a kludge. I'd rather diagnose the problem and fix it than just hide the symptoms.

Does anyone know how much a (genuine) new MAF sensor is - or a replacement EGR valve? How many megabucks?

Andrew

 

[Timbo_1975]

One thing I do know is that the boost is going well over 20 psi under some circumstances and the ECU then generates an overboost error - so the MAP sensor seems to be doing what it should. I'd be more concerned if the boost pressure was way too high and there was no overboost error.

I *think* the shortlist of potential culprits (for crime of generating an excessive boost pressure) is now down to two - the MAF sensor (still my personal favourite) or (as suggested by a number of folks) a sticking EGR valve, though I still don't really understand the means by which a sticking EGR can cause it.

As the MAF is very conveniently located and easy to test, I will have condemned it or pronounced it innocent of all charges before tomorrow lunchtime. If found innocent, I'm going to have to develop a strategy for sorting out the EGR valve. Has anyone managed to (successully) remove/clean/unstick an EGR valve on a ZD30? Any tips on the best/easiest way to do this?

I know that many folks block EGR valves to fix this kind of problem - but (apart from it being illegal in the UK) I would then be catapulted into having to install an exhaust gas temp gauge and re-calibrate the turbos VNT actuator - which in my book is a kludge. I'd rather diagnose the problem and fix it than just hide the symptoms.

Does anyone know how much a (genuine) new MAF sensor is - or a replacement EGR valve? How many megabucks?

Andrew

There is no legal requirement for a working egr valve, not here and not in any country. All you have to do is comply with the emissions standard, which in this country is via an opacity test.

Correct that with disabled egr you would need to make some other knock-on adjustments for it to run right, and really for serious use of a ZD30, an egt gauge would be essential anyhow.

I have had some success cleaning egr valves in an ultrasonic bath using a mixture of degreaser and washing soda, followed by some lapping paste, but not all would work reliably after that so were replaced with new.

For the version of the ZD30 (detuned slightly) that is used by vauxhall in the movano the egr valve is around £160+vat which is a guide.

 

[andrewk]

Let's just summarise where I think I am up to in trying to understand what is going on here ........

1) The turbo itself is probably OK, at least there's no evidence that the vanes are sticking. The VNT actuator itself is holding vacuum. On tickover, the vacuum in the hose from the VNT solenoid to the VNT actuator is 18 " Hg which is fine - and the turbo VNT lever is against the stop.

2) At around 3000rpm at full throttle in 4th heading towards the Saddleworth turnoff on the M62, the boost pressure whips up to 25 psi just after I floor the throttle. I get an overboost/limp in just a few seconds. This is repeatable.

3) At 2600/2700 rpm on light throttle in 5th on the flat, boost pressure stays somewhere between 10 and 15 psi. No sign of an overboost or limp mode. (MAF sensor probably OK)

If my understanding is correct, at full throttle and high engine revs, the EGR valve is supposed to be closed. At 2600 rpm on light throttle, it is supposed to be open. If the 'Troll misbehaves when the EGR is supposed to be closed and behaves when it is supposed to be open, then the odds are that it is stuck open. If that were so then at full throttle and high engine revs, recirculated exhaust gas might cause the inlet manifold pressure to rise uncontrollably (and that is what I'm seeing).

From what you say, it seems that trying to repair/clean/unstick the EGR valve might not be a very workable solution to my problem. I think the options I have are either:

a) Buy a new EGR valve or

b) Block the EGR valve and deal with any consequential increase in boost pressure at lower engine speeds (i.e. cruising at 2600-2700 rpm on the flat) - probably by adjusting the VNT stop screw to delay the opening of the turbo vanes a bit. Hopefully, I will not have to do anything else. I am now seriously considering this option.

Am I making sense here? Does this sound like a plan?

Andrew

 

[andrewk]

Right, folks - MAF sensor.

According to the book of words (dealer workshop manual), the MAF voltages should be:

The readings I get are:

Ignition on: 0.42 v
750 rpm: 1.57 v
1000 rpm: 1.78 v
1500 rpm: 2.07 v
2000 rpm: 2.6 v
2500 rpm: 3.3 v
3000 rpm: 3.6 v
3500 rpm: 3.6 v

This is for the original genuine Nissan MAF installed when the car was new. I have another (non-genuine) and it measures pretty much the same. They were measured between battery earth and the actual MAF terminal (red wire) using an accurate DMM.

Maybe some will disagree - but I don't think there is a lot wrong with those readings.

I'm now running out of options ..... the high revs overboost problem almost has to be a goosed EGR valve - and I *think* it has to be stuck open. If it were, it wouldn't affect cruise boost pressures (as the EGR is supposed to be open) and those are OK - but it could affect the large throttle/high revs pressure (as the EGR is supposed to be closed) which is when the overboost/limp is occuring.

As a temporary measure (might become permananent), blocking the EGR seems to have lots to be said for it. At least it will confirm (or otherwise) that a stuck-open EGR valve is the cause of the high revs/large throttle overboost/limp. Does anyone have a template for a blanking plate (ZD30 with stepper motor EGR valve) .......... so I can make one before I do any dismantling.


Andrew

 

[rayf3262]

If its any use for comparison here's a snapshot of MAF vs RPM for my ZD30:

 

[andrewk]

That is interesting. I don't think those are very different from the readings I took - about 0.2 V higher than mine up to 2000 revs, but otherwise very similar.

How did you produce that?

Andrew