Renault Zoe | Warning message “STOP: Electric failure DANGER”

Technology is changing at a relentless pace, presenting challenges to technicians worldwide. For those who are not brand-specific, diagnosing anything and everything, these challenges are beyond consuming. The following case study is one such example. I would like to share the hurdles I had to be overcome when I attempted to diagnose a Renault Zoe with virtually zero product knowledge. Throw into the mix an “All Electric Drive Train” and you just know you are going to reach that point during the diagnosis where you ask yourself why.

Customer’s description

The customer reported that the following warning message appeared in the instrument panel: STOP: Electric failure DANGER”. This warning message cleared itself and did not prevent the vehicle from driving.

Technical description

Verifying the customer complaint is an essential step in the diagnostic process but it is also quite often a time-consuming task without success. On this occasion, it was noted that a different warning message would appear after the vehicle had been sitting for a considerable amount of time. The warning message “Check ELECTRIC system” would be displayed in the instrument panel at the point of unlocking the vehicle with the remote key before the driver’s door was opened.

NB: Modern vehicles may “power up” once they are unlocked via the remote key/keyless entry systems. Be aware that “power up” with electric vehicles translates to (and should be considered as) “Vehicle live and active”.

After talking to the customer again, we learned that the warming messages would clear themselves from the display, without requiring intervention from a scan tool.

Diagnosis

With the customer complaint verified, we confirmed the Vehicle’s ID and Specification.

Confirmation of vehicle specification is of the utmost importance when it comes to diagnosis, as there is often a temptation for customers to modify their car with fashionable accessories that lack the fundamental quality control and engineering that was intended for the vehicle. I am referring here to accessories such as parking cameras, dash cams, trackers, speed trap sensors, etc., all of which have the potential to consume the precious 12 V battery reserve, or worse still, interrupt communication lines where accessories are integrated into the vehicle network. The customer confirmed no such accessories were installed to their knowledge.  

The Customer Interview followed the 4 targeted, open questions principle below to establish the facts.

1. How long has the problem been evident?

Periodically for 6 months

2. When did you first notice the problem?

During the winter months (cold weather)

3. Has any work been carried out on the vehicle recently?

No work other than typical maintenance where a new 12 V battery was installed. (Replaced due to a failed routine test.)

4. When do you experience the problem?

When the vehicle has been left standing either overnight or over a weekend

Based on the answers I was already thinking about Parasitic Drain but I did not want make any assumptions.

The Basic Inspection confirmed no fluid leakages, no visible signs of damage to hoses, connections or wiring harnesses and no accident repair. Other than the new 12 V battery, the vehicle looked 100% original.

A vehicle scan of all onboard control units revealed the eight fault codes listed below:

Now, here is a curveball: In the capture above I used a Bosch scan tool, which revealed eight fault codes. A second scan, using a LAUNCH diagnostic tool, revealed the following three codes:

To summarize the vehicle scan, communication codes appear to be the order of the day:

• Bosch indicates three CAN codes at Motor Control Unit (MCU)
• LAUNCH indicates three CAN codes at Power Electric Block (PEB)

Both scan tools use a different description for the same unit under the bonnet (Combined motor, charger/inverter assembly) which Renault refers to as Power Electronics Block (PEB).

This is a typical hurdle to overcome with different scan tools and their interpretation/translation of vehicle codes/components.

Before diving in here, it is important to take a step back and check for Technical Bulletins (Recalls & Campaigns etc.). In our case, none were relevant. Based on the vehicle history and symptoms we could move on to possible causes.

Possible causes

• CAN network wiring
• CAN network components (ECUs)

The action plan

The action plan was predominately governed by accessibility, probability, and cost. Based on the acquired Bosch scan data we had five fault codes referring to either “CAN communication errors” or “Bus faults” (MCU 4.0 & EV ECU 4.0 respectively).

The action plan, therefore, focused on CAN:

• Measurement of CAN network activity
• Wiggle testing of CAN network wiring and connectors

To recap

Periodic warning messages appear in the instrument panel suggesting either:

• “Check ELECTRIC system” or, “STOP: Electric failure DANGER”
• Both error messages will clear without using a scan tool
• The vehicle will drive and perform normally with or without the error messages displayed

A variety of trim panels and under-shields were removed for access to numerous ECUs and their associated CAN wiring. The EV CAN connector at the HV battery was chosen as the measurement point for CAN circuit evaluation. Why choose such a remote connector for capturing EV CAN? My line of thinking here was water Ingress, given its location under the vehicle. However, we found no such water damage.

A point to note here when working with EVs: safety is paramount, along with the necessary training, PPE and live working certification.

Below, we have EV CAN activity captured at the HV Battery while being charged via the Electric Vehicle Service Equipment (EVSE or Charge Station).

Note the noise present on the CAN circuit while the HV battery was being charged. Could this be our communication fault? When using the math channel, A-B, we can see the noise has cleared and proves once again how fault-tolerant CAN has to be in such electrically noisy environments. By using the PicoScope CAN decoder below, we could go one step further and qualify that the data was decoding correctly during the “noise” period with only minimal errors due to the sample rate settings. The noisy CAN signal was, therefore, not the cause of our fault.

The wiggle test of CAN wiring and connections was up next and here we used the math channel A+B in conjunction with a mask. For more information on capturing intermittent CAN errors, check out this forum post, which includes a video tutorial. Below we can see how the noise only intrudes into the mask upon disconnection of the HV battery CAN connector, which is to be expected. Before and after disconnection, the math channel A+B remained inside the permissible “white” graph area regardless of wiring and connector manipulation.

Therefore, we concluded that our CAN communication codes are not relevant to a wiring or connector failure at this stage. While checking the harness routing throughout the vehicle, we discovered a Tracker/Telematics style device behind the fascia, which includes connections into a CAN network other than EV CAN! Could it be possible for the Tracker to influence or disrupt messaging on the EV CAN when not directly connected to this network? It’s a question I could not answer honestly but faced with a variable like this, the easiest option was to remove the device and continue the testing. We informed the customer who had no knowledge that such a device had been installed.

Before further testing, we erased all fault codes and here I discovered another concern with generic scan tools. The CAN communication codes revealed by the Bosch tool (U1000, U1001 & U1002) could not be erased, suggesting the faults were still present! However, they could be erased with the LAUNCH tool so highlighting the need for scan tool awareness when it comes to their ability to communicate on all levels with all vehicles. (This scenario could have had me chasing my tail for days.) 

Further testing and use of the vehicle over several weeks confirmed the inevitable, the warning messages had returned along with the following codes captured using the Bosch scan tool:

I have added colours to the second round of fault codes to indicate the changes that have occurred since the original vehicle scan:

Yellow: CAN or BUS errors

Red: Specific/Descriptive codes

Green: System codes

White: Not relevant

The fault codes were also checked using the LAUNCH scan tool:

EVC: 5 Codes

Df018 Consistent Multiplex signal for CC/SL, the data supplied to the cruise control or the speed limiter are not correct

Df126 Battery Health Status, internal electronic fault

Df125 Battery health status OK

Df125 Battery health status confirm

Df113 Pulse box signal incorrect

BMS: 1 code

Df010 CAN communication no further description

Note: There are no longer communication codes for MCU (U1000, U1001 and U1002) as revealed by Bosch and PEB (Df020, Df021 and Df022) as acquired by LAUNCH in the initial vehicle scan. While CAN/Bus faults were still present (using Bosch) the previous wiggle tests and EV CAN decoding had confirmed the integrity of the network. We shifted our focus to the specific/descriptive codes.

Using multiple scan tools (as above) most certainly introduces variables in the form of “translations” and “interpretations” of manufacturer DTCs and component names given by engineers who develop such tools! For example, MCU Motor Control Unit (Bosch) & PEB Power Electric Block (LAUNCH) are the same component! These descriptions could have you searching for components that do not exist!

To summarize

• Bosch provides an improved description of DTCs but assigns them to “P” codes (powertrain) that could be misinterpreted as generic OBD-II codes.

For example, P0510 is described as HV battery charge “Internal control unit fault” however, P0510 OBD-II code is 02 Sensor Circuit Malfunction (Bank 2 Sensor 1).

• LAUNCH provides unique DTCs but with minimal descriptions.

So which tool should we use and which codes are relevant? I have to say, at this stage, I do not know. Establishing an occurrence order to the fault codes was proving difficult and may or may not be possible using the vehicle manufacturer scan tool as we did in this case study. 

Establishing an occurrence order is often the key to finding the root cause of the original fault, allowing you to discard the numerous irrelevant DTCs that appear as a consequence. Once again, product knowledge and familiarity of the vehicle would help here, I had neither, and so I required an element of reverse engineering.

I used the Bosch scan tool and began to sequentially disconnect a number of EV ECUs in order to determine their titles as listed in the vehicle scan report. While time-consuming, we discovered three key pieces of information:

1. We know the location of the ECU listed in the vehicle scan report
2. We learn the title given to the relevant ECU by the scan tool manufacturer
3. We discover the ECU responsible for reporting the fault code to the scan tool.

For example, disconnecting the low voltage multiplug from the HV battery revealed the “HV Batt. 4.0 Master” is an ECU that resides inside the HV Battery assembly (as it was no longer displayed within the vehicle scan list). This feels far more comfortable from a diagnostic point of view than “System Name: HV Batt. 4.0 Master (Bus Fault)”.

Armed with this new-found knowledge of all the ECU titles, location, and codes, our next step was research. Love it or hate it, Google diagnostics is near impossible to ignore when you have drawn a blank with your diagnostic process!  After what seemed like hours of study, it appears there is a tenuous correlation between these OBD-II style codes listed by Bosch and the Vehicle Manufacturer fault codes. This link led me to the following discovery!

P0510 translates to Renault fault code 0510/ followed by a test number

For example:

Renault code 0510/F1 = 12 V Battery: Test plan F1

Bosch claimed P0510 = EV ECU 4.0 Internal control unit fault, Bus Fault and Unknown fault

P0512 translates to Renault code 0512/67 12 V Battery charge control: Test plan 67

Bosch claimed P0512 = HV System Signal invalid

The information above proved to be diagnostic gold and reinforces the need for manufacturer training (product knowledge) accompanied by access to technical information via the VM portal.  

Thanks to Renault for allowing cost-effective access to their technical information portal, I discovered that when dealing with multiple DTCs you must prioritize DTC 0512. This is not so much occurrence order, more of a “priority order”.

To recap

• P0512 was acquired by the Bosch scan tool
• DTC 0512: “Pulse box Test 67” was described in the link to the SPEAK EV forum above
• Pulse box is reported by the LAUNCH scan tool “Df113 Pulse box signal incorrect”
• Renault portal requests priority is given to DTC 0512 and indicates a production improvement to the Pulse box after 30/09/13

So, what is a Pulse box and how do we diagnose it?

Results

Following the Renault test procedure, the 12 V battery had to be tested before we could test or replace the Pulse box (Interesting further discovery here, Renault DTC 0510 96 = 12 V “Battery State of health”).

Above, we have the results of the conductance test, revealing the SOH as “low” for a SOC of 100%. In addition, it became apparent the battery did not have the right specifications for this vehicle.

• Failed battery specification 40 Ah EN 330 CCA
• Recommend battery specification 54 Ah 500 CCA
• The failed battery was therefore 26% below the recommended capacity.

We could not use the battery test procedure in PicoDiagnostics, as it requires that the 12 V battery is under load using the starter motor in order to determine the characteristics of the battery. You can find more information here.

This was an EV and so there was no conventional starter motor to load the battery. Therefore, to qualify the performance of the new 12 V battery against the old, we used PicoScope in conjunction with a traditional “12 V battery load tester”.

Below we plot the volt drop and current flow from both batteries (under load conditions) where the difference in capacities are revealed. Note how the old and new 12 V batteries commence the load test at similar open-circuit voltages, yet the old battery continues to fall under load while the new battery voltage plateaus and recovers to 12 .62 V post-loading.

The Pulse box (confirmed post 30/09/13) is located behind the LH headlight and responsible for load testing the 12 V battery before the customer switches to ready mode. Think of this as a load test across your 12 V battery before you drive the vehicle (similar to cranking an engine). This is achieved by connecting the battery positive to ground via a substantial resistor inside the Pulse box (protected via a Maxi fuse) while monitoring the voltage and current from the battery (see the set up in the image below).

The behaviour of the Pulse box was discovered during routine parasitic drain testing (after replacing the 12 V battery), when unlocking the vehicle with the remote key while the vehicle was in sleep mode (see channel D below).

Returning to the Technical Description of the customer's symptom, the warning message “Check ELECTRIC system” would be displayed in the instrument panel at the point of unlocking the vehicle with the remote key, before opening the driver’s door! Knowing now what we know from the above capture, shortly after unlocking the vehicle, the 12 V battery is load tested for approximately 200 ms and if a “failure” is determined, the instrument panel will warn the driver even before entry to the vehicle! Our discovery fits the symptom perfectly.

Confirmation of repair

Given the intermittent nature of the fault, and the conclusive evidence found above, the best course of action was to return the vehicle to the customer and to normal service/duty. After 5 months there has been no return of the symptoms/faults described above and no warning messages within the instrument panel. The customer has confirmed that the usage pattern remains the same.

List All Parts Fitted:

12 V battery

Additional comments

We cannot underestimate the vital role played by the 12 V battery system linked to EVs. There is a mindset that because EVs have huge capacity batteries, the 12 V system can be “side-stepped” as cranking current is no longer required and so “any old battery will do!”

As you can see from the carnage above, the incorrectly specified and inferior battery generated hours of study, days of work and hundreds of miles of travel.  (I know what you must be thinking, “all that for a battery!”) 

Moving onto generic scan tools, as we know, there is no one scan tool that fits all models and using 2 scan tools on one vehicle threw a number of curveballs into the mix. Unfortunately, they all have their strengths and weaknesses and with this vehicle, it was necessary to use the LAUNCH tool to erase the CAN codes discovered by BOSCH (U1000, U1001 and U1002)

By no means is this case study an assault on generic scan tool manufacturers (far from it) as without them a number of repair centers could not offer a diagnostic service. However, I hope it highlights the hurdles and variables that are introduced to diagnosis as a result of incoherent translations and explanations of DTCs and components, along with sporadic coverage of features such as reading and erasing fault codes.

One simple example would be “Battery system fail” as a DTC description on an EV, if we had “Low Voltage Battery system fail” or “12 V Battery system fail” this would be a huge step in the right direction.

Hindsight is a wonderful thing, and I can see how I went off-piste with this case study at numerous stages! In a perfect world, product knowledge, product training, access to a dedicated VM scan tool the relevant technical portal should have been the correct approach, but the real world is far from perfect. It could do with fewer hurdles for sure.

Many thanks to Steve Winn at Autocare and Pete Melville at HEVRA for their invaluable support and technical input.

Thank you again for taking the time out to read and view this study, I sincerely hope it helps.