Skip to content

Latest commit

 

History

History
135 lines (116 loc) · 10 KB

README.md

File metadata and controls

135 lines (116 loc) · 10 KB

taptap

This project implements the Tigo TAP protocol, especially for the purpose of monitoring a Tigo TAP and the associated solar array using the TAP's communication cable. This allows 100% local offline data collection.

The TAP protocol is described at docs/protocol.md. This system uses two networks, a wired "gateway network" and a wireless "PV network":

                     Gateway                PV device
                   device (TAP)            (optimizer)
               ┌─────────────────┐     ┌─────────────────┐
       PV   ┌─▶│   Application   │     │   Application   │   Proprietary
  network   │  ├─────────────────┤     ├─────────────────┤    │
            │  │     Network     │     │     Network     │    │
            │  ├─────────────────┤     ├─────────────────┤
            │  │      Link       │     │      Link       │   802.15.4
            │  ├─────────────────┤     ├─────────────────┤    │
            │  │    Physical     │     │    Physical     │    │
            │  └─────────────────┘     └─────────────────┘
            │                  ▲         ▲
            │                  └ ─ ─ ─ ─ ┘
            │  ┌─────────────────┐
  Gateway   └─▶│    Transport    │                           Proprietary
  network      ├─────────────────┤                            │
               │      Link       │                            │
               ├─────────────────┤
               │    Physical     │                           RS-485
               └─────────────────┘

Connecting

The gateway network runs over RS-485 and can support more than two connections. An owner may therefore connect a USB RS-485 adapter, or an RS-485 hat, or any other RS-485 interface without interrupting communication.

The gateway network supports a single controller. Most owners use a Tigo Cloud Connect Advanced (CCA), but there are alternatives, including older Tigo products and similar controllers embedded in GoodWe inverters. taptap can observe the controller's communication, without ever transmitting anything; as far as the other components are concerned, it does not exist. This allows owners to gather real-time information from their own hardware without going through Tigo's cloud platform and without modifying the controller, their TAPs, or any other hardware in any way.

Placement considerations

This system uses a 4-wire bus: ground (– or ⏚), power (+), A, and B. These wires are intended to run from the controller to a TAP, and possibly to another TAP, and so on. The A and B wires carry RS-485 signals. Tigo recommends putting a 120Ω resistor on the last TAP's A and B wires to terminate the far end of the bus, and they built a 120Ω resistor into the controller to terminate the near end of the bus.

If you are adding a monitoring device to an existing install, it would be best to move the controller's A and B wires to the monitoring device, and then to run new wires from there to the controller. Having said that, it should be fine to connect short wires from the controller's A and B terminals to the monitoring device, especially if you plan never to transmit. (Your monitoring device may also have a "ground" or "reference" terminal, which should go to the controller's gateway ⏚ ground.) In either case, make sure the RS-485 interface you're adding does not include a third termination resistor. The bus should always be terminated at the controller and at the furthest away TAP.

┌─────────────────────────────────────┐      ┌────────────────────────────┐
│                 CCA                 │      │            TAP             │
│                                     │      │                            │
│ AUX  RS485-1  GATEWAY  RS485-2 POWER│      │                    ┌~┐     │
│┌─┬─┐ ┌─┬─┬─┐ ┌─┬─┬─┬─┐ ┌─┬─┬─┐ ┌─┬─┐│      │   ┌─┬─┬─┬─┐   ┌─┬─┬│┬│┐    │
││/│_│ │-│B│A│ │-│+│B│A│ │-│B│A│ │-│+││      │   │-│+│B│A│   │-│+│B│A│    │
│└─┴─┘ └─┴─┴─┘ └│┴│┴│┴│┘ └─┴─┴─┘ └─┴─┘│      │   └│┴│┴│┴│┘   └─┴─┴─┴─┘    │
└───────────────│─│─│─│───────────────┘      └────│─│─│─│─────────────────┘
                │ │ │ │                           │ │ │ │
                │ │ │ ┃───────────────────────────│─│─│─┘
                │ │ ┃─┃───────────────────────────│─│─┘
                │ └─┃─┃───────────────────────────│─┘
                ┃───┃─┃───────────────────────────┘
                ┗━┓ ┃ ┃
              ┌───┃─┃─┃───┐
              │  ┌┃┬┃┬┃┐  │
              │  │-│B│A│  │
              │  └─┴─┴─┘  │
              │  Monitor  │
              └───────────┘
Future work: controller-less operation

In the absence of another controller, taptap could request PV packets from the gateway(s) itself. The gateway and PV modules appear to function autonomously after configuration, so for a fully commissioned system, receiving PV packets from the gateway without ever transmitting anything to the modules would likely be sufficient for monitoring.

Software-based connection method for owners with root access on their controller

Some owners have root access on their controller. These owners could install tcpserial_hook on their controller to make the serial data available over the LAN, including to taptap, without physically adding another RS-485 interface.

This method has several disadvantages: it requires root access, it requires (reversibly) modifying the files on the controller, it might stop working in future firmware updates, it only works when the controller is working, etc. It is a fast way to get started for some users, but consider wiring in a separate RS-485 interface instead.

Project structure

taptap consists of a library and an executable. The executable is a CLI:

% taptap
Usage: taptap <COMMAND>

Commands:
  observe            Observe the system, extracting data as it runs
  list-serial-ports  List `--serial` ports
  peek-bytes         Peek at the raw data flowing at the gateway physical layer
  peek-frames        Peek at the assembled frames at the gateway link layer
  peek-activity      Peek at the gateway transport and PV application layer activity
  help               Print this message or the help of the given subcommand(s)

Options:
  -h, --help     Print help
  -V, --version  Print version

% taptap observe --tcp 172.21.3.44
{"gateway":{"id":4609},"node":{"id":116},"timestamp":"2024-08-24T09:16:41.686961-05:00","voltage_in":30.6,"voltage_out":30.2,"current":6.94,"dc_dc_duty_cycle":1.0,"temperature":26.8,"rssi":132}
{"gateway":{"id":4609},"node":{"id":116},"timestamp":"2024-08-24T09:17:01.691683-05:00","voltage_in":30.75,"voltage_out":30.4,"current":6.895,"dc_dc_duty_cycle":1.0,"temperature":26.8,"rssi":132}
{"gateway":{"id":4609},"node":{"id":82},"timestamp":"2024-08-24T09:16:41.686961-05:00","voltage_in":30.55,"voltage_out":30.2,"current":6.845,"dc_dc_duty_cycle":1.0,"temperature":29.3,"rssi":147}
{"gateway":{"id":4609},"node":{"id":82},"timestamp":"2024-08-24T09:17:01.691683-05:00","voltage_in":30.95,"voltage_out":30.6,"current":6.765,"dc_dc_duty_cycle":1.0,"temperature":29.3,"rssi":147}
{"gateway":{"id":4609},"node":{"id":19},"timestamp":"2024-08-24T09:16:41.686961-05:00","voltage_in":30.35,"voltage_out":29.9,"current":6.865,"dc_dc_duty_cycle":1.0,"temperature":28.7,"rssi":147}
{"gateway":{"id":4609},"node":{"id":19},"timestamp":"2024-08-24T09:17:01.691683-05:00","voltage_in":29.85,"voltage_out":29.4,"current":7.005,"dc_dc_duty_cycle":1.0,"temperature":28.7,"rssi":147}
{"gateway":{"id":4609},"node":{"id":121},"timestamp":"2024-08-24T09:16:41.686961-05:00","voltage_in":29.8,"voltage_out":21.9,"current":5.25,"dc_dc_duty_cycle":0.7607843137254902,"temperature":29.8,"rssi":120}
{"gateway":{"id":4609},"node":{"id":121},"timestamp":"2024-08-24T09:17:01.691683-05:00","voltage_in":30.55,"voltage_out":22.8,"current":5.3,"dc_dc_duty_cycle":0.7725490196078432,"temperature":29.8,"rssi":120}

As of this initial version, the observe subcommand emits taptap::observer::Events to standard output as JSON rather than emitting metrics for InfluxDB or Prometheus, and it does not persist its own state, meaning the gateway and nodes are identified by their internal IDs rather than by barcode. These are the next two features to add.