Granny charging

“Granny charging” is used for ultra slow, normal plug charging. So it’s not about charging up grandma, but charging AT grandma’s place, if she lives just a tad over half the range away and no decent public chargers on the way, read: emergency charging.

And there is another reason why I keep it in the trunk: I have had occasions where a flaky grounded charger put my Zoe in the “red nose” mode. I’ll talk about the reset procedure in another post, giving me back my precious cruise control, but one part is: a successful charging session, however short.

Here are two video’s of my home-build. It cost me roughly 150 Euro’s. A “real”, clunky one could easily set you back 400 Euro’s and Renault retails (retailed?) theirs here for more than 700 Euro’s.




For the technicians: no, this is NOT a fake system that simply puts the proper pilot signal on the CP pin. It is a decent OpenEVSE system, doing all the checks and balances.

LED schematic (thanks to user “seti”)



I took the liberty of posting a picture of a rebuild that reader “seti” made, see the comment thread below.

Granny Cable1

And another one by Andy Fraser.

Thank you all for rebuilding!

Posted in ZOE
88 comments on “Granny charging
  1. feschy says:

    Wow, I’m impressed!

  2. Jeroen Meijer says:

    Fun huh?

    And it allows me to drive to my sis living in Northern France from the last decent 22kW charger in Viroinval near the French-Belgium border.

  3. seti says:

    Hi great !
    Can you publish a schematic ? From charger and from Schuko-Plug ?

    Thanks !

  4. Crf says:

    I suppose the original Renault Granny does not care about the schuko polarity?
    Can you explain how I can tell if the phase and neutral wires are switched? How does Zoe behave?

  5. Jeroen Meijer says:

    Absolutely correct. I assume you made it (thank you!) and I copied it here. If that’s a problem, just let me know please.

    • Jeroen Meijer says:

      Hi Seti,

      I disagree on all items, sorry.

      Neon lamps are easy to break. LED’s are sturdy!

      Neon’s are impossible to notice in daylight.

      As you can see, the resistors are not much higher than for LED, so the leak current to ground is roughly the same (actually, about twice as much, the difference between nothing and nothing really).

      Furthermore, as the LED side is connected to earth, there is not more than 2 volts close to the housing. Neon goes up to 90 volts.

      And the LEDs are far easier to integrate in the plug. So, the LEDs are actually (in my opinion) better and safer than neon’s.

      Which brings me to your other suggestion: NO! πŸ™‚ That thing will accumulate water like hell. And if you meant: put all the electronics in there: No no no, it would require an expensive 4 wire cable to the plug and a more complicated GFCI.

  6. Andy Fraser says:

    That’s a very neat job. I’ve seen reports that the ZOE is not very happy at 6A, so 10A would be better, faster and more efficient (assuming that the relay can handle it).

    I’ve built two 10A granny leads using the Mainpine EPC (one lead lives at MiL’s house).

  7. seti says:

    Hi Jeroen,
    what version of OPEN-EVSE Software you used ?
    the actuall v3.10.4 is too big for my mini-pro …
    or do you comment out anything in the software ?

    Thanks and best wishes for 2016 !

  8. Jeroen Meijer says:

    Hello seti. I used a somewhat older version (3.3.2) and indeed I did not include all options, mainly because a few functions were either useless or unneeded. For instance, GFCI is useless, as it would only protect the few cm’s between the electronics and the car. No LCD is needed either. A current coils is not included either. It measures if the car takes more power than allowed. Not many commercial chargers have that function btw. The sketch only use 11K now, less than half the size of the mini.

    Happy 2016 πŸ™‚

    EDIT: just for clarity: I am not saying a GFCI is unneeded, on the contrary, it is VERY needed indeed. I am saying it doesn’t make sense to have one at the end of the cord. The outlet you are using for ANY granny charging (in essence, every outlet, always) should be protected by a decent GFCI.

  9. seti says:

    Hi Jeroen,

    Thanks for reply!
    I tried now for hours to scale the OPEN-EVSE software to minimal size (you spoke abt 11K) Because the software is not very modular built, i only can shrink it on abt 17K , I am not able to complete comment out GFCI, button and LCD parts, they are spreaded over many source files … For me that “granny” hardware isnt the problem only the software-programming, like C++ πŸ™
    Maybe you could publish your 11K version ?

    Thanks and greets

    • Jeroen Meijer says:

      I would be happy to share my version with you, but I am slightly confused to the “why”. Are you trying to use a 16KB mini?

  10. seti says:

    Yes right, i want to use the Arduino Pro Mini with 16K Flash and ATmega168 Controller. Only 14336Bytes are free to use …


    • Jeroen Meijer says:

      I will send a zip to the email address you left behind when making comments. I am pretty sure this needs to be compiled with an old Arduino development environment, 1.0.5

  11. Andy Fraser says:

    Could you send me the zip too please?

  12. seti says:

    o.k. now i am testing πŸ˜‰
    Should i terminate some undiscribed Arduino-inputs to high or to low ?
    All components onboard. Arduino is running, green LED ist briefly on, then red LED comes … πŸ™ an staying on

  13. Jeroen Meijer says:

    The red LED is (of course) an error situation. without any further info I would guess the GFCI indicator is not happy. Did you use “my” software drop? Because it is disabled there. Otherwise, you can try to disable it in the .h file.. Grounding input pins in general is a good idea, but I didn’t with mine and I have seen no problems. These Atmel chips are builld like armoured tanks somehow and hard to break. I miswired an ATtiny85 and it heated to bad I burned my finger on it. Kept working.

  14. seti says:

    yes, its your software drop … GFCI is disabled … I will look at the serial CLI … maybe to localize the error more exact …

    • Jeroen Meijer says:

      If it is the drop as I send it to you, about the only problem can be your analog circuit, as stuck relay detection is off too. Just my 2 cents.

  15. seti says:

    Yes, it was the analog circuit. you can believe it or not, but 7 (!) pieces from 10 pieces *NEW* ordered P6KE15CA were defective !
    (only working in one direction)
    and of course i put a defective part in :-((
    now it seems working … next test tomorrow …


    • Jeroen Meijer says:

      Pfff, bummer.

      But I don’t understand where you put that part? I think it was only specified in the GFCI part, to protect peaks coming from the coil?

  16. seti says:

    P6KEs are very good to suppress higher voltage peaks to a circuit. I think it is only a precaution to protect the circuit from outside.
    I ve built many ham radio stations on higher locations like towers and masts on hills and mountains (danger of lightnings) . The P6KEs allows to reduce the maintenance service significant. I m a great fan of these items! … But not to the distributor of these defective parts …. πŸ™


    • Jeroen Meijer says:

      I get that, so just to ensure that I am correct, you added that to the build yourself right? Not something I am overlooking in the schematics?

      • Jeroen Meijer says:

        Ah OK thanks. Those blasted schematics are so “small print” (and my eyes so old πŸ˜‰ ). I indeed have one of those on the CP.

  17. seti says:

    No, the P6KE15CA is in the schematic at the Pilot-Pin to GND. Schematic of the analog circuit. And in Partlist ,there is a P6KE16CA also, but only the P6KE15CA is available. CA-Type is bidirectional Type, not like a simple Z-Diode as shown in schematic.

    Look here:
    and here:


  18. seti says:

    Well … its done !!! πŸ™‚ And its works !

    Thank you Jeroen for your passionate help !!!


  19. seti says:

    too early happy …….
    circuit works perfect on table…. signaling faultless (tested with EVSE-Tester) …. all o.k. ,so far i can say …
    but my ZOE does not react to the granny charger …. only LED turns from green to yellow while connecting….
    nothing else happens ….
    another mobile Charging Station (Bettermann) works perfekt on same power outlet ….
    currently i am at a loss …..

    Or does the ZOE eventually needs a Resistor PP to GND ? I do not think so ,but what do you think about it Jeroen ?
    The Betterman Box who works, does need the original Charging Cable incl. builtin Resistors.


  20. seti says:

    Ahh, ok. i ll try first a 1k5 / 13A Resistor. wondering these R is not in documentation …

    Thank you !

    • Jeroen Meijer says:

      Because it is the specification of how much the CABLE can handle. Your charger may do 64A, your car may do it to, but if you put in a 3x 1.5mm2 cable, well…….

      So it is not part of the charger. However, our Granny charger has no cable! πŸ˜‰

  21. seti says:

    ok,thanks! I ve read many docs about PP. The PP ,in my opinion, is not very well discribed. i understand that the PP Resistor tells the Charging Station how much the cable can handle, thats clear, but its not clear that the car also needs this information. In the description of the original charging-cable i saw, that both sides are terminated PP to GND.
    I will try tomorrow !

    Thanks again

    • Jeroen Meijer says:

      Nope, it tells the CAR how much it can take. The PP pin is not even connected to the charger. If my memory doesn’t fool me, if you open up your blue cable you will find that PP doesn’t even run to the other end.

  22. seti says:

    o.k. now its clear! The charger only “tells” the maximum of available current (with PWM-Signal) and the car regulate the charging-current depending on its own requirements. To these own requirements belongs to know how much current the cable can handle.
    Thanks for explanation!

    • Jeroen Meijer says:

      Yes, it is the car’s logic that takes the PWM, the cables’s max rating and it’s own battery-black magic to determine what it will do. All three values are hard limits. Some chargers measure what the car takes and will kill the charging process if the car takes more than what it tells the car through the PWM signal. It’s of course rather different with DC charging……

  23. seti says:

    Hi Jeroen,

    halfway lucky …
    The charger charges with 1k5 or 680Ohm R-PP to GND exactly with 9.0 amps. No difference detectable.
    (when 1k5 equals to 13A and 680 eq to 20A ,thinks thats o.k.)

    But CanZE says “Max AC charge pilot amp” = 9.0 Do you know where to adjust this value in arduino-software ?
    In open_evse.h i tried in line 154 to 162 to make the values higher (16A) an re-program the controller. But nothing changed. The charger charges exactly with 9.0amps. Batterie is abt 75%full.


    // current capacity in amps

    // minimum allowable current in amps

    // maximum allowable current in amps
    #define MAX_CURRENT_CAPACITY_L1 10 // J1772 Max for L1 on a 20A circuit
    #define MAX_CURRENT_CAPACITY_L2 10 // J1772 Max for L2


    // current capacity in amps

    // minimum allowable current in amps

    // maximum allowable current in amps
    #define MAX_CURRENT_CAPACITY_L1 16 // J1772 Max for L1 on a 20A circuit
    #define MAX_CURRENT_CAPACITY_L2 16 // J1772 Max for L2

    Do you know the difference between:
    // current capacity in amps
    // maximum allowable current in amps

    Thanks for patience

    • Jeroen Meijer says:

      // current capacity in amps

      Default values for the PWM signal (depending if it’s mode L1 or L2)

      // maximum allowable current in amps
      #define MAX_CURRENT_CAPACITY_L1 16 // J1772 Max for L1 on a 20A circuit
      #define MAX_CURRENT_CAPACITY_L2 16 // J1772 Max for L2

      Maximum to which the UI should be able to push it. Since you don’t have a UI, it is not very relevant.

      I bet your next question would then be: why the heck is in not doing 16A? I am not sure to be honest. Isn’t the firmware remembering the last setting in the EEPROM? To test that, either enable the one-button UI (almost impossible to do without display), bind the CLI, or erase the EEPROM.

      Other than that I really, really advice you not to go past 12A on a normal outlet.

  24. seti says:

    Ahh, good idea to erase the EEPROM!

    Made it at the moment (eeprom_clear) , tested , but nothing changed … 9.0amps
    (I set DEFAULT_CURRENT_CAPACITY_L1 and DEFAULT_CURRENT_CAPACITY_L2 to 15amps. (15A are 3A higher than predefined 12A , when i get 9A + 3A then i expect to get 12A … )

    If anyhow possible i would go to 12A with SchuKo.
    And no worry, for testing i have a 16A CEE socket. (blue one)

    o.k. enough for today , the charger charges , thats essential.

    Thank you very much again !

  25. seti says:

    Not enough for today yet πŸ˜‰

    Now it works perfectly at 12A !

    The solution is to set ALL values in open_evse.h at lines 154 to 162 to the desired amp-value.
    Setting all values to 12 (amps) the charger charges with exactly 11 amps in my ZOE.
    Settind all values to 13 (amps) the charger then charges exactly 12 amps !!!

    (measured with external amp / watt – Meter AND CanZE via OBD2)

    You can see here:

    Now i allow myself a little drink πŸ˜‰

    Best regards !!!

    aka Juergen

  26. seti says:

    For all who wants to rebuild the “granny” Charger:

    1kHz PWM with duty cycle 16,7% results in a charging current of 9A
    1kHz PWM with duty cycle 20,0% results in a charging current of 11A (Value in open_evse.h is “12”) (20% should be 12A, look link below)
    1kHz PWM with duty cycle 21,6% results in a charging current of 12A (Value in open_evse.h is “13”)

    Maybe this applies only to _MY_ old Zoe from 2013 at 1Phase 230V at Schuko-Plug.

    Details here:

    Good luck and success with rebuild !


    • Jeroen Meijer says:

      Only now I notice you also put in the control LED’s (I didn’t). Did you drill holes through the grip or do they simply “shine through” the white plastic?

  27. seti says:

    The “Three-Colour” LED is in an drilled hole. Ist very helpful to indentify the Charger-State.
    The onboard LED of the Power-Supply at the tail end of the grip shines through the grip .
    Nice to identify if or if not grid is on.

  28. fisnik says:

    Wow! Looks really cool. How much for you to make me one? πŸ™‚

    • Jeroen Meijer says:

      Thank you for the offer. To be honest, I am not too enthusiastic about building them. If you ignore the LED indications, it could put the car in an (user) unrecoverable state. And if things go seriously wrong…….. The project is not very difficult for an electronics enthusiast. Really the hardest thing is fitting it! Part costs are roughly 150 euro’s (and most of that is the connector).

  29. Andy Fraser says:

    I bought my connector here:

    It cost just under Β£62 ($87) including postage ($25) and arrived in about a week. Or, there’s one on eBay for Β£69 delivered:

    I am just beginning to construct mine. I’m using a 12V supply and relay (and a 12V to +/- 12V converter for the op-amp supply). I found a 12V relay on eBay which is already mounted on a circuit-board with driver transistor, diode and led:

    There are 5V versions, but they seem to be mounted on the board differently, so that it wouldn’t fit inside the connector so well.

    If anyone is interested (especially in the UK), I could order in a stock of the electronic parts and produce a kit, including a pre-programmed Arduino Mini-Pro.

    As mains plugs in the UK are not reversible, there shouldn’t be a mains polarity issue, so no requirement for LEDs in the plug.

    • Jeroen Meijer says:

      I know the three-prongs are not reversible, but are non-ground pins always wired the same way (i.e. neutral left, phase right)? I mean, ALWAYS? If not, I’d put in the LEDs anyway Andy. My first one blew the red LED because I had omitted the reverse diode. And as fate had it while I had given it to someone else for a while; he was in a bit of an emergency. He or his wife put in the plug the wrong way, red LED did not light up and he ended up having the dealer over who fought with the car quite a while.

    • Jeroen Meijer says:

      That relay looks fine. The component itself is the same as mine. I hope it will fit, it can be quite the puzzle. And please make sure the HV lines are tinned. Those PCB’s are often not up to snuff and you don’t want a fizzle right against your car.

      Do you have a link for your power supply? I couldn’t find one small enough for 12V and then also adding the -12V. The regulator on the pro mini will work fine with a VCC of +12, though if I were you I would not add the three LED’s right now. It can always be done later. Check the regulator’s heat first would be my unsolicited advice πŸ™‚

    • Paul Allen says:


      I know this post is AGES old, but I would be interested in purchasing a kit with a pre-programmed Arduino Mini-Pro (especially if it charges at 12A)

      If not can somebody please send me the “pruned down” version of the open EVSE firmware?

      I need to have one of these compact EVSEs in my life πŸ™‚


  30. Andy Fraser says:

    As UK 13A plugs have a fuse in the Live side (right-hand pin), the sockets should always be wired that way. It is of course possible that a mistake has been made in the wiring, so it would be advisable to check on each different socket where the cable is used. Perhaps including indicators on a UK plug is still worthwhile (it would also double as a house-wiring check).

  31. Andy Fraser says:

    I’m still working on a source for the 12V power supply. I’ve ordered one of these, which looks as though it should be small enough:

    Once I’ve got it apart, I’ll be able to see if it’s really capable of 240V input, but as it’s also available with a Euro plug connection, it should be OK:

    • Jeroen Meijer says:

      Those will have a switched power supply and that should be fine. Whether it fits is an entirely different ball game! Good find though! Don’t forget to post a few pics please.

  32. Andy Fraser says:

    Sadly, it was too big.

    I’ve found one that’s a suitable size, but it’s only 9V @ 500mA. I’ve ordered one in the hope that I can change a resistor in the feedback to get it up to about 11V. The 12V relay should pull in at 9V, but I’d like a bit more to be sure.

    It’s coming from China, so will probably take a couple of weeks.

    • Jeroen Meijer says:

      Well, why don’t you also order a “backup plan”. It will cost you no more than a few pounds: my reasoning at the time was the relay would be the main consumer and the 12V rails would be virtually power-free: the op amps use nothing, and the protocol wire, I think the craziest resistor value you’ll see on it is 220 ohms, so 20 mA max. It just cried for a symmetrical DC-DC converter somehow. I also didn’t loose the (admitted, rather models) losses in the linear power regulator on the nano, and the relay (same as your in essence) was available for 5V. The coil was rated 27 ohms, so 200 mA was rather doable. The 5V to +12/-12 (@50 mA) was available. It’s called a SIM1-0515D. It set me back a horrendous 6.50 euro’s.

    • Andy Fraser says:

      The PSU arrived today. It looks very similar to the one in @Seti’s picture. I have worked out that I need to change the 8.66kohm surface mount feedback resistor for a 12kohm in order to increase the voltage from 9V to ~11.5V. I experimented with 10kohm just to make sure that it did what I expected, and it raised the voltage to 10.1V, which is not quite enough for the +/-12V DC-DC chip. I will have to order the correct value resistor (Β£0.19 for 100).

      I will also be putting in a 3 colour LED, but I’m going to try drilling almost through the casing at the top in order to make it transparent enough to see the light without compromising the waterproofing.

    • Andy Fraser says:

      My granny cable is now completed and working. I can just see the 3 colour LED through the reduced thickness casing. I used a 3mm green LED in the 13A mains plug to indicate correct live and earth. If I do another, I’ll use a red/green LED to check both L & N to earth.

    • Andy Fraser says:

      If I don’t include the FedEx admin fee (mentioned above), the total cost including cable and RCD has been about Β£95 (about €120). Most of the internal parts came from China, so were very cheap – e.g. Pro Mini Β£1, PSU Β£1.50, Relay Β£3.

  33. Andy Fraser says:

    Don’t buy the connector from the ev-istore as mentioned above. Today I got an invoice from FedEx requesting an additional Β£18.42, comprising Β£6.42 VAT + Β£12 admin fee.

    • Jeroen Meijer says:

      Requesting, you are funny! Their phone number is +86…. = China. Never ever order something from China (actually, from outside the EU) over 20 Euro’s or you’ll get that. Good luck with the BREXIT on that one. I got mine for 95 euro’s locally + a few euro’s (from memory 8) shipping.

  34. Andy Fraser says:

    It’s not the VAT I object to; it’s the Β£12 admin fee.

  35. Jeroen Meijer says:

    @Seti You’d be happy to know I took your suggestion and put in a three color LED exactly the same way you did. Only difference is I put them in the “bottom half” of the grip, but that’s totally irrelevant. I seated it with an overdose of two component resin to keep it water tight and put tiny shrink tube over all 4 soldered wires at the LED ends and slightly wider ones over the free floating driver resistors. Works like a charm, thank you for the suggestion.

  36. cose385 says:

    I’m currently driving DIY electric car and I’m planning to get a Zoe this year – your web site is treasure trove of useful information! πŸ™‚

    Regarding this granny charger hack – I’m considering making the universal charging box which could be used both for regular schuko granny-charging and for connecting to standard industrial 3-phase plug to enable fast emergency charging when charging stations are not available. Industrial 3-phase plugs can be found in various workshops and they provide 16-32A on each phase (11-22kW) – much faster then schuko. In that case standard Mennekes charging cable would be plugged into this box and schuko or 3-phase would is plugged into the box on the other side – so the box would simulate charging station with ability to adjust / monitor current (LCD would be necessary in that case).

    Do you think that Zoe could charge that way and this is doable?

  37. Lars Nilsson says:

    Nice build and good to have a backup charging solution.
    I have an arduino pro mini, a cable mounted gfci and it would make me very happy if you could send me the code for the evse.
    I have same problem as Seti with disabling the lcd, gfci and buttons.


    • Jeroen Meijer says:

      I would happily send you that very old code drop, but have a look first at the chip on the pro mini. If it is a 328 instead of a 168. Only very old Pro Mini’s are 168 which are memory limited. If you have a 328, you better get more recent code from the openEVSE project.

Leave a Reply