X trail hard to start when cold. Out of ideas :(

[Lee]

This is the same X trail I posted about a while ago, 2.2 Dci.

A lot of work later (new chains,valves/head work,new turbo) and it now
*runs perfectly* but is difficult to start when cold.
Hard to start when cold seems to be a common complaint with these...

Once it has fired up, it behaves impeccably, fast idles as it should
whilst warming up and no trace of missing or unevenness.

This car has had a new battery, glow plugs and main ground strap and
have also tried the starter motor from a known good vehicle.
Injectors have been tested by a diesel specialist for leakage, fuel
delivery and response times.
Alternator also tests ok.
Obviously it has had the fuel filter changed and the pump seems to
deliver the required rail pressure. (Target and actual FRP agree in
diagnostics graphing)

If you do the usual trick of waiting for the glow light to go out and
then repeat and wait 20 secs (by design the plugs are powered for 20
secs after light goes out) it will start after about 4 -5 seconds of
cranking.
(This engine also lights the glow plugs during cranking, which it is doing.)
If you don't do this and try to start it as soon as the light goes out,
it will crank for 30 secs or more before starting. Despite the glows
lighting during cranking. Sometimes at this point it will throw spurious
codes, usually crankshaft sensor - although it still shows the rpm and
the waveform is still correct,(measured at ECM terminals using a 'scope)
or EGR valve (replaced with known good) and occasionally various moaning
about loss of CAN communications to various modules.
I'm guessing this is because the ECM is upset about the voltage drooping
from extended cranking.

The battery voltage is 12.6, dropping to 11.5 with the glows on and down
to 9v when cranking.
Now, while this sounds a little low to me, an otherwise identical model
exhibits the same behaviour - except that one starts within a second,
hot or cold and without needing the 'double glow trick'.

Have tried as many diagnostics comparisons to the other vehicle as I can
think of, cranking rpm, fuel rail pressure, target pressure, temp
sensors, crank and cam sensors, injector waveforms, air mass readings
and there are no obvious differences.
For the purposes of clarification the compressions are 320PSi, *but* so
is the comparison vehicle...

Out of ideas now :(

[Mrcheerful]

If it hits 9v under load: there is your problem, the battery is poorly.

[Chris Whelan]

On Tue, 04 Feb 2014 16:30:47 +0000, Lee wrote:

[...]

> This car has had a new battery, glow plugs and main ground strap

OEM glow plugs, not pattern?

Chris

--
Remove prejudice to reply.

[Lee]

On 04/02/2014 16:35, Mrcheerful wrote:

> If it hits 9v under load: there is your problem, the battery is poorly.

Well it is a generic 'no name' from a certain low cost parts suppliers...

I would have swapped it out already but it isn't my car and the owner
isn't convinced.
I don't have a spare battery large enough for a proper test so I've left
it up to him to decide if he wants to source a decent battery :)


Lee

[Lee]

On 04/02/2014 16:59, Chris Whelan wrote:
> On Tue, 04 Feb 2014 16:30:47 +0000, Lee wrote:
>
> [...]
>
>> This car has had a new battery, glow plugs and main ground strap
>
> OEM glow plugs, not pattern?
>
> Chris
>

ADL/Blueprint, sadly the OEM ones have been disposed of.

Lee

[Chris Whelan]

Don't know about the X Trail, but some diesels are really fussy about
their glow plugs, and are slow to start when cold without OEM ones.

With everything else that has been done, replacing then with genuine
might be the cheapest next step?

[Lee]

The OEM Nissan ones are could hardly be described as cheap, but point taken.

Lee

[The Revd]

On Tue, 04 Feb 2014 17:28:20 +0000, Lee <cyber...@ukonline.net>
wrote:

>On 04/02/2014 16:35, Mrcheerful wrote:
>
>> If it hits 9v under load: there is your problem, the battery is poorly.
>
>Well it is a generic 'no name' from a certain low cost parts suppliers...

Ah, a piece of shit, in other words!

[The Revd]

On Tue, 4 Feb 2014 21:50:59 +0100, The Peeler
<finish...@themoronicRevd.invalid> wrote:

>On Tue, 04 Feb 2014 12:43:32 -0800, The Rectum, the resident psychopath of
>sci and scj and Usenet's famous sexual cripple, FAKING his time zone again,
>farted:

>
>
>>>> If it hits 9v under load: there is your problem, the battery is poorly.
>>>
>>>Well it is a generic 'no name' from a certain low cost parts suppliers...
>>
>> Ah, a piece of shit, in other words!
>

>Like me, you mean?

Yes anus! LOLK!

>I'm piece-of-skata Grik psycho! <BG>

You, ARE anus! <GB>

[spam>@btinternet.com RHJ Croxton & Son (Farms) Ltd.]

"Lee" wrote in message news:blciu4...@mid.individual.net...

What is cold cranking RPM? Needs to be 280rpm at least.

Do you have crank and cam sync within afew crank cycles on your scan tool?

I would like to know what rail pressure target v actual is during
cranking...

Glow plugs should be irrelevant til its well below zero.

Tim..


Tim.

[Duncan Wood]

Quick diagnosis is does it start instantly if you jump start it?

[Mrcheerful]

or give it a tow start (assuming manual)

[Lee]

On 07/02/2014 21:45, Mrcheerful wrote:

>>
>> Quick diagnosis is does it start instantly if you jump start it?
>
> or give it a tow start (assuming manual)

Tow start is instant, whereas jump start exhibits the same issue. I
suspect there is more going on than a simple voltage drop.

[Duncan Wood]

What's the voltage at the starter terminals when cranking?

[Lee]

On 07/02/2014 17:06, RHJ Croxton & Son (Farms) Ltd. <the.farm. wrote:

<snip>

>
>
> What is cold cranking RPM? Needs to be 280rpm at least.

That could be a clue as it's about 200rpm

>
> Do you have crank and cam sync within afew crank cycles on your scan tool?

Don't remember seeing that entry on the scan tool, but
injector waveform is present within a couple of turns so assuming cam
and crank sync is obtained and rail pressure is attained, otherwise it
wouldn't drive the injectors?...
Admittedly I only checked for the presence of the waveform, I must
remember to check the drive amplitude and injector current when I check
it next.

Cam and Crank sensors are in sync when checked with a dual channel
scope, when cranking and when running. We were *extremely* careful when
replacing the chains :)

>
> I would like to know what rail pressure target v actual is during
> cranking...

I'll have to re-check the target pressure when cranking as for some
reason I haven't noted that down. 'Actual' I have written down as 25-30MPa

Once it's started, and warmed up, target is 24/26 and actual is 24/26MPa
That's actual as reported by the ECM, don't have a suitable gauge and
adapter to measure the 'real' pressure.

>
> Glow plugs should be irrelevant til its well below zero.

Allow the manual says otherwise, I tend to agree and think it's a red
herring and that something else is going on.

>
> Tim..
>
>
> Tim.
>

[Lee]

Should have access to a high current clamp sensor for the weekend so I
was going to check the starter current and recheck the terminal voltage,
but last time I noted it, it was 9v. I also measured 9v at the battery
terminals (the battery itself not the clamps, but they were the same
anyway), but I didn't measure them simultaneously, which I will do, in
case there is an issue with the positive cable. Ground lead has been
changed already.

Have already tried another battery and get the same thing, 9v at the
battery (both itself and and at the clamps) when cranking - either
that's what it should be or both starters are suspect....

[Graham J]

[snip]

> Glow plugs should be irrelevant til its well below zero.
>

I have a Vectra 2 litre diesel - at plus 10 degrees if I disconnect the
glow plugs starting is difficult - it cranks for several seconds then
coughs and splutters into life, and runs rough for many seconds.

By contrast with the glow plugs connected starting at +10 is quick - it
starts after less than one complete turn of the crankshaft and
immediately runs smoothly.

So I reckon glow plugs are essential at even quite warm temperatures ...

--
Graham J

[Mrcheerful]

and some diesels don't have glow plugs (or equivalent) at all, older
Transit engines for example. So much must depend on engine design, and
of course nowadays, emissions control.

[Duncan Wood]

If the starters sticky then two batteries in parallel will often spin it
fast enough to start immediatly, & you can bypass the engine earth lead
with the jump lead straight to the block.

[Tim]

From what you have said now about cranking voltage dropping to 9v;

a) this is too low- where are you measuring it? At battery or at starter?
b) I would be surprised if the ECU is still firing the injectors at 9v.

Test and report back.

Have you missed an earth strap off the engine block?

[Lee]

On 07/02/2014 17:06, RHJ Croxton & Son (Farms) Ltd. <the.farm. wrote:

>

> I would like to know what rail pressure target v actual is during
> cranking...
>
> Glow plugs should be irrelevant til its well below zero.
>

Right, it's been a while since I've been able to get back to this but
I've just checked now..

During cranking the *target* rail pressure is dancing about all over the
place between 20 and 50Mpa and the actual rail pressure is following it.
During this the crank/cam/throttle/air mass and temp signals are all as
expected.
Once it starts, then the target rail pressure stabilises and the actual
rail pressure along with it.

Thinking suspicious thoughts toward the pump, or at least the SCV valve,
but if it is, why would it bump so easily and run perfectly? Doesn't
make sense...

[Mrcheerful]

does this have an electric lift pump? autodata says it does. if so,
does that actually run as it should? and give 3.5 bar delivery pressure.

[Lee]

Main earth strap has already been changed, it runs from battery to
chassis and then onto the engine..

9v at the battery terminals themselves (and the same across the clamps,
which are clean and tight) and 8.5v at the ECM main power terminals when
cranking, this is with yet another new battery. (ECM voltage is 11.5
when not cranking and 14.4 when the engine is running - measured with a
(trusted) DMM at the ECM power relay adjacent to the ECM)
Starter has been examined by a local specialist and pronounced fit, not
surprised since both starters do exactly the same thing...
Engine doesn't feel excessively 'tight' when turning by hand, (feels the
same as the other one) and when it cranks it is at a fairly even speed.

It seems completely obvious that it is the starter, but the one which
works perfectly on the comparison vehicle still exhibits the same
symptoms on this one.

[Tim]

What is Battery or ECU feed voltage doing during cranking, and also -ve
verses battery -ve on the block and at ecu? Is the ECU ground good?
Obviously the current draw of the starter is upsetting something.

I would be 'scoping the battery feed under cranking to see if a dodgy
starter is spiking the ecu with EMF, or the rail pressure sensor signal
line.

IME, a problem with the suction valves creates a bad rattle, lack of power
and copious blue smoke, rather than a no-start.

Tim..



"Lee" <cyber...@ukonline.net> wrote in message
news:bms206...@mid.individual.net...

[Lee]

Autodata says it has, but it hasn't. The Nissan service manual shows an
external "fuel transport pump" which pumps fuel between the two U-shaped
sections of the tank. It's not easy to get at but I guess I should
eliminate that as well, even though it should in theory have no bearing
on the problem.

[Lee]

On 22/02/2014 17:13, Tim <flexie23 wrote:
> What is Battery or ECU feed voltage doing during cranking, and also -ve
> verses battery -ve on the block and at ecu? Is the ECU ground good?
> Obviously the current draw of the starter is upsetting something.
>
> I would be 'scoping the battery feed under cranking to see if a dodgy
> starter is spiking the ecu with EMF, or the rail pressure sensor signal
> line.
>
> IME, a problem with the suction valves creates a bad rattle, lack of
> power and copious blue smoke, rather than a no-start.
>
> Tim..

The voltage at the ECM power relay is 8.5v when cranking,(9v at batt)
11.5 (11.9 at batt) in Ign on, engine off and 14.4v (14.4 at batt) with
engine running. About the same as the ECM reports over ODB.

Not sure where it's dropping that 0.5v, if you run a thick patch cable
from the battery to the relay then the 0.5v drop disappears but it
doesn't make any difference to the behaviour.
ECM ground is good, it's bolted directly to the chassis anyway.
DC waveform at the ECM is a little spikey when cranking, but it doesn't
look any worse to me than other vehicles I've checked.
Maybe this one is more sensitive - I'll have to think about decoupling
it with something suitable.

[Mrcheerful]

why not leave the onboard battery to run the ecu and connect a spare
battery directly to run the starter, if it is then starting and running
ok then you will know it is a power problem

[Lee]

On 22/02/2014 19:34, Mrcheerful wrote:

>
> why not leave the onboard battery to run the ecu and connect a spare
> battery directly to run the starter, if it is then starting and running
> ok then you will know it is a power problem

Exactly what I was going to try, later :)

Lee

[Tim]

"Mrcheerful" <g.odon...@yahoo.co.uk> wrote in message
news:8Z6Ou.26910$cK2....@fx11.am4...

I was about to say the same thing. 8.5v at the ecu is just not enough. It
needs to generate ~80v to fire the injectors...

Power the ECU directly from a separate battery via some good quality jump
leads / patch cable and try again.

Tim..

[Mmeldrum29]

replying to Lee, Mmeldrum29 wrote:
Hi. Did u get to the bottom of this problem. I imagine having exact same
symptoms with mine

--
for full context, visit http://www.motorsforum.com/maintenance-uk/punto-annoyance-47689-.htm

[Lee]

On 05/05/2017 01:18, Mmeldrum29 wrote:
> replying to Lee, Mmeldrum29 wrote:
> Hi. Did u get to the bottom of this problem. I imagine having exact same
> symptoms with mine
>

Yes, but the actual final cause will certainly not be relevant to your
issue. (it was an incorrect part than had been previously fitted)

Obviously you need to have checked all all the usual suspects, ie
injectors/temp sensors/maf/ crank and cam sensors etc

Wear in the chain can cause poor cold starting (as I found out)

What I found useful was viewing the timing correlation of the crankshaft
and camshaft sensors on a 2 channel oscilloscope.
The correlation varied too much between hot and cold and pointed to the
(new) chain being badly out of spec.

[alan_m]

On 05/05/2017 01:18, Mmeldrum29 wrote:

> replying to Lee, Mmeldrum29 wrote:
> Hi. Did u get to the bottom of this problem. I imagine having exact same
> symptoms with mine
>

What problem? The crap web site you are using links your question into
a post about gear changing problem on a fiat punto that was posted 12
years ago!

--
mailto: news {at} admac {dot] myzen {dot} co {dot} uk