Actions
Forum Thread Details
  • Replies 18 replies
  • Subscribers 278 subscribers
  • Views 2483 views
  • Users 0 members are here
Related
See a helpful answer?

Be sure to click 'more' and select 'suggest as answer'!

If you're the thread creator, be sure to click 'more' then 'Verify as Answer'!

Need help reverse engineering a control box for an EV charger

Wolfmane

Hi folks,

I'm at a loss. I have a 2016 Zero SR electric motorcycle. A couple of years ago I ordered a high-powered charger kit on IndieGoGo from a company that had been around for quite a while. It was three 3.3 kW AC chargers in a custom belly pan, with all the wiring and a control box for plug-and-play installation. I was led to believe that I would be able to adjust the chargers for whatever situation I was in so that I could charge with anything from a regular wall socket to a 3-phase Type 2 EVSE. Past versions of these chargers could be connected via Bluetooth to a mobile phone and have the settings changed at will. However, when I received it it turned out to be hardcoded to run the chargers full blast. That means I can _only_ charge from 3-phase 10 kW+ EVSEs, which is far from ideal. There were other problems as well, including insufficient cooling that caused the chargers to overheat and throttle down to a trickle after only a few minutes, meaning I couldn't actually complete a charge session in one go. It took 3 or 4 tries with cooling pauses between in order to fully charge the battery pack. I got fed up with it and went back to totally stock.

The problem is that the company that made the kit promptly left the market and completely ceased all support. They stopped responding to any contact. That leaves me with a charge kit that cost me several thousand dollars just sitting collecting dust, and a bike that is otherwise awesome that is only good for commuting and day joyrides and can't be road-tripped as I was hoping to do when I ordered the kit. (The factory charger takes 8.5 hours to charge, vs. 45 minutes to an hour with the kit.)

I can solve the overheating by not using the belly pan that came with the kit (it was too deep anyway and caused some clearance issues with speed bumps) and mounting the chargers in another location with better airflow complete with heatsinks and fans. However, the control issue is far beyond my capability. I'm good with basic electronics and have played around with Raspberry Pis and even dabbled with ESP8266 (I have plans for a LOT more of that) but when it comes to analyzing signals and protocols and programming them I'm completely out of my depth. But I don't know where to turn for help.

What I really need is to make a new control box with a few buttons and knobs that will allow me to set the modes of the chargers quickly and easily without having to get out my phone or another device and go through the hassle of connecting and fiddling with that other device. Especially since using touch screens with motorcycle gloves is a far from optimal experience.

There are some things I know: I believe the chargers were based on, if not made by, Elcon. The control box connects to them via 4 wires. The control box doesn't care how many or which chargers are running, the control wires are simply connected in parallel to the chargers and whichever chargers are physically connected to power will run at whatever settings the control box is sending out. I believe it's a CAN Bus protocol. The communication is probably not even 2-way; the controller just sends out the settings and the chargers' internal control circuitry takes care of the rest.

Can anybody help me, or give me some idea of where I can go to get help reverse engineering the control protocol and programming an ESP8266 or similar to control the chargers? I would be happy to pay a reasonable fee for any help. I can't afford to pay a large percentage of what the kit cost, though. Any help would be immensely appreciated.

  • Having been down this road myself (1985 engine control unit with zero documentation) I'd say collect what info you can and then start sniffing the control wires with a basic scope to see what is there. Capture what the controller outputs and what if anything is returned.

    A quick web search for Elcon charger protocol gave this:
    https://evwest.com/support/Elcon%20CAN%20Specification%202019.pdf

    Might not be the same but can give you some clues as to what type of data may be present.

    As for probing, something like a low cost Picoscope can get you quite far. It has decoders for CAN should it turn out to be that.

  • I might start by looking up the datasheet on the existing controller.

  • in reply to EagleFox

    There's no datasheet. It's custom in-house made and in a box potted with epoxy. (For water and vibration resistance.) Not even an infinitesimal chance there. There weren't even any installation instructions with the kit. No paperwork or information whatsoever.

    It can't be all that complicated. The guys producing these were not electrical engineers. They were really only hobbyists who only got into the business because they needed faster charging so they could race their own bikes. I'm sure they had somebody else design the circuits but they were doing all the assembly, setup and testing themselves.

  • in reply to beacon_dave

    Even a 'low cost" USB oscilloscope is getting into the territory of not being worthwhile for a single project. There is very little chance I'll ever use an oscilloscope for anything else.

    There is a "makerspace" in the area, although it's far enough to not be convenient. I may go there and ask around. They should at least have diagnostic tools I can use for much less than buying.

  • in reply to Wolfmane

    I suspect you will be end up spending a lot more than the cost of an entry level USB oscilloscope in your own time working on this and I suspect even more working blind without one.

    An entry level Picoscope is around £100 new, seems a small price compared to  "...a charge kit that cost me several thousand dollars just sitting collecting dust..."   The Picoscope software is a free download, so if you can pick up the hardware second hand, it will work out even cheaper.

    Makerspace is a good second option though. If you can capture the activity on the control bus and save it for reference, then you may get most of the info that you really need in a single visit. Enough to be able to hook it up to a computer and to start communicating with the charger in code.

    It looks like the Elcon chargers came in two types though - the original appeared to require an external adapter in order to use it on a CAN bus, whereas the later ones had the CAN interface built-in, If it turns out to be an Elcon based system, then you'll need to work out which type it is before attempting to hook it up to a homegrown controller. If it's not the later built-in CAN type, then you will either need to source the adapter or alternatively work out what the native control interface and protocol is.

  • in reply to Wolfmane

    I suspect it's a variation on a switch mode power supply, I'm not sure what you can see on the board at all if anything, but SMPS's pretty much have the same schematic though they may be laid out differently, maybe look up the general schematic for an SMPS and see if you can figure out what's what based on what if anything you can see. YouTube channel Haseeb Electronics has a lot of videos on SMPS....the guy is a whizz and maybe able to even build you what you want as he is accessible and seems easy to make contact with and is willing to help.

  • in reply to EagleFox

    It's not a power supply at all. It's a controller. It sends out signals to tell the charger what voltage to target, what amperage to run at, how fast to ramp, etc.

  • There are people on E14 who could probably help you. Since you don't have the right gear to do the sort of testing needed you will need to get the whole caboodle to someone who does. So you need to tell us where you are located and see if anyone volunteers to give you a hand.

    If you can get to Castle Douglas in South West Scotland I'd be willing to have a go myself.

    MK

  • Some pictures would help to understand with what we are dealing here.

  • in reply to beacon_dave

    Investing even 100 quid on a device the likes of which I haven't touched since the mid 1980s and will never again after this project isn't a wise investment at all. Putting that money toward actual help from someone who really knows what they're doing makes a lot more sense. I'm decent with a soldering iron and basic circuits, Ohm's law and such, but when it comes to complex circuits or analysis I am, as I said before, way out of my depth. Even if I did get an oscilloscope I would need help adjusting settings, connecting correctly without damaging anything and interpreting what I'm seeing, never mind successfully capturing anything.

    If I were capable of doing this stuff myself, I wouldn't be asking for help. Believe me, I always prefer to DIY when I can. But I sincerely need help with this one.

    There is a _slim_ but real chance that given the protocol I can work out the new control box myself. There is some fairly detailed information at https://zeromanual.com/wiki/DigiNow_Super_Charger_V2.5 (the "Unofficial Zero Manual") about how the immediately prior charging system's _adjustable_ controller worked. But knowing what information to send and how to send it (baud rate, how to mark the beginning and end of a message, etc.) are two very different things.