Repairing an electric smart car

  • The 3rd generation smart electric drive is a very robust car. Usually it reaches 200,000 km without any noteworthy problems, neither it's high-voltage technology nor general.

    For workshops, there is hardly anything to do that they are not familiar with from normal cars already. The high-voltage technology is practically maintenance-free and things like brakes, steering and suspension are nothing new for them. With this blog post I want to take away the fear from the independent workshops of working on electric cars.

    As I learned myself, it's not difficult. It started with the teardown of the 22kW charger of an EQpassion follower. At the same time, I also learned how to replace it and what is important. All in all, it is not as complicated as you hear regularly from inexperienced workshops.



    Repairs to the electric smart are no rocket science, even if they involve high-voltage components themselves. In order to make it easier for independent workshops, I would like to give an overview in my blog:


    Repairing an electric smart car


    smart fortwo Coupé 451er ED 3,7kW-Lader, 05/2013; > 250.000km (bei 250.000 km 22 kW nachgerüstet)

    smart fortwo Coupé 453er EQ 22kW-Lader, 05/2018 - 05/2021; 109.000km

    smart fortwo Cabrio 453er EQ 22kW-Lader, EZ 04/2021; > 15.000 km


    "Aus Gaspedal wird Spaßpedal"

    - Smart

  • Hi,

    In your blog article you mention the following :- Using DAS / Xentry for retrieving diagnostic info and also an ED3 service manual.

    I have just purchased a new project, a non- runner ED3 (lo-cost) which I suspect has a bricked main battery.

    The vehicle was parked during covid and was apparently OK before this, this may or may not be true.


    With my Smart 451 CDI's I am using a C3 interface and XP laptop running a VM of DAS/Xentry version 2011-08. This unit was given to me by a garage owning friend who no longer needed it.

    I've plugged it in to the ED3 but it is a later VIN (~ mid 2013) and although DAS/Xentry sees the ED3 it does not provide information on the important systems.


    I can communicate using the Arduino/CAN module described on the forum and the HV battery appears dead. (attached text file)

    Googling the web for DAS / Xentry is a minefield of dubious links and vendors of interfaces/clones, it is very difficult to know what to buy.

    Do you have any thoughts on a way forward in order to communicate with the ED3 using a later DAS/Xentry or any other software available.

    Which hardware /software version have you used to communicate with your ED3?

    I've seen references to the J2534 interface via the OBD port but I am uncertain whether DAS/Xentry can use this interface and where I would obtain a later version of DAS/Xentry anyway.


    Any thoughts on this?



    Regards

    George

  • I've plugged it in to the ED3 but it is a later VIN (~ mid 2013) and although DAS/Xentry sees the ED3 it does not provide information on the important systems.

    It doesn't know it's control units. You would have to get an update for it to do so.


    I can communicate using the Arduino/CAN module described on the forum and the HV battery appears dead. (attached text file)

    That's a bug in the tool. Communication works, but the battery is so deeply discharged, that it can't show the values properly.


    Which hardware /software version have you used to communicate with your ED3?

    I've got the same hardware, but I'm using 2020 software. Google for a newer software, there are multible companies offering that.



    What we can already say by looking at the text file from the tool:

    The battery seems to have been good, bevor covid hit it. Range was close to back then, when it it was new. (I don't know how it's going to look like when you've manually recharged it) But it's defenitly deep discharged now.


    What I would do, if I where you: Start two projects at the same time

    1. Get new software for the Xentry tool
      • This way you can use it with the smart, when it's been recharged.
      • It's now 10 years old. An upgrade is overdue
      • Usually those upgrades aren't that expesive
    2. Remove the battery and start charging it
      • From the arduino tool we can already see, that the battery is defenitly deep discharged
      • By waiting for xentry you are risking cells beeing permanantly damaged
        • every day counts here


    Use the blog post I wrote about repairing the battery:


    smart ED3 17.6kWh battery repair (feat.P18051C / bricked BMS)

    smart fortwo Coupé 451er ED 3,7kW-Lader, 05/2013; > 250.000km (bei 250.000 km 22 kW nachgerüstet)

    smart fortwo Coupé 453er EQ 22kW-Lader, 05/2018 - 05/2021; 109.000km

    smart fortwo Cabrio 453er EQ 22kW-Lader, EZ 04/2021; > 15.000 km


    "Aus Gaspedal wird Spaßpedal"

    - Smart

  • Hi,

    Thanks for sharing your valued opinion on this matter, your blog/forum and the articles therein are invaluable to anyone who is interested in the electric Smart models.

    I will drop the battery pack as soon as possible and charge the cells manually as indicated in the blog post suggested.

    At work I can 'borrow' a lab psu required for this and I am comfortable working with HV electrical systems.

    It is good to know that a later version of Xentry will work with the C3 interface, I will find someone who has the 2020 version.

    Thanks again for your help, I will update this post when further information is available.


    Regards

    George


  • Hi,


    I have made some progress reading and clearing the fault codes of my ED3 using a 2020 version of Xentry (VM) with a J2534 pass thru interface which I have 'borrowed'.

    The codes in the pic above are the codes that, when erased re-appear a few moments later. All others have been cleared.

    There were many fault codes showing in many modules to begin with but I think that this was mainly due to the 12V system battery being drained.

    I have now made a start removing the HV battery and hope to have it out over the weekend.


    Regards

    George

  • I have removed the battery pack from the ED3, (eventually, work got in the way ), and started charging the cells individually. The pack is in good physical shape with no signs of damage anywhere, electrically however it's a different story.

    The total voltage across the pack is 10V.

    Module 1 - 4.5V

    Module 2 - 3.0V

    Module 3 - 2.5V

    Recovery of this pack is going to be interesting!

    The modules have been removed from the battery enclosure and are on the bench with the cooling plates removed to have better access to the individual terminals.

    All plugs and busbars have been removed including the BMS, current sensor,HV contactors etc.
    The CSE does not seem to be easily removable from each module and I won't attempt that but my question is, should the purple earthing plugs be disconnected when charging ?

    I think they should.

    If the CSE can discharge the module by a command from the BMS (P18501C) and the CSE can 'remember' this when the BMS is removed then it is likely that this discharge would be through the purple connectors to earth.

    This is speculation on my part as I don't have a schematic of the BMS/CSE and how they interact.

    I have started charging each cell individully at low voltage and low current (1V@10mA), I have had some success using this method with other Lithium based cells I have experimented with. (no guarantees though! )

    The V/mA is then increased depending on the cell state until the cell V is in the 'normal' range (>3.3V). That is the plan.

    The P18501C code may still be lurking in the BMS as the J2534 interface may not show in Xentry as it is not the MB dealer interface.

    The C3 multiplexer interface I have combined with Xentry may show it or I have read some posts on the Web about Verdiano via CAN which may allow direct access to the BMS on the bench.

    One thing is for sure this is going to take a long time.... and may not work is not likely to work but I will give it a go.

    Exciting stuff !

  • Recovery of this pack is going to be interesting!

    It definitely will. First charge the battery, then buy the reset, after you know for sure that all cells survived it.

    All plugs and busbars have been removed including the BMS, current sensor,HV contactors etc.

    Very good. :)


    The V/mA is then increased depending on the cell state until the cell V is in the 'normal' range (>3.3V). That is the plan.

    Sounds good.

    Exciting stuff !

    Definitely. Keep us updated. :)

    smart fortwo Coupé 451er ED 3,7kW-Lader, 05/2013; > 250.000km (bei 250.000 km 22 kW nachgerüstet)

    smart fortwo Coupé 453er EQ 22kW-Lader, 05/2018 - 05/2021; 109.000km

    smart fortwo Cabrio 453er EQ 22kW-Lader, EZ 04/2021; > 15.000 km


    "Aus Gaspedal wird Spaßpedal"

    - Smart

  • Update.

    Good news :-)

    The HV battery is being slowly charged and all cells look good so far.

    Each cell has 'absorbed' the following :-

    ~ 1 A @ 10-50mA charging rate to get to 1V

    ~ 1 A @ 50-100mA charging rate to get to 2V

    ~ 1 A @150-200mA charging rate to get to 3V

    I don't know if this charging method is more or less likely to improve the capacity but at least the cells are responding OK so far.

    It is still technically discharged at the moment at ~ 3V per cell,

    I am also trying to establish if any cells are discharging through the CMS as they are being charged and by what value.




    Not so good (but interesting) news, :-(

    After the HV battery was removed I locked the car and left it.

    There was no reason to access the car until Wednesday last as I had something in the boot I needed.

    Opened with the remote and all seemed OK then.

    Yesterday I tried to get in and the 12V battery was flat (< 5V). The 12V battery is/was new and had been fully charged.

    Prior to HV battery removal, I checked for a current drain at the 12V battery and there was almost nothing.

    Further investigation is needed but I read about something similar happening with an ED3 when the Norwegian owner removed the speedo display and the battery went flat overnight.

    This may or may not be relevant to the original cause of the HV battery depletion, but it's certainly interesting.


    George

  • Hi

    I had a dead battery too, I charged the 3 module with an adjustable Transformer, bridge rectifier and some incandescent Light bulbs from 25 to 300 Watt, as inline resistors. My Battery showed also no Communication with the three modules. While charging the battery manual and i made sure the individual cell stayed below 4 Volt and they stayed balanced. I got the 3 bank to 113 Volt and reinstalled the battery. Now I could read the voltage of each cell through the can bus with my OBD2 reader. But P18051C was now also present and various other HV circuit fault. The BMS is on it way to be reset. The Battery is open in my garage. It over a month since I charged the battery and it still show 111Volt per bank.

  • Sounds good. Only thing that's not left is reseting the BMS.


    smart fortwo Coupé 451er ED 3,7kW-Lader, 05/2013; > 250.000km (bei 250.000 km 22 kW nachgerüstet)

    smart fortwo Coupé 453er EQ 22kW-Lader, 05/2018 - 05/2021; 109.000km

    smart fortwo Cabrio 453er EQ 22kW-Lader, EZ 04/2021; > 15.000 km


    "Aus Gaspedal wird Spaßpedal"

    - Smart

  • Hi Stefan,

    It is interesting to hear from someone else in a similar situation.

    My HV battery is still looking good, I am targeting 345V (115V per module). I used 3 x 60V Lab PSUs borrowed from work to get the all cells to ~ 3V evenly by serial linking the PSUs. I then connected all the modules together and I'm now using a 400V 1A PSU I bought (HSPY-400-01) current limited at 500mA to get to the target. (I am at 315V at the moment)

    I could not retrieve the P18051C code when the HV battery was in the car but I am going to make a few cable extensions to test the HV battery without physically re-installing it in the car and see if the P18051C code is there.

    If it is I will send it off to be fixed.

    There are a few Youtube videos showing the ability to access the BMS/HV battery data using Canbus and Vediamo which I would also like to try.


    I have a question to ask if I may.

    Do you know what triggered your HV battery to deplete in the first place?


    My 12V battery went flat rather quickly after the HV battery was removed, I charged the 12V battery and fitted it back into the car with a wireless current and voltage transducer. The info is sent to a small display outside the car which I can see easily. If I open the car with the remote I can see the current jump up to 16A briefly as various systems become active and then drop down to almost zero when locked. A week has passed and nothing unusual has happened but I will continue to monitor this.

    The voltage is dropping slowly < 0.1V per day


    George

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!