mirror of
https://github.com/uhi22/pyPLC.git
synced 2024-11-10 01:05:42 +00:00
added FAQ. more details in sequence description.
This commit is contained in:
parent
6794100922
commit
162f2c1ed0
1 changed files with 47 additions and 3 deletions
50
readme.md
50
readme.md
|
@ -26,7 +26,11 @@ In this project, we call this mode *ListenMode*.
|
|||
- Install wireshark to view the network traffic
|
||||
- Install Pcap-ct python library
|
||||
- Patch Pcap-ct to support non-blocking operation
|
||||
- Run `python pyPlc.py` and use keyboard to trigger actions
|
||||
- Get and compile the exi decoder/encoder from http://github.com/uhi22/OpenV2Gx
|
||||
- Run `python pyPlc.py` and use keyboard to trigger actions, or
|
||||
- Run `python pyPlc.py E` for EVSE (charger) mode, or
|
||||
- Run `python pyPlc.py P` for PEV mode, or
|
||||
- Run `python pyPlc.py L` for Listen mode
|
||||
|
||||
## Architecture
|
||||
![architectural overview](doc/pyPlc_architecture.png)
|
||||
|
@ -109,7 +113,8 @@ Now, in the IDLE shall 3.10.6, the import works:
|
|||
Pcap-ct does not work with Python 3.4. After update to Python 3.8, it works.
|
||||
|
||||
## Example flow
|
||||
This chapter describes the start of a charging session, considering all layers.
|
||||
This chapter describes the start of a charging session, considering all layers. It is NOT the description of the currently
|
||||
implemented features, it is just a reference for understanding and further development.
|
||||
|
||||
Precondition: On charger side, there is a homeplugGP-capable device present, which is configured as CentralCoordinator.
|
||||
1. The charger (Supply entity communication controller, SECC) creates a "random" value for NID (network ID) and
|
||||
|
@ -190,7 +195,10 @@ min and max current. Now, the initialization phase of the charging session is fi
|
|||
is, to adjust the chargers output voltage to match the cars battery voltage. Also a current limit (max 2A) is sent.
|
||||
48. The charger confirms with PreChargeResponse. This response contains the actual voltage on the charger.
|
||||
49. The charger adjusts its output voltage according to the requested voltage.
|
||||
50. The car measures the voltage on the inlet and on the battery. If the difference is small enough (less than 20V?), the car
|
||||
50. The car measures the voltage on the inlet and on the battery.
|
||||
51. The above steps (PreChargeRequest, PreChargeResponse, measuring physical voltage) are repeating, while the
|
||||
physical voltage did not yet reach the target voltage.
|
||||
51. If the difference is small enough (less than 20V?), the car
|
||||
closes the power relay.
|
||||
51. The car sends PowerDelivery(Start)Request.
|
||||
52. The charger confirms with PowerDeliveryResponse.
|
||||
|
@ -259,6 +267,20 @@ https://github.com/uhi22/pyPLC/tree/master/results.
|
|||
As summary, the concept with the python script together with the compiled EXI decoder works. Further efforts can be spent on
|
||||
completing the missing details of the V2G messages.
|
||||
|
||||
### 2022-11-11 [*EvseMode*] Ioniq in the PreCharge loop
|
||||
The EVSE state machine, together with the EXI decoder/encoder, is able to talk to the Ioniq car until the PreCharge loop. The
|
||||
car terminates the loop after some seconds, because the intended voltage is not reached (no physical electricity connected to the
|
||||
charge port). Maybe it is possible to convince the car to close the relay, just by pretending "voltage is reached" in the
|
||||
PreCharge response. But maybe the car makes a plausibilization with the physical voltage, then further development would require
|
||||
a physical power supply.
|
||||
|
||||
### 2022-11-15 [*PevMode*] Draft of SLAC sequencer
|
||||
In PevMode, the software runs through the SLAC sequence with a simulated Evse. After SLAC is finished, we send a software
|
||||
version request broadcast message, to find out, whether we have at least two homeplug modems in the network (one for the
|
||||
car and one for the charger). If this is fulfilled, we should use the SDP to discover the chargers IPv6 address. But this
|
||||
is not yet implemented.
|
||||
|
||||
|
||||
## Biggest Challenges
|
||||
- [*ListenMode*] Find a way to enable the sniffer mode or monitor mode in the AR7420. Seems to be not included in the public qca/open-plc-utils.
|
||||
Without this mode, we see only the broadcast messages, not the TCP / UDP traffic between the EVSE and the PEV.
|
||||
|
@ -274,6 +296,28 @@ Any idea how to enable full-transparency of the AR7420?
|
|||
|
||||
## Other open topics
|
||||
- [*EvseMode*] [*PevMode*] Fill V2G messages as far as needed, to convince the car to accept it.
|
||||
- [*PevMode*] Implement SDP to get the chargers IPv6 address
|
||||
- [*PevMode*] Testing on real charger
|
||||
- improve docu (update layer diagram, improve hardware docu, add link to evse which provides the 5% PWM)
|
||||
- Resolve the todo-markers in the code
|
||||
- (and much more)
|
||||
|
||||
## FAQ
|
||||
|
||||
### Q1: What is the minimal setup, to catch the MAC addresses in the communication between a real charger and a real car?
|
||||
- Hardware: A TPlink TL-PA4010P homeplug adaptor, with the configuration for PEV. Modified according to the hardware manual https://github.com/uhi22/pyPLC/blob/master/doc/hardware.md
|
||||
- Software: Wireshark. Only wireshark. (The pyPlc project and the exi decoder is NOT necessary to sniff the MAC addresses.)
|
||||
|
||||
### Q2: Is it possible to use this software, to make the car closing the relay, so that I'm able to draw energy out of the car?
|
||||
Good question. This depends on how strict the car is. This first hurdle is to convince the car, to close the relay. This is
|
||||
done after a successful PreCharge phase. And it depends on the implementation of the car, whether it needs physically correct
|
||||
voltage on the inlet before closing the relay, or whether it relies on the pretended voltage in the PreChargeResponse message.
|
||||
The second hurdle is, that the car may make a plausibilization between the expected current flow (charging) and the actually
|
||||
measured current flow (discharging). The car may stop the complete process, if the deviation is too high or/and too long.
|
||||
|
||||
However, the software will help to explore and understand the behavior of the car.
|
||||
|
||||
### Q3: Is it possible to use this software in a car without CCS, to make it ready for CCS charging?
|
||||
That's is definitely a goal, at it looks reachable. Of course, two aspects need to be considered:
|
||||
- This project is not a final product. Further development will be necessary to ensure compatibility with chargers, and make is flexible for practical use.
|
||||
- Some parts are not covered by this project at all, e.g. communication with the BMS, connector lock, safety considerations.
|
||||
|
|
Loading…
Reference in a new issue