Electric cars are becoming more and more common on the road, but when they are parked in a driveway or garage, there are still some problems to be solved when charging them. Of course, there are many charging stations on the market, but they all have different features, functions, and even ports, so in order to really ensure full control of car battery charging, you may need to enter the parts box and pull out a reliable Arduino.

This project comes from [Sebastian], he needs this level of control over the charging of the Leaf, and he also has the skills to implement it from large high-voltage switch contactors to software running its network connections and web applications. This charging station also has all the available functions. It can tell the car to charge at different rates and can restrict it to charge at different times (for example, if energy is cheaper at night). It can monitor the charging status and other information of the car through the communication bus with the vehicle, and even has a front-end Web application for monitoring and controlling the device.

This project is based on Arduino Nano 33 IoT, and all the codes are available on the GitHub page of the project. Although we recommend taking extra care when handling power supply voltages and interfacing with high-priced products such as EVs, at first glance, the build seems to have surpassed all Ts and may even become a good prototype for the production unit in the future. However, if you do not need all the functions of the charging station, you can modify the car itself at any time to add some more advanced charging functions.

I originally wanted to criticize this person for not including GFI in his design, but his project page does say that he is inserting a GFI protection circuit.

However, I did not see any GCM in his design, which is another important safety feature in EVSE, because the vehicle always sits on four rubber tires.

I strongly urge those who want to DIY EVSE at least look at OpenEVSE, their grandfather, and at least understand the purpose of all its security features. The OpenEVSE derivative design has been approved by UL.

I think there is no difference between sitting on 4 rubber tires (a car) and sitting on a non-conductive wooden floor (my washing machine), so one of GCM or GFI seems to be sufficient. But I am not really expressing an opinion here...I will not claim expertise here. I really want this to make me wish my washing machine had GCM.

I am here because this is the first time I have heard of GCM. How does this pair of cars work? Does it have two different ground pins that should be connected to different points on the chassis and check the continuity between them?

GCM is ground continuity monitoring. It fits perfectly with GFI. If the frame of the vehicle/electrical/anything does not have a good ground connection, the GFI cannot detect the residual current until-for example-the user is electrocuted by completing the grounding circuit.

For your washing machine, it is at least conceivable that if the grounding conductor of the power cord fails, there may still be some grounding paths through the water supply system, so the GFI inserted into it may trip.

Oh, I didn't say how GCM works. GCM actually measures very, very small (and limited) ground currents to make sure it is possible. In my J1772 Hydra and OpenEVSE II design, the current is limited to 1 mA, but if the ground is not enough to allow 1 mA to flow, then it will malfunction and stop charging.

Where do you do the test? What is the loop length it tests? On the one hand, it needs to know the ground all the way to the chassis. On the other hand, if it is not, in failure mode, it needs to cut off the heat source before it leaves the wall-mounted box. If there is no dedicated return pin or something similar, I would not think of a solution to this problem.

In any case, I don't want to believe that running current on the ground is how it works, because Bill describes the problem below this reply. Your GCM opposes your GFI. You say that they get along well, but they are obviously destroying each other. For things with a metal chassis, I may prefer GCM to GFI, but the two together are obviously toxic. Some people will find that it will trigger their GFI and bypassing them will bypass their GFI.

I like the idea of ​​GCM, but if it forces end users to destroy their GFI, then it's not the beginning for me.

I hope it is really good, such as startup or periodic inspection, it disables all power outputs, performs a ground test, and then only enables power when it passes. This will not detect sudden failures, but it may be very bulletproof for open connections that remain constant anywhere in the circuit, and it will not counter GFI. The system you describe...let us say, I will not humiliate anyone by leaving it!

But of course the reality is a series of compromises...

This is why GCM and GFI belong to EVSE-so that GCM can be placed before GFI detection. You should not plug the EVSE into the GFI protection circuit, not only because of the GCM, but also because the GFI tolerance of the EV (according to the UL specification) is higher than that of the household GFI outlet.

> GFI cannot detect the residual current until-for example-the user is electrocuted by completing the ground circuit.

GFI does more in Canada, even without a ground wire. A long time ago, GFI was changed to GFCI, not just triggering on the stray current on its ground. Every GFCI I have seen for more than ten years has a common method of detecting stray current flowing to its ground, but there is a coil on each conductor (here is Live & Neutral). The current imbalance between the two will also trigger a fault to compensate for the situation where the output current does not return to the GFCI, because the defect provides another path. Therefore, when the current finds a different ground path, the GFCI in the old building where the wiring and sockets are not grounded will cause the GFI to detect an imbalance. (Like the one protecting my aquarium, it has been broken twice; two skimmers with faulty pumps.)

Please note that if you have multiple devices "protected" by a single GFCI, you can fail in one device to maintain the input/output balance in the GFCI and connect the circuit to another on that GFCI in a way that is not faulty. A device is detected. Each device should have its own GFCI to prevent this from happening.

GFCI aka Energy Conservation Detector/Analyzer "25 millikus passed through wire A, but did not return through wire B or C! Energy is not conserved, panic, turn it off!" It sounds like what a black countertop should have until You realize that wires are not electrically isolated and that current can flow out "other ways"

This debate can go on forever, and both sides are arguing. My opinion is not intended to be debated. If you use a reliable and well-proven method to keep the value at <30ma / <40ms response time, you can protect the user through the RCD/GFI method without any problems. On the other hand, the "newer" system GCM will help you detect whether the ground connection is loose, which is not new in industry as a method of continuous ground monitoring equipment in IT systems (or floating), but it is also It is not necessary because the user is protected by RCD/GFI; in addition, as Greg mentioned, you can only perform this operation on the uncovered charger side car (you will need an extra ground wire, which will make the design and (Charging standards are complicated) So in other words, GCM only checks when you are connected to the wall socket to ground, assuming that the ground path to the car is healthy. At this point, I agree with Greg in my opinion that Nick is trying to upgrade his OpenEVSE project ($$$ :))), and pointed out that electric car manufacturers are providing this feature for their chargers (I have tested BMW and Volkswagen’s new charger) they have it) but again just tick a box function, it will depend on what the user wants...

Please also note that the author is not satisfied with the voltage reading through the small VT. I haven't tried this project yet, but I have used 3PH power measurement on my solar inverter grid power supply (9 phases in total) and it works perfectly (using 220/10V VT). The trick lies in the calculation algorithm, I suspect it is EV RFI because My 12KW inverter will exist as RFI level. This summer I will try to make my own design plan: RFID, local OLED and WIFI connection (not for WAN, so I need to dig more to see which servers I can use if MQTT? Is it still free or my own? ) I have some loopholes in the APK part, so...ANYOANE ON BOARD? ? :)

I want to thank the booth authors for their work in the open project; this will help me because I will not start from scratch.

Nick, Bryan is a Norwegian, and GFI is an obligation, just like in the European Union. https://dehnes.com/electronics/2021/03/31/dehneevse_charging_station.html NL210 63A (bottom left corner of the box) is GFI, with high immunity, high frequency immunity and DC immunity, such as server protection or The inverter is a good choice. This is really safe. The grounding continuity monitor is not an obligation in the European Union. The reason is the type of grid. All EU countries have the same neutral point treatment. But it may be safer. For EU standards, we must have a yellow/green terminal on the box. Suitable for all cables Y/G. I think you can see the second picture. The continuity of G/Y is important, because your system cannot be Type II. (Your car is Grade II.) Brian, for safety, please attach photos and diagrams and handle them with your neutral point.

I am an electrical inspector. This is true. Please save your life.

This electrical schematic diagram is only applicable to EU, with TT network. 3-phase 240V phase and neutral. Please refer to this page: https://en.wikipedia.org/wiki/Earthing_system

And all parts must be directly grounded. More than 16 square millimeters. The resistance is R = U / I = 50 / 0,03 = 1666 ohms U = 50V (safety voltage for humans, beware of cattle and other animals...) I = 0.03A (GFI value) R = Test your ground Value, the value must be <1666 ohms (usually 100 ohms)

@Bryan: Your cable is too small, you must use 16mm² and 63A/Ph, or prove your choice is reasonable (it is possible to use the short-circuit cable on the box, but I prefer you to use 16mm², please).

Okay, this is a good project, but it is dangerous, please contact me, we have to replace some wires or protection. Sorry for the hype

This is my hope that professional comments on Hackady will bring fruitful and reasonable critics, not like an arrogant bastard who despises DIY stuff.

Thanks for your feedback. For clarification, the rated current of these stations is 32A (even though GFI is 63A), and I used 6mm^2 wires, which is enough for 32A. There is also a 32A circuit breaker in front (not shown in the figure).

Thanks! I am here to say basically the same thing. The cost of screwing up is quite high—property damage or possibly someone’s life in the end. Don't try this project unless you have received training and skills.

Ironically, the Clipper Creek EVSE cable that came with my Chevy Bolt apparently has a GCM (ground continuity monitor) circuit, and the test current that deliberately leaked into the ground conductor caused an imbalance, which caused me almost every day The 120 volt GFCI outlet in the garage has tripped. This has never happened to the EVSE power cord that came with LEAF that I used before.

The Clipper Creek EVSE has a maximum current of 12 amps, but it can operate at 120 volts or 240 volts, so I installed a 240 volt dryer outlet in the garage and made a pigtail so I can plug in the EVSE . The charging speed of the car is much faster, and the i-square R loss in the indoor wiring EVSE is approximately constant (*), because i=12 at any voltage. The overall efficiency has increased by a few percentage points because it takes less time to charge at 240 volts and wastes power. The charger inverter in the car will be more efficient when the input voltage is higher.

(*) 12 AWG wire is approximately 1.5 ohms per 1000 feet. My garage is independent, so it’s about 200 feet back and forth from the circuit breaker panel. At 12 amps, this means I consume 43 watts in the indoor wiring. The dryer plug circuit is 40 amps and uses 8 AWG wiring, approximately 0.6 ohms per 1,000 feet. When the car was charged at 12 amps, I only burned 17 watts in the 8 AWG wire of the dryer circuit.

The effect of GCM on GFI sensitivity is why in general, your EVSE should have its own GFI (and *test residual current after GCM) and should be on a circuit without its own ground fault protection.

This is why I designed Hydra and OpenEVSE II to use 1 mA GCM. But even so, the UL standard of EVSE GFI sensitivity is more like 17 mA, (I believe) slightly higher than the standard of ordinary household GFI. Therefore, you may find more annoying trips depending on the car (after all, the "battery charger" is actually part of the car. The EVSE is just a complicated power switch).

Oh, I saw your Pi EVSE hat, but I didn’t expect that you also have OpenEVSE II...

I have to look at that later this year. (I am considering the one or Pi hat in the project...)

Interestingly, the DC fast charging standard also stipulates the use of HomePlug GreenPHY for "smart" AC charging-have you done any experiments on this? Do vehicles that support CCS also support the use of HPGP/ISO 15118 for AC control? Or does the communication control through 15118 fall into the category of "someone has standardized it but there is zero implementation"?

15518 fort AC is real, but it is not widely used. IMO's "killer" function is plug and play and finer charging control.

Disclaimer, I work for EVBox.

I am very satisfied with the hardware on the EVSE hat, but I personally have no use for it, so I have not yet sat down to write software for it. I kind of hope that the user community can gather around the platform and do this for me. :D

I'm pretty sure that Canadian electrical codes require GFCI plugs (or circuit breakers) for outdoor plugs, and most likely in the garage.

I won't argue about the security requirements, but I don't think there is a need for a second PE line. If I remember correctly, PP and CP (boot and control pins, used for communication between EV and wallbox) use PE as ground.

Therefore, the failure of PE will result in the inability to communicate, and thus the power contactor of the EV cannot be charged/disconnected.

This is at least what I remember when designing some charging cables for commercial wall boxes.

OpenEVSE has all these features (I believe there is an open API), and it also incorporates more security features. I don’t want to belittle another DIY open source EVSE project, just because there are more complete functions, but the risk of messing up here is high enough, I will think twice, use similar things regularly without too much thinking about the security of the system And its expected failure mode.

Another thing I strongly recommend is that if you DIY handle things with tens of amperes of current, you should make the lid of the box transparent, and *regularly* check the wiring for signs of heating or degradation or whatever. The thermal damage caused by high current is a positive feedback loop, which can easily lead to fire.

In view of the fact that the possibility of a charged vehicle catching fire is very small, but it is still non-zero, consideration should be given to mitigating the fire.

I really wanted to build an OpenEVSE, but then the car arrived, so the next best thing is Wallbox Pulsar Plus with RPI calculation module. After the warranty expires, I will take over the computing module. Although there is now an easy-to-use API, it is not documented but it is easy to sniff using a web portal. I am controlling the charging current, lock/unlock unit and pause/resume based on my electricity bill (Octopus Go) and remaining solar energy (change the charging rate to the lowest to ensure maximum solar absorption).

Wow... Pulsar Plus uses Pi CM? ? ?

That just put it on my list.

Please be kind and respectful to help make the comment section great. (Comment Policy)

This website uses Akismet to reduce spam. Learn how to handle your comment data.

By using our website and services, you explicitly agree to the placement of our performance, functionality and advertising cookies. learn more