A shout for help again

Post by **Steve Smith** » Tue Jan 26, 2021 11:53 am

Vehicle details:
2005 Vauxhall Vectra-C Auto
Engine code Z32SE (V6-Petrol)

We have been diagnosing a failure of the above vehicle to crank when turning the ignition switch/lock to the cranking position.

To cut a long story short, the starter motor energizing wire from the starter relay (pin 87) is not stable and pulsing when cranking the vehicle. (I will explain why later)

The symptom therefore is the starter motor repeatedly trying to engage and sounds very similar to a discharged battery

The video below displays the symptom perfectly:

SYMPTOM.mp4: Symptom; (1.29 MiB) Downloaded 945 times

The 12 V battery is fully charged & stable with the above symptom present. Adding battery support will occasionally allow the vehicle to crank as normal

The voltage drop on the starter motor supply and ground are also correct
The above symptom is not related to battery supply or indeed chassis grounding

Fault codes indicate numerous ECU’s reporting a loss of communication with the engine ECU

With the symptom present (Starter motor engaging/disengaging when cranking) the engine MIL also illuminates and extinguishes at the same frequency

With the scan tool attached and communicating with the engine ECU, when attempting to crank (with the symptom present) communication between the scan tool and engine ECU fails. (i.e., communication drops out)

The cranking circuit relies upon the engine ECU to ground pin 86 of the starter motor relay, the image below courtesy of ALLDATA will help clarify the circuit

: Image 1

To explain the diagram above and the missing circuits, the ignition switch supplies the gear position switch with a 12 V supply when held in the cranking position.

If the gear position switch is in either the Park or Neutral position then the supply from the ignition switch is allowed to pass onto pin 85 of the starter motor relay.

The starter relay (pin 86) is then grounded via the engine ECU (so closing the relay contacts) and cranking can commence. Here is where we have the issue as the engine ECU is not holding the starter relay pin 86 to ground, but pulsing the ground instead.

The result therefore is the rapid energizing and deenergizing of the starter relay and of course, the engaging/disengaging of the starter motor as a direct consequence

The waveform below captures this event

: Image 2

The image above demonstrates how the engine ECU is only grounding the starter relay for a very short period, see channel E. (It appears to be pulsing the starter motor relay)

Note: Channel E “channel label” indicates “starter relay pin 85” yet the wiring diagram indicates pin 86, this is because the wiring diagram pin configuration for the starter relay differs to the pin configuration embossed into the engine bay fuse box assembly (Another variable to tackle)

What I find very interesting is the behavior of the in-rush current into the starter motor. “Current” momentarily flows into the starter motor at the point where the starter relay is de-energized?

Moving on with additional testing:

Given the engine ECU is bolted to the engine (and in turn acquires an earth path via the mounting bolts) a volt drop test (ground line) was carried out to ensure the ECU is earthed correctly. Note for volt drop testing a 4425A was used to take advantage of the floating inputs. More information on floating inputs can be found here: https://www.picoauto.com/library/traini ... ing-inputs

: Image 3

During peak inrush we capture an average volt drop of 785 mV which I initially thought was high (for the engine ECU) however, the addition of a ground “jump lead” did not change this value

Engine ECU power supplies and grounding during the fault condition were captured and qualified below:

: Image 4

Please ignore channels A, B & C as these are only relevant during engine run conditions

Channels D, E & F ensure our +12 V, Engine control relay supply and Ignition supply are present when the fault occurs

Channel G qualifies the engine ECU can ground the engine control relay during the fault condition. This confirms our grounding issue (within the ECU) is only related to the starter motor relay

Channel F confirms the engine ECU CAN signal is still evident, even though we lose a number of CAN ID’s when the fault occurs. I have confirmed the missing ID’s are related to the engine ECU

Additional symptoms and notes:

Periodically this engine will crank and start as normal

Removing the starter relay and joining pins 30 & 87 will allow the engine to crank and start (with ignition on)

The engine will periodically cut out when running.
I have tried to capture this event but of course, when connected to the scope the engine will not cut!

The questions I would like to ask:

How does the engine ECU receive the cranking request from the ignition switch?

As a cranking request input, I can only see pin 86 from the starter motor relay going to 12 V during cranking. This could be interpreted as a cranking signal but it seems very odd?

The engine ECU could also receive a cranking request signal via CAN, but again this seems overkill. Any input here would be most welcome.

Finally has anyone experienced a similar issue of the engine ECU failing to ground the starter relay pin 86?

Thank you again for taking the time to read through the diagnosis carried out to date

Take care…..Steve

liviu2004 · Post by **liviu2004** » Tue Jan 26, 2021 5:13 pm

I’ve fixed same model car, 2003 same engine same problem, automatic gearbox, with another ecu. Let me try to find my records...

Technician · Post by **Technician** » Tue Jan 26, 2021 7:15 pm

The ECU's on the engine are notorious for filling up with water. Have you disconnected the multi-plugs and looked at the pins to see if they are swimming? Yes I've seen a few vauxhalls like this and they still started and ran.

liviu2004 · Post by **liviu2004** » Tue Jan 26, 2021 7:15 pm

All right, got my info back, I wanted to do a Case Study here at that time, but not sure, I think I forgot. To confirm, ECU receives a key on signal from CIM, and the start command is a CAN-BUS signal from CIM to ECU.

However, I made a topic on another forum, and I will translate it herewith.

Car, Opel Vectra C Station Wagon, 2003, engine Z32SE V6, engine ecu Motronic ME 3.1.1., automatic gearbox 5 speed, sitting for the last 2 years in a parking lot.

Fresh Owner, cheap car, complain, engine stops random, either cold or warm. Parts cannon shoot at, with crankshaft position sensor, new battery and now he was ready to change the Column Integrated Module, CIM. Which btw is a known problem on this car generation.

Owner also complained about the following faultcode: B3601-04 Key capture relay circuit malfunction.

Immediately realized this is not the issue, as this fault is generated when the CIM coil, which retains the key in the cylinder slot, when the gearbox is not in Park, was missing. This is a plug&play part and plays no role in starting.

Diagnosis

With engine started, water at 64C, engine stopped. In the same time, my GM MDI2 lost comm with the engine ecu, but rapidly came back.

When cranking, engine MIL flashed random, start motor was either engaging or disengaging random.

1st I've scoped the CIM to ECU ignition voltage, with contact on. Second channel at the ECU side, in order to exclude some wiring issues. You can clearly see, when the engine stops, that this is not a CIM problem.

Next step was to check if ECU loses constant power and or ground. Blue channel on ecu body and red channel on permanent plus. No problems here. Yes, I've noted the ground noise here as well, but the pictures shows a filtered channel.

Then I noticed all kinds of noises at the throttle body, as it was trying to calibrate the throttle. Keep in mind, ecu will calibrate the throttle body on each ignition on event. But we know the ignition voltage does not change state when this problem occurs.

I took the throttle body plug out and well, of course, it forced some fault codes, but guess what, engine did not stop anymore. So two options here, or the throttle body has an issue or the ecu is defect, how to prove it.

Checked throttle body movement by hand, all smooth. Lets see what current draw is when ECU activated:

Weird signal from what I would expect, but 6A is not a lot. Can I say throttle body has an issue, mm, not really.

Lets look to the voltage command, we must have a PWM signal.

Nothing unusual so far. What about now?

Is it me, or it seems that the IGBT in the red line does not shut off? I suspect in that moment, there is a short between plus and minus, voltage drops and ECU reboots.

This was a tough call, but I condemned the ECU as defect. Second day Owner got hold of an ECU, I had reprogram it to the car and it was a confirmed fix. 3 May 2020 and I never heard from him since.

liviu2004 · Post by **liviu2004** » Tue Jan 26, 2021 7:34 pm

If we were to think that the ECU throttle body design looks something like this, as an example, when we close the throttle body, transistors 1 and 4 would be energized in PWM control, when we open the throttle body, then transistors 2 and 3 would be energized in PWM control.

What I think happens is that one of the transistors remain conductive, either due to control fault, which I believe more, either the transistor itself. Then one of the chain 1+2 or 3+4 remain conductive, which would create a short circuit, would drop the supply voltage to the ecu, and thus cause a reboot.

Steve, I am very curious if you can reproduce the measurements.

liviu2004 · Post by **liviu2004** » Tue Jan 26, 2021 7:50 pm

Starter current delay is caused by electromagnetic and mechanical inertia of the start relay. Cannot be anything else.

Post by **Steve Smith** » Wed Jan 27, 2021 8:32 am

WOW, thank you so much for taking the time to reply with such detail and clarity

My confidence is now building once again after taking a knock over the past few weeks wrestling with this vehicle

You mentioned 2 items in your reply that stand out to me and I never highlighted in my original post (Sorry)

1. This vehicle had been standing for many months without turning a wheel and only restarted for its annual MOT
This is when the fun started as prior to its hibernation the vehicle was fine. (Similar scenario to your vehicle Liviu)

2. I have heard the throttle body repeatedly cycling during the numerous test I have carried out and assumed this was preparation for restart (Setting to base position)

I hope to be back on this vehicle later today and will feedback my results ASAP (which maybe the weekend)

Once again thank you so much, take care......Steve

liviu2004 · Post by **liviu2004** » Wed Jan 27, 2021 6:29 pm

Steve, I would be more than happy to assist you with an reset ecu (not tested), free of charge, considering how much Pico has helped me in the past, I would be honored to be able to help you.
I am not sure though how to deal with customs shipping from EU to UK. But let me know.

When the vehicle occasionally starts on battery support, my theory is that the increased voltage would overcome the voltage drop due to driver short circuit, thus enabling ecu software to stay alive.

From what I read from others, a transistor either works or not, so my suspicion on the root cause is a defect of ultrasonic soldering of the base of one of the transistors, thus ecu not being in full control of. Known Opel / Vauxhall problem with ecus mounted on the engine and aluminum conductors.

I hope you're able to move along in the diagnosis and keep in mind, while symptoms can be alike, always follow your instincts and procedures.

Post by **Steve Smith** » Mon Feb 01, 2021 7:43 am

Thank you once again Liviu for your kind feedback and generous offer which I will accept. Everything we produce here on the forum, newsletter, website, software and hardware is all achieved through team work and its great to know that we are not only supporting but educating/helping one another too

Every time I read through the forum, I learn something new and realize very quickly you simply cannot retain everything. The best we can do is document our diagnostic adventures here for all to see and this way, we are no doubt helping someone somewhere. My shout for help here is a typical example and has opened my eyes to a product of which I am not familiar, so thank you again

On with our diagnosis and throttle control:

You are onto something Liviu with throttle control, the waveform below captures the initialization/ calibration of the throttle at ignition on. The throttle plate is moved slightly to ensure “no stick” condition
Note the instability of the PWM signal at ignition on which appears to have no ground reference and “floats” (Remember, the ECU also fails to ground the starter motor relay) This “instability” appears to stabilize approximately 12 seconds after the initialization/calibration of the throttle!

: Image 1

The greater concern is the sporadic throttle motor current which appears to fluctuate between – 6 & + 6 A. Below we see this huge fluctuation during the initialization of the throttle

: Image 2

The sporadic deviation in throttle motor current is also present without any movement of the throttle plate and would appear to be in response to PWM command

: Image 3

Moving onto throttle motor behavior during the fault condition of the starter motor repeatedly engaging/ disengaging during cranking:

Below we see the repeated initialization/calibration of the throttle whilst holding the ignition lock/switch in the cranking position. (This appears to be the Engine ECU continually resetting)

: Image 4

Above I have included the CIM signal as I now understand how the engine ECU receives the cranking request signal. (See below)

1. ECU pin 64 (CIM signal) falls to 0 V from 12 V when cranking (Ignition key in crank position)
2. CAN network delivers cranking request from CIM to Engine ECU
3. ECU pin 39 (Stater relay pin 86) increases to 12 V when cranking which the Engine ECU then pulls to ground to energize the starter relay

Moving on:

This vehicle will start as normal on the odd occasion and here is one such event. Notice how the throttle is only initialized/calibrated once at ignition on

: Image 5

With the engine running I have captured a WOT event (see below) where the instability of the PWM and sporadic current flow is clearly visible

: Image 6

Based on the sporadic current I would like to inspect the inner workings of the throttle motor assembly as we have seen a similar condition here https://www.picoauto.com/library/case-s ... power-loss

I need to determine if the instability of the PWM signal & sporadic motor current is consequence of failing throttle motor components or indeed a fault within the Engine ECU throttle control circuitry

As a final test, the throttle motor was disconnected from the Engine ECU and the throttle plate moved through a full cycle by hand. Here we measure the induced voltage within the throttle motor windings. Note, the spikes in the capture below occur when the throttle plate changes direction and is allowed to “spring back” (uncontrolled) to the rest position

: Image 7

My line of thinking with the above test is that we can possibly reveal a motor winding fault based on the induced voltage waveform. If this waveform resembled the sporadic current we have seen in earlier captures, we may be onto something with the throttle body motor. We most certainly have a saw tooth effect in the induced voltage and this supports dismantling the throttle body/motor assembly for further investigation! (I will keep you informed)

One final curve ball, I noticed the connected scan tool indicates a continual deviation in throttle motor PWM duty (82 to 62 %) when in reality, there is no activity present on the throttle motor wiring! (See channels A & B below)

: Image 8

To summarize where we are now:

1. Engine ECU drops communication with the vehicle CAN network when trying to crank
2. Engine ECU fails to ground stater motor relay during attempted cranking
3. Engine ECU continually initialize/calibrate the throttle motor during attempted cranking
4. Adding battery support will occasionally clear the above faults and vehicle starts OK
5. Throttle motor PWM control is unstable during attempted cranking (Throttle plate is stable)
6. Throttle motor current is sporadic in nature which appears to respond to unstable PWM control. This tends to improve once the engine/ECU is warm
7. Vehicle periodically cuts out without warning (This fault appears to have improved over the last week)
8. Whist the throttle motor PWM control is unstable and current is sporadic in nature, the throttle plate is stable and does not “flutter”

Subject to removal and inspection of the internal workings of the throttle motor, my gut instinct is to replace the Engine ECU. This is a tough call for a vehicle of this age, however, based on all the evidence and support above I believe we are justified

Liviu, thank you again for your support and please feel free (everyone) to add anything I have missed above. I will be in touch as soon as I am happy with the throttle motor

I hope this helps, take care….Steve

liviu2004 · Post by **liviu2004** » Mon Feb 01, 2021 11:03 am

Well done so far, we’ll be in touch by email.

PicoScopes
& kits

For vehicle manufacturers

EV kits and
accessories

Noise, Vibration and Balancing

Milliohm tester

Pressure sensors

Current clamps

Breakout
leads/boxes

Accessories

Automotive Forum