LDV extended crank starting

Everyone loves a happy ending. Share your diagnostic success stories here.
Post Reply
MarkStammers
Newbie
Posts: 2
Joined: Sat Jun 21, 2014 7:42 pm

LDV extended crank starting

Post by MarkStammers »

LDV Maxus 2008: Engine code: R2516L VM39/40B 2.5CRD

I was called out to this vehicle, as it was causing the garage a lot of frustration.

The vehicle drives perfectly ok, once started, it just struggles to start especially from cold.

The garage had replaced the glow plugs and the relay to no avail. They did manage to pull a code from the injection ECU, using a generic EOBD scan tool. A P0335 crankshaft sensor malfunction. They replaced the sensor and, whilst in the area, checked and cleaned the target wheel. Still no change.

I was able to communicate with the injection ECU, using Bosch KTS, and confirmed there were DTC's for crankshaft sensor, amongst other codes. However the interpretation of the codes present were not reliable. I could see live data and selected PID's for engine RPM, fuel rail pressure, and cam/crank synchronization. I gave the engine a start command and was able to witness all PID's reacting, including engine RPM and fuel rail pressure. Unfortunatley, I couldn't rely on the cam/crank sync PID, as it displayed a numeric code status instead of yes/no status. It did however prove that the crankshaft sensor was working, because of RPM.

So I set up the scope to monitor some key information. Crankshaft sensor signal, Camshaft sensor signal, fuel rail pressure and battery current. At this time I was particularly interested in the starter motor current, as this is easily overlooked and can cause just such symptoms.
Starter current.jpg
The battery current checked for relative compressions and starter motor draw and as you can see proved fine. The engine clearly failed to start. I then took another capture, only this time replacing the battery current with injector no.1 current. This showed me that the injectors were not being switched on, either because there was a fault with the voltage supply to them or because the ECU was inhibiting them.
Starter up sequence with injector no start.jpg
I had noted that the crankshaft sensor fault, which I had cleared previously, returned. This made me suspicious of the valve timing, as the fuel drop off and no injector operation, is a fairly normal strategy for a cam/crank sync issue.

I checked my library files and found no entries of such a vehicle, so I checked out the Pico library. 'Result'! I found two samples on file and they both matched each other. That gave me confidence in their accuracy. They didn't have an engine code against them but they did match my sample exactly! Oh dear.. That made me think; the valve timing appears to be correct, but the ECU keeps complaining of crankshaft sensor and that works.

I took another sample, during which time the engine started after a shorter crank time. I captured the sample and compared the crankshaft signal to the known good library sample.
Starter up sequence with injector and start.jpg
After I took a closer look at the library captures, concentrating on the crankshaft signal, I noticed my capture clearly had a lower amplitude to the sensor. I checked how the library sample scope probes had been connected and noted they had used a common ground connection, as had I. (That's a really useful notation, so thanks to the provider of the waveforms). So I took a closer look at the sensor and noticed how it didn't seem to sit flush in it's mounting, even although it was tightened down securely.

I removed the sensor and inspected the mounting surface. I attempted to clean it up and refit the sensor. Unfortunately I don't carry the necessary engineering tools and so was a little restricted in what I could acheive. I found that the sensor appeared to sit flush if I rotated it away from it's fixing position. So I left it in the rotated position and re-captured a start sequence.
Starter up sequence with injector and start after fix.jpg
That did it! Instant start and the crankshaft signal was much better with a stronger amplitude.

I left the garage with the job of perfecting the mounting surface.

User avatar
STC
Banned
Posts: 790
Joined: Sun Sep 13, 2015 2:10 am

Re: LDV extended crank starting

Post by STC »

Good catch there Mark :) With the ECU against you - Showing RPM
I noticed my capture clearly had a lower amplitude to the sensor.
We are often taught and see that circa 600mV is enough to switch a FET and consequently the ECU can see it. In your case, RPM is visible and correct in Actual Values.

This was for sure a curved ball thrown at you - P0335 and RPM in live data. They wont go to heaven ?

I speak with the benefit of hindsight having had this dilemma with a W211 E270 Mercedes. This had an aftermarket Crank Sensor fitted, filing the mounting base so it sat closer to the reluctor equalled a start every time.
Similar scenario with an Audi TT 225bhp BAM.

I will add that due to the non existence of P0336 I did not chase timing on either vehicle, I did use brake cleaner to:
1. Confirm to some extent Mechanical Integrity of the Lump
2. Increase Amplitude / Magnitude of CKP signal - Looking at the scope to record at what point of FRP and RPM it unleashed the vitamins and minerals to make that engine spin autonomously.

Once again - Good Catch. Some see P0335 / 336 and fit a new crankshaft or ECU.

Not forgetting the resource - Pico Library - Thank You PICO for that - it is a godsend :)

MarkStammers
Newbie
Posts: 2
Joined: Sat Jun 21, 2014 7:42 pm

Re: LDV extended crank starting

Post by MarkStammers »

Yes, it is always a problem when an input is just a little off the mark. Sometimes we get caught out by not paying attention to the details of a signal and only looking at the presence of a signal. I guess like most things, we get complacent, because most of the time the sensor failure will be obvious.
I guess that's why it's important to record details of all captures and how they are obtained. Easier said than done!

Regards

Dave Hill
TwoWaves
TwoWaves
Posts: 63
Joined: Fri Jun 22, 2007 8:15 am

Re: LDV extended crank starting

Post by Dave Hill »

Nice work mark

A good job & a very nice write up too.

Worthy of a place in the Case Studies section I reckon! :D

Post Reply