[SOLVED] Yet another power loss and black smoke question (updated)

[OldManBeard]

Hi All,

This is my first post but I've spent a fair bit of time reading the forum beforehand. I know this subject has been covered a lot but after trying all the suggestions I've come across so far I'm still stuck. I've been working with a mechanic on this and he's stumped too.

The vehicle: A 2013 (2012 MY) Thai built Navara (126KW). Total weight around 2700 kg. Done just over 100,000 k. I recently installed a long range tank but made sure to flush it right out a couple of times before installation. Besides, the problem started well after that.

The symptoms:

• Going up a moderately steep hill (25+ degrees), where I need to put the foot down a bit more just to maintain 60kph, after a short while there's a massive power loss, even to the point of spluttering, and enough black smoke that I can see nothing behind me
• Unable so far to reproduce the problem on a flat road with the pedal hard to the floor
• Haven't yet tried with the 2T camper hooked up, as I can't be sure I would be able to get it back up the hill to get home
• No error codes have appeared at any time since this problem started

What the mechanic or I have done/tried so far:

• New air, fuel and even oil filters
• New injectors, as two tested faulty
• Tried cleaning the SCV, without success, so put in a new one
• Intercooler hot hose replaced with hard pipe from turbo to intercooler and cold hose replaced with a new one from Forefront
• Cleaned and checked the function of the MAF and MAP (turbo boost) sensors
• Cleaned and leak tested the intercooler
• Tried an EGR blank but removed it when it made no difference, installed a catch can instead
• Cleaned and tested the EGR
• Emptied the fuel tank and refilled it from a different servo, in another suburb, temporarily running on the contents of a jerry can that I had filled up elsewhere
• Replaced the fuel filter again after changing the fuel
• Temporarily using a DPF that has been hollowed out, so as to eliminate the exhaust restriction

We did a length road test with the mechanic while he had his professional gear hooked up and recorded it all. Neither of us could see a problem. The mechanic sent the recording to other mechanics he knows but they couldn't see anything wrong either. During road tests, I continue to monitor things like intake airflow, boost pressure, etc. but have not been able to spot anything wrong there. I use graphs, rather than gauges, as I reckon variations should be more obvious that way.

At least I now know a bunch of things that are not the cause but I can't think of what else to try, so I'm open to new ideas.

================

UPDATE: My brother just popped in and I roped him in to watch one graph at a time on Torque while I drove tests up and down the same hill, where I can always reproduce the problem. He spotted something that I had somehow managed to miss; when the car starts mucking up the fuel rail pressure does a nose-dive. We did this a few times to be sure we were getting consistent results.

The question now is; what causes a sudden drop in fuel rail pressure when the car is under load up a hill?

 

[GunMentalGrey]

A couple of things that caused power loss and extra smoke for me have been vacuum hose and the fuel primer bulb perishing allowing air to leak into the system.

The primer bulb leak symptom was a really bad flat spot on acceleration, and needing priming every morning.

Vacuum hose leaks were lack of turbo boost, but that was apparent on the boost gauge. Only getting 10psi boost max, but returned to 18-20psi once repaired.

 

[OldManBeard]

Thanks for the reply but my model has the primer built into the fuel filter, rather than the bulb. I've checked that each morning over the last few days and the system has remained fully primed. i.e. I can't push the pump down any further without applying absurd force.

As for the vacuum, like you say that would/should be evident from the turbo boost, which remains what I believe is normal. Depending on which app I use, it peaks at either around 24 PSI (Torque) or 30 PSI (Piston). Don't ask me to explain the discrepancy between those two when nothing else gets changed and the road tests are as near identical as I can make them.

 

[Old.Tony]

While the problem itself is fairly simple, diagnosis sometimes isn't.

At first glance, with a fuel rail pressure diving like it does, you'd have to wonder why there's so much black smoke (partly combusted diesel) when there's no fuel being delivered! However, this is likely to be an ECU-saving-the-engine thing and may not be as simple as cleaning a battery terminal or wiping the seagull poop off the left rear brake light.

On Torque, try monitoring these items (you can actually store this in a CSV file for importing into OpenOffice or Excel and subsequent graphing):

* Throttle position
* Engine RPM
* Vehicle speed
* Engine load
* Fuel rail pressure
* Turbocharger boost pressure

This last one should rise in response to throttle and engine load particularly above 1500rpm and is THE one to keep an eye on.

I suspect it might be the ECU commanding a drop in the fuel rail pressure when the boost pressure gets too high. In my own car, if I adjust my Tillix valve to allow extra boost, I generally go just fine until I put a little load on the car (I'm often towing a 2.5T caravan), then the boost pressure rises over 23psi and the car takes a nose-dive moments later. I've adjusted my Tillix valve to peak boost at around 21psi and I no longer have any problems.

If you're using a stock BCS (Boost Control Solenoid), this may be operating poorly, opening to vacuum too much and pulling the turbocharger vanes open too far and thus producing too much boost.

There's no simple way to "adjust" the ECU to command less boost. The only way to deal with this is to either replace the BCS with a new unit (perhaps $240 from Nissan, definitely cheaper on eBay (example: here) or get a Tillix valve or a Dawes valve (like this one).

So if you find your boost is hitting 24psi just moments before the fuel rail pressure plummets, it's the ECU's way of saying "Cap'n, she kinna take ne more o' this!".

I gave up on replacing the BCS. The Tillix valve works just fine.

 

[Old.Tony]

Oh, and put the DPF back in (for now). A DPF-related fuel problem generally causes white/grey smoke - and LOTS of it. Even while idling. Personal experience talking!

 

[user 10312]

made sure to flush it right out a couple of times before installation. Besides, the problem started well after that.

Ok, from that we can deduce the problem is not fuel tank related.- Maybe. Is there any difference from the tank chock a block full to say a 1/4 of a tank? Pretty wierd it only does it up a hill, there must be something in that. It does sound like it is something boost/turbo related. Like fuel arrives in great proportions but boost pressure is not there yet. These engines tend to do that anyway but this seems way excessive.

Does it continue to splutter and carry on going up the hill or does it come good after a while of wide throttle?

 

[OldManBeard]

While the problem itself is fairly simple, diagnosis sometimes isn't.

It's certainly been anything but simple.

At first glance, with a fuel rail pressure diving like it does, you'd have to wonder why there's so much black smoke (partly combusted diesel) when there's no fuel being delivered!

I really wondered about that too but it appears the smoking starts when I back off and tease the accelerator to find that point where the car can still produce enough power to continue forward, albeit with no guts at all until I get over the crest. I then back off completely very briefly and it's all back to normal.

However, this is likely to be an ECU-saving-the-engine thing and may not be as simple as cleaning a battery terminal or wiping the seagull poop off the left rear brake light.

Quite possibly but I've also started wondering whether the release valve might be sticking a little. I've worked with hydraulics in the past and I've come across similar issues with high-pressure relief valves on machinery.

On Torque, try monitoring these items (you can actually store this in a CSV file for importing into OpenOffice or Excel and subsequent graphing):

* Throttle position
* Engine RPM
* Vehicle speed
* Engine load
* Fuel rail pressure
* Turbocharger boost pressure

This last one should rise in response to throttle and engine load particularly above 1500rpm and is THE one to keep an eye on.

I have monitored those and captured several of them, including turbo boost, in the same file as the fuel rail capture. I left them off the graph to keep it visually simpler. The boost certainly does rise. Nothing weird, just a nice steady and smooth.

I suspect it might be the ECU commanding a drop in the fuel rail pressure when the boost pressure gets too high. In my own car, if I adjust my Tillix valve to allow extra boost, I generally go just fine until I put a little load on the car (I'm often towing a 2.5T caravan), then the boost pressure rises over 23psi and the car takes a nose-dive moments later. I've adjusted my Tillix valve to peak boost at around 21psi and I no longer have any problems.

If you're using a stock BCS (Boost Control Solenoid), this may be operating poorly, opening to vacuum too much and pulling the turbocharger vanes open too far and thus producing too much boost.

There's no simple way to "adjust" the ECU to command less boost. The only way to deal with this is to either replace the BCS with a new unit (perhaps $240 from Nissan, definitely cheaper on eBay (example: here) or get a Tillix valve or a Dawes valve (like this one).

I'll file this suggestion for now. I would very much prefer to find the problem than hide it like that. The issue didn't exist in the past, so it should be possible to get it back to that condition.

So if you find your boost is hitting 24psi just moments before the fuel rail pressure plummets, it's the ECU's way of saying "Cap'n, she kinna take ne more o' this!".

I should mention that if I put my power chip back in-circuit I can easily get the boost pressure up there on a flat road without it triggering the power loss and smoke. I disconnected it while trying to resolve the problem, as I wanted to eliminate as many variables as possible.

 

[OldManBeard]

Ok, from that we can deduce the problem is not fuel tank related.- Maybe. Is there any difference from the tank chock a block full to say a 1/4 of a tank?

I had similar thoughts but have tested with everything from a near empty to a completely full tank. Makes no difference, except it triggers sightly earlier on the same hill with a full tank but I believe that's simply due to the extra weight, as it's a 140 litre tank. It was the same when I had my brother in the car; the problem started even a bit earlier on the hill.

Pretty wierd it only does it up a hill, there must be something in that. It does sound like it is something boost/turbo related. Like fuel arrives in great proportions but boost pressure is not there yet. These engines tend to do that anyway but this seems way excessive.

This is the bit that puzzles me the most. Why have I not been able to trigger the same problem on the flat? Admittedly flat road testing must be a bit limited due to the lack of somewhere I can really push the car as hard as I'd like. There's a distinct lack of roads I can do this kind of testing on that don't also have a lot of traffic and/or speed cameras or cops. I don't want to be dangerous to myself or others, nor do I want to get booked.

I might get more meaningful results if I was game to hook up the camper and run tests but I don't want to get into the situation where I can't tow it home again, as I live near the top of the hill. I may need to see if my neighbour would be willing to help in this regard, as he has the same coupling and trailer connections. If so, I'll post the results.

Does it continue to splutter and carry on going up the hill or does it come good after a while of wide throttle?

That's a very interesting question, to which I think I have to answer no. When the problem starts I have to back off and then tease the accelerator to find a point where I get enough power to get up the rest of the hill, at a greatly reduced speed and billowing black smoke. If, when I reach the flat bit at the top, I put the foot down too much it's still powerless. I have to briefly back off completely for half a second or so before it will come back to normal. I should probably have mentioned this in my original post.

 

[[email protected]]

probably irrelevant to do a forced regen drive you car 2nd gear 4000rpm for 20 minuits

 

[OldManBeard]

probably irrelevant to do a forced regen drive you car 2nd gear 4000rpm for 20 minuits

Yes, not really relevant with a hollowed-out DPF.

 

[Old.Tony]

To be clear on this, once the issue triggers, does the boost level fall along with the fuel rail pressure? What's the engine load % at this time (and additionally, what is the load % in the moments leading up to it)?

Just an additional thing: what's the coolant temperature at this time (reported by the ECU, not by an external gauge)?

If it's not the ECU directly intervening, what could it be? An electrical fault (dirty connection etc) would not be this specific, surely? Past experience tells us that faults generally target random things.

With the chip, is it doing a full intercept of the injector control and fuel rail pressure, or is it a simple modifier of the fuel rail pressure sensor (like a No Limits chip)? Which chip is it? There's a chance that the chip is masking a problem that the ECU thinks is dangerous to the motor (whether it is or not). Diagnosis, as you say, is needed without the chip, but the fact that the chip hides the problem might point to where the problem lies if we know what the chip is doing.

 

[OldManBeard]

To be clear on this, once the issue triggers, does the boost level fall along with the fuel rail pressure? What's the engine load % at this time (and additionally, what is the load % in the moments leading up to it)?

Just an additional thing: what's the coolant temperature at this time (reported by the ECU, not by an external gauge)?

If it's not the ECU directly intervening, what could it be? An electrical fault (dirty connection etc) would not be this specific, surely? Past experience tells us that faults generally target random things.

With the chip, is it doing a full intercept of the injector control and fuel rail pressure, or is it a simple modifier of the fuel rail pressure sensor (like a No Limits chip)? Which chip is it? There's a chance that the chip is masking a problem that the ECU thinks is dangerous to the motor (whether it is or not). Diagnosis, as you say, is needed without the chip, but the fact that the chip hides the problem might point to where the problem lies if we know what the chip is doing.

Thanks Old.Tony, I appreciate you sticking with this one so far and offering your various ideas.

I'm going to do more testing and I'll use a shorter capture interval to get more detail. So far I've not detected any other reading doing anything out of the ordinary when the problem starts. I'll post more when I have finer-grained results.

If it was an electrical fault it would have been evident under other circumstances. I wish it was electric but this fault is completely predictable and reproducible. Having worked with electrics and electronics for more than 50 years I'm confident I could have resolved an electrical problem weeks ago, even an intermittent one.

The DTI chip is dual channel, controlling fuel rail pressure and turbo boost. It doesn't control the injectors themselves. I've been running it for two years, using all settings from maximum economy to maximum power. Even when pushing the car hard in mountain country with the camper hooked up, I've experienced no problems when using the chip. It's disconnected at the moment because I always, where feasible, remove add-ons when diagnosing problems. It's the simple rule of eliminating what I can and then just working with what's left.

 

[Macs550]

Hi, First time posting from me but been following for some years now.

The symptom you have indicates excessive fueling but you mention your fuel pressure is dropping. Can you install a temporary high pressure guage near the fuel pressure sensor and see if the pressure is actually dropping? To me I would consider these two scenarios. 1 The ecu is adding more rail pressure to compensate for the loss of pressure signal it is receiving, creating over fueling. 2 Rail pressure is actually dropping because fuel is being dumped or leaked into the combustion somewhere.

 

[redkite]

The vehicle: A 2013 (2012 MY) Thai built Navara (126KW). Total weight around 2700 kg. Done just over 100,000 k. I recently installed a long range tank but made sure to flush it right out a couple of times before installation. Besides, the problem started well after that.

The symptoms:

• Going up a moderately steep hill (25+ degrees), where I need to put the foot down a bit more just to maintain 60kph, after a short while there's a massive power loss, even to the point of spluttering, and enough black smoke that I can see nothing behind me

How sure are you about the steepnees of the hill?

Physics tells that driving 2700kg of vehicle up a 25 degree hill at 60km/h requires at least 186kW to counteract gravity, let alone wind and rolling resistance.

Obviously, that ain't gonna work with a 126kW engine. So I wonder whether you are not overstressing the engine, causing the RPM to drop, and the ECU cutting out to protect the engine.

A simple test would be to find out what is the maximum speed you can climb that hill. And check the steepness.

 

[OldManBeard]

Hi, First time posting from me but been following for some years now.

The symptom you have indicates excessive fueling but you mention your fuel pressure is dropping. Can you install a temporary high pressure guage near the fuel pressure sensor and see if the pressure is actually dropping? To me I would consider these two scenarios. 1 The ecu is adding more rail pressure to compensate for the loss of pressure signal it is receiving, creating over fueling. 2 Rail pressure is actually dropping because fuel is being dumped or leaked into the combustion somewhere.

First posting or not, I do appreciate you, and all other responders, taking the time and offering ideas. Thank you.

Adding a gauge will probably require installing an appropriate fitting into the fuel rail, unless someone knows of a test point already available for this. While that's certainly not out of the question, it is a bit of work as I'de wouldn't want to be fitting a test point while the rail is on the engine. I don't have any high-pressure gauges and I'm not even sure where I could source one around here. I'm long out of touch with that sort of thing. This might be something I need to go to a diesel specialist for, who will have the needed equipment. I'm trying to avoid that if possible for two reasons: Cost and the fact that, this being my first diesel, I'm trying to learn as much as I can about it because the more I learn, the more chance I can keep it going when I'm in the middle of nowhere.

In regards to the smoking, you may well be correct because that appears to happen after I've backed off and try to find a point where I still get some power, so yes, it ould be the ECU compensating for that sudden drop in pressure. More testing, which I wasn't able to do today, might shed more light on that.

I don't believe fuel is dumping into the combustion chamber. The only normal way for it to get there is through the injectors and they've just been replaced with a new set. Any other way in would have to be due to some pretty catastrophic failure in the head, which would mean I would have really major problems all the time, not just under specific conditions. Same thing for an injector sticking open, although I acknowledge that could potentially happen even with a brand new one.

I definitely need, and intend, to collect more info, with a much shorter time between sensor scans, so that I get a better resolution in the output. More information must improve the chance of nailing this. Even if I do have to take it to a professional, I'll share what I have and it might shorten the time they need to spend diagnosing it.

 

[Old.Tony]

I had a silly thought about this during the day, when I recalled what had happened to a small electric motor's shaft that was somewhat loose in its bearings. As the motor sped up, the shaft started to wobble, and it reached a point where the engine speed fell back dramatically regardless of how much power was applied to the motor (and the motor overheated as a result).

So I then thought about your turbocharger.

See, the black smoke thing is going to be caused by "too much fuel to be completely burned". Now "too much fuel" is a distinctively relative thing - for the engine to burn without too much soot, you need combustion of a certain amount of fuel with a certain amount of air and creating a certain temperature. A little more fuel has exactly the same result as a little less air - black smoke.

There are three primary things that affect this.

1) Injector commands. Let's assume for the moment that the ECU is getting this right, but it bases the decision about injector operation on the fuel rail pressure, boost pressure, MAFS flow rate and MAFS temperature (as well as coolant temperature, engine load, throttle position and in an auto, some feedback from the TCM).

2) Turbocharger output. It's happened a bit (and to me) that the turbocharger has let go (wobbly shaft) and at a certain level of demand the turbo's shaft starts to harmonically wobble and dramatically slow down. This reduces the air flow rate (removing air = black smoke).

3) EGR. This nasty little blighter is designed to prevent the combustion temperature from getting high enough to produce oxides of nitrogen. The sad reality is that EGR also dampens performance. I know you've tried it with and without an EGR block but I have to include it in the list of things that cause black smoke.

I would ask you to take a look at the turbocharger's shaft and see if there's any lateral movement. It doesn't seem like any of the other likely candidates are ... well, likely!

 

[OldManBeard]

Snip... snip...
I would ask you to take a look at the turbocharger's shaft and see if there's any lateral movement. It doesn't seem like any of the other likely candidates are ... well, likely!

Thanks for the idea. Never discount any idea until you're sure it won't/can't work.

Although not mentioned in my post, and not knowing the mechanic had also already done it, I did indeed check the turbo. I used a dial indicator, as that'll detect slop the hand can't even feel. Just normal and correct clearance. Regardless, I do believe any performance loss due to shaft wobble should be readily apparent in the boost pressure readings.

Regarding the likelihood of any of these ideas explaining the issue, the fact remains that we've pretty much ruled out all the usual candidates and must therefore be prepared to consider just about any possibility (although I will draw the line at checking the tyre pressures ).

 

[Macs550]

If it was mine I'd start by looking at the fuel pressure sensor. When you look at your graph and if it happens at the same pressure every time I'd say you could replicate this more with your camper on (more load). My guess is you are putting the engine under more sustained load up the hill than you can achieve accelerating on the flat. Once the rail pressure hits a certain (un-specified) point it's like the sensor crashes until you get it back operating in its sweet more common spot. When the sensor crashes the ecu is still trying to push fuel in to make the required pressure. It could also be the pressure relief doing the same thing valve but I'd say its the sensor due to the fact that the fuel is getting to the engine (black smoke). When you inspected the first scv was it a bit rusty or scratched? If so those contaminants may be affecting the pressure sensor also.

 

[OldManBeard]

If it was mine I'd start by looking at the fuel pressure sensor. When you look at your graph and if it happens at the same pressure every time I'd say you could replicate this more with your camper on (more load). My guess is you are putting the engine under more sustained load up the hill than you can achieve accelerating on the flat. Once the rail pressure hits a certain (un-specified) point it's like the sensor crashes until you get it back operating in its sweet more common spot. When the sensor crashes the ecu is still trying to push fuel in to make the required pressure. It could also be the pressure relief doing the same thing valve but I'd say its the sensor due to the fact that the fuel is getting to the engine (black smoke). When you inspected the first scv was it a bit rusty or scratched? If so those contaminants may be affecting the pressure sensor also.

While you may well be thinking in the correct area, I don't believe it's the sensor. A fault there could definitely result in a sudden low reading but that shouldn't be causing the immediate and dramatic loss of power that I'm getting. I'm more inclined to suspect the pressure relief valve could possibly be sticking a bit when it opens. I have read where some people suggest the two are actually the same device but they're separate and on opposite ends of the fuel rail. I'd be very surprised if the two are combined into one unit on any engine.

ECU doesn't know about the relief valve opening (it's a simple spring valve with no sensor), other than detecting the pressure drop. In such a situation, what you said about the ECU continuing to push the fuel in makes complete sense. Anyway, to the ECU it would appear exactly the same, whether it's the sensor or the relief valve. The only difference is whether or not there's a brief shortage of fuel, followed by a surge, or just the surge on its own. I, and presumably the ECU, would expect the normal pressure drop due to the valve opening to be quite minimal. The valve's purpose is to let out just enough fuel to limit the maximum pressure in the rail, not to dump it all. Hence my suspicion.

I may pull the valve out, dismantle and inspect it. Perhaps I should shim it to increase the opening pressure, just for testing. That should hopefully prove the point one way or the other. I have to either prove or eliminate both sensor and valve from my list of possible causes.

You are of course correct in that I can't put the engine under the same sustained load on the flat, at least not without a much heavier load. As mentioned elsewhere, I haven't been game to hook up the camper for the road tests because I live near the top of the hill and I'm not at all confident the car would currently be able to get it home. I may yet be able to get my neighbour to help with this, in which case I expect we'll be able to learn more about what's going on.

 

[OldManBeard]

UPDATE: My brother just popped in and I roped him in to watch one graph at a time on Torque while I drove tests up and down the same hill, where I can always reproduce the problem. He spotted something that I had somehow managed to miss; when the car starts mucking up the fuel rail pressure does a nose-dive. We did this a few times to be sure we were getting consistent results.
View attachment 34897
The question now is; what causes a sudden drop in fuel rail pressure when the car is under load up a hill?

Please disregard the quoted update and everything I wrote about the fuel rail pressure. I was completely misled as a result of logging with too coarse granularity. i.e. The sampling interval of 10 seconds was far too long to tell the real story. Logging at one second intervals reveals there was no problem there after all. I've done a number of additional test runs under various conditions, logging all the sensors that Torque was actually able to get readings from, and nothing is obvious at all. Far from it, if those readings were all I had to go by I'd have to conclude there is no problem, even though there clearly is.

I'm attaching today's in case someone wants to have a look at them. Although I didn't experience the really bit power loss and smoke belching, due to the traffic conditions slowing me down too much to get to that point, I did have a number of times where there was a distinct flat spot. I can't recognise those points in the logs though.

Had this been a petrol and/or LPG engine I would have been at home, as I've been keeping those running smoothly for five decades. However, this is my first diesel and my first turbo. I'm working with no experience and far too little information, so I'm going to have to admit defeat. I think I'll be contacting Diesel Tech later this week to see if they can help.

Thanks again for the help you've each given me.