Vehicle details: BMW E46 318i
Engine code: M43TU
Year: 2000
Symptom: No OBD communication,
Non-starter
Author: Andy Cape | Cape Auto Diagnostics, Cape Town

BMW E46 318i | Non-starter

The problem

The vehicle had been driven through a flooded area, and water had entered the block via the air intake. The car cut out with the engine flooded with water. The cold water entered the hot cylinder block, and caused it to crack due to thermal shock. The engine had been replaced, but could not be started. The vehicle was towed to our garage, but we soon discovered that no diagnostic scan could be performed due to no communication to the modules.

The investigation

After a thorough investigation of the relevant fuses, checking supply voltages, ground (earth) signals, immobilizer signals, and the primary ignition voltage, we discovered that:

  • The primary ignition was being supplied with 12 V, but the ECU was not triggering the coils, which resulted in no spark being produced.
  • The relevant fuses were intact.
  • The immobilizer signal was available as soon as the ignition key was turned and therefore was not at fault (see Figure 1).
  • The Crankshaft Sensor did not deliver a valid signal. The power supply for the sensor was only 500 mV once the car was started. (see Figures 2 and 3).
signal from the immobiliser to the ECU

Figure 1: A good signal from the immobiliser to the ECU

Figure 2: Blue = Crank shaft sensor, Red = Hall sensor supply (should be 12 V)

crank shaft signal waveform

Figure 3: The crank shaft signals zoomed

The solution

We checked the wiring of the crank shaft sensor and could not find any faults. The sensor itself had been replaced by the client already. We did notice that without the sensor connected — there was no sufficient power supply to the sensor from the ECU. This led us to believe the ECU must be faulty.

The result

We opened the ECU and found the proof. The power connector was destroyed by extreme overload (see Figure 4). We were confused as to why the fuses did not trip to protect the ECU.

Figure 4: ECU with damanged power connector

Comments

10 comments | Add comment

Magnus
May 26 2012

Nice one!
Why is Stuart Duff searching and reading all this if he already knows how everything works??

DOM (ALFAAUTOMOTIVE)
April 27 2012

NICE ONE ANDY TAKE HEART THERE WILL ALWAYS BE SCEPTIC OR THOSE THAT THINK THEY KNOW MORE AND I HAVE BEEN IN THE MOTOR TRADE FOR OVER 25 YEARS AND SEE THESE TYPES MANY TIMES BEFORE

Martin
December 10 2011

Thanks Andy for your answer. I didn´t know that the signal is constant 12V, when the key is wrong or something else. I have to make more reference signals to learn much more about the diagnostic with Oscilloscope. I even have a question: When I connect my Pico with 2 Channels to two different places (at example channel 1 to battery + with ground on battery - and channel 2 without the ground to battery - only with the red lead to battery +) I get a Signal on Channel 2 12V although I don´t have there a ground for the channel. Is there a problem with my pico? Thanks for every answer.
Best regards

Andy
December 07 2011

I would like to answer the question from Martin (on top). For my understanding is any signal good from EWS to ECU when it looks like the one I captured. When there is no valid signal from key than the EWS OUT signal to the ECU IN is constant 12V. I made me some reference files with transponder and without transponder in key. From transiver module to EWS and from EWS to ECU. Useful if you suspect wiring or module faults.
regards
Andy

dave
December 02 2011

Thanks for taking the time with this one It makes me wonder why some people are still willing to buy flood damaged cars good job

Sheraaz Suliman
December 02 2011

Nice and thorough, good job
we’re based in cape town as well, finding it difficult to contact you, pls drop me a mail with ur contact details .(JavaScript must be enabled to view this email address)
best regards

Dave
December 01 2011

I`ve seen a crank signal like that with a non BMW part that another shop installed.Caused all kinds of issues.

Raul
November 30 2011

It is posible the water got in to the computer?

Mike
November 30 2011

Excellent job tracking this down. I would suspect that the fuses would allow more than enough current for this to happen. Most are fused to at least a 10 to 15 amp fuse. That’s plenty for circuit welding. The ECM wouldn’t have to be flooded, just one of the solenoids it controlled. Good job!

Martin
November 27 2011

Hello, how did you know that the signal from the immobiliser was right? Had you got a comparatively vehicle where the engine was good working and so you record from this the immo signal and compare it with the problem car? I hope you can answer my question thank you sorry for my english i am german smile

Your email address will not be published. Required fields are marked *

Case study: Non-starter