In this article, we will use the InfluxDB APP in ctrlX to set up InfluxDB, send Queries from Node-RED and TwinCAT3 TF6420 to a time series database, and create a simple Dashboard in InfluxDB.

Let’s get started!

InfluxDB?

The InfluxDataTM* data platform InfluxDB is designed and optimised to process large volumes of time-stamped data from sensors, applications, users and systems. It also facilitates the collection, storage and visualisation of large volumes of time-series data.

ctrlX AUTOMATION – InfluxDB

InfluxDB is available as an app for ctrlX AUTOMATION InfluxDB is not just a database, it collects and analyzes data from almost all sources, including MQTT, OPC UA and Modbus, so a number of Open source Telegraf Plug-ins are provided.

Why use InfluxDB?

Using InfluxDB offers a number of benefits. For example,

Fast processing of time series data
Data analysis using Data Explorer
Built-in dashboards for direct viewing of data

InfluxDB is also written in Google’s programming language Go (Golang), and InfluxDB 2.0 is written in a new language called Flux Flux is optimized for database extraction, transformation and loading processes.

Its syntax is based on JavaScript, a common language, and the main advantage of Flux is the ability to integrate different data sources using, for example, third-party APIs.

Install APP

Open the ctrlX web server and click ‘Install an app’ to install the APP.

Click on ‘Install from file’.

InfluxDB Appを選び、App FileをOSにUploadしてください。

Install the InfluxDB App with the “Install” button.

just a second..

Done!

To the left of the web server, there will now be an entry called ‘InfluxDB’.

Configure Click screen?

Next, click “GET STARTED” to build the InfluxDB App installed on the ctrlX OS,

The initial User settings screen appears.

Enter the User and Bucket names and proceed with CONTINUE.

Done!An API Token is displayed, which is required to access the InfluxDB App from an external device, and can be copied and pasted on Notepad or similar.

Done!This is the first InfluxDB App screen.

Click on ‘Data Explorer’ in the right-hand Slide Bar.

This is the Data Explorer screen, which allows you to visualise the time series data stored in InfluxDB.

Reference Video

This is the video description Version for this article.

Reference Link

PLCNEXT#Play with Influxdb

PLCNEXT#Play with InfluxDB x telegraf

Project#Beckhoff TwinCAT3 x Siemens S210 Servo Drvie_Part6

Implementation From Node Red

Before connecting to the Beckhoff TwinCAT3 TF6420, a communication test is first performed from Node-Red.

This is the library used in this article to test the communication between Node-Red and the ctrlX InfluxDB App.

npm install node-red-contrib-influxdb

You can find more information on the relevant Node from the Link below.

https://flows.nodered.org/node/node-red-contrib-influxdb

Node-Red Side

After installing the InfluxDB Node, you will find the InfluxDB-related Node in the Editor screen > Storge item in Node-Red Add Influx bath to Flow.

As shown in the diagram below, the Trigger Node of the TimeStamp flows into the input of the Function Node and the output of the Function Node is connected to the Influx batch.

Configure Database Configuration

The next step is to connect InfluxDB.

Server

Server is the IP address of the ctrlX OS.

Organization

The Organisation is the name of the organization that was initially set up in ctrlX InfluxDB earlier, in this case “mana”.

Bucket

The bucket name will be the bucket name initially set in ctrlX InfluxDB earlier, this time “mainBucket”.

Token

Token is the authentication key that Node-Red needs to access InfluxDB.

Fnction Code Node

msg.payload = [
{
measurement: “weather_sensor”,
fields: {
temp: 5.5,
light: 481,
humidity: 51
},
tags:{
location:”garden”
},
timestamp: new Date()
},
{
measurement: “alarm_sensor”,
fields: {
proximity: 1003,
temp: 19.5
},
tags:{
location:”home”
},
timestamp: new Date()
}
];
return msg;

Result

Deploy Flow, click on timestamp and send Data to InfluxDB.

In the Data Explorer screen > mainBucket, you can now see the data you have just published in Node-Red! Select the data you have visualised.

Click on the SUBMIT button on the right.

Done!

Implementation From beckhoff

Now that the communication between Node-Red and ctrlX InfluxDB has been confirmed, the next step is to access it using the TF6420 with the Beckhoff TwinCAT3.

Beckhoff Side

Before proceeding with the actual Tutorial, a brief introduction to the specifications of the TF6420 is in order: the Beckhoff TF6420 also supports InfluxDB versions 1.x and 2.x. The database configuration is done via the “TwinCAT Database Server Configuration of the database is done via the “TwinCAT Database Server Configurator”.

When connecting to InfluxDB 2.x, “InfluxDB2” must be set in the TwinCAT Database Server Configurator and the authentication method can be “None” The authentication method can be “None”, “Username/password” or “Token”.

This is the variable type supported by the TwinCAT3 TF6420. (It should be noted that the WSTRING type cannot be supported).

As a time series database, InfluxDB2 has some special features and each measurement of a time series contains the following items

Time Columns
Tag Columns
Field Columns

Time Columns are stored in InfluxDB as UNIX epoch times; the Function Block provided with the TF6420 automatically converts TwinCAT time to UNIX epoch time, so the Programmer can send free TimeStamp Programmer can send a free TimeStamp.

Activate License

To test the connection with the InfluxDB App installed on ctrlX with the TF6420, activate the TF6420 license in advance.

Add Database Configuration

To add a new Database project to the TwinCAT project, click Add>New Project.

Select Empty TwinCAT Datase Server Project and proceed to Next>.

Enter a Data Server Project name and create the project with Create.

Done!

Add New Database Configuration

To add a new database, click Add New Database.

Done!A new DB Configuration has been added.

Configure DB

Set Database Type to InfluxDB2.

The settings related to InfluxDB2 are now displayed. Note that if you set another Database Type while you are doing the connection settings, the original settings will disappear.

Server

Server is the IP address of the ctrlX OS.

Org

The Organisation is the name of the organization that was initially set up in ctrlX InfluxDB earlier, in this case “mana”.

Bucket

The bucket name will be the bucket name initially set in ctrlX InfluxDB earlier, this time “mainBucket”.

Authentication

Set Authentication to “Token” and paste in the authentication key needed to access InfluxDB.

Check

Finally, click the Check button to test the communication.

Done!It can be connected to TwinCAT3 and ctrlX InfluxDB App.

Check Sql Query Editor

Before we actually create the programme, let’s take a look at the contents of the ctrlX InfluxDB App database using the Sql Query Editor tool on the TwinCAT TF6420.

This is the TF6420 Sql Query Editor tool screen.

Click the Refresh button.

There was a time series of data that had just been sent by Node-Red.

Right-click>Query to send a Query to the ctrlX InfluxDB App.

By Default, the TF6420 Sql Query Editor also generates the Query already.

Query and click FB_SQLCommandEvt.ExecuteDataRetrun.

Done!!Data returned from InfluxDB App.

DUT

T_QueryOptionTimeSeriesDB_Insert

This is a structure for writing data to a time series database, and arrays of structures can be specified by symbols. The individual array elements are mapped as Rows with their own TimeStamp in the database and are generated by specifying a start timestamp and a time cycle between data sets.

TYPE T_QueryOptionTimeSeriesDB_Insert
STRUCT
pSymbol : PVOID;
cbSymbol : UDINT;
sDataType : STRING;
nDataCount : UDINT := 1;
sTableName : T_MaxString;
nCycleTime : UDINT := 100000;
nStartTimestamp : ULINT := 0;
END_STRUCT
END_TYPE

Variable	Type	Descirption
pSymbol	POINTER TO BYTE	Symbol Pointer written to database
cbSymbol	UDINT	Length of the specified Symbol Pointer
sDataType	STRING	Specified Symbol name
nDataCount	UDINT	Number of data sets in the transferred array.
sTableName	T_MaxString	Name equivalent to table
nCycleTime	UDINT	Time interval of the data set
nStartTimestamp	ULINT	Timestamp of the first dataset in TwinCAT time: 100 ns intervals from 1 January 1601.

Function Block

FB_NoSQLQueryBuilder_TimeSeriesDB

This is a Function Block for defining a Query for a TimeSeries database, and the Query is sent via Function BlockFB_NoSQLQueryEvt.

VAR_INPUT

Variable	Type	Descirption
pQueryOption	POINTER TO BYTE	Address Query Options.
cbQueryOptions	UDINT	Length of Query Options.

Tc3_Eventlogger

Since TwinCAT3 (4022.20), the Database Server TF6420 can now support the TwinCAT3 Eventlogger. This allows the details of the Function Block to be read out via the interface.

RequestEventText

This Method returns the event text.

VAR_INPUT

Variable	Type	Descirption
nLangId	DINT	Specifies the language of the event, 1033 = English.
sResult	REFERENCE TO STRING	Variable size (in Bytes) where the event text is stored.

VAR_OUTPUT

Variable	Type	Descirption
bError	BOOL	TRUE = there is an error.
hrErrorCode	HRESULT	S_OK=Method execution OK

Return

Variable	Type	Descirption
RequestTextEvent	BOOL	TRUE = Execution complete.

Program

This is the program.

PROGRAM MAIN
VAR

State: E_DbLogState;
bWriting: BOOL;

dbid: UDINT := 1;

//Queryparameter
QueryOption_TSDB_Insert : T_QueryOptionTimeSeriesDB_Insert;
//api
fbNoSQLQueryBuilder_TimeSeriesDB : FB_NoSQLQueryBuilder_TimeSeriesDB;
//execute queries
fbNoSqlQueryEvt : FB_NoSQLQueryEvt(sNetID := ”, tTimeout := T#15S);

// databuffer for 1 second with 10 ms time delta
Data: ARRAY[1..100] OF DUT_Data;

// error handling helper values
TcResult: Tc3_Database.I_TcMessage;
bError: BOOL;
sErrorMessage: STRING(255);

i: INT;
rand : DRAND;
nrand: LREAL;
TON:TON;
END_VAR

FOR i := 1 TO 100 BY 1 DO
rand();
nrand := rand.Num*5.0;
Data[i].Direction := 240.328 + nrand;
Data[i].Speed := 1200.3292 + nrand;
Data[i].Power := 1500.1975 + nrand;
Data[i].ID := ‘Unit13’;
END_FOR

TON(
IN:=State = E_DbLogState.idle AND bWriting
,PT:=T#1S
);

CASE State OF

E_DbLogState.idle:
IF TON.Q THEN
//bWriting := FALSE;
State := E_DbLogState.init;
END_IF

E_DbLogState.init:
fbNoSQLQueryBuilder_TimeSeriesDB.pQueryOptions := ADR(QueryOption_TSDB_Insert);
fbNoSQLQueryBuilder_TimeSeriesDB.cbQueryOptions := SIZEOF(QueryOption_TSDB_Insert);

QueryOption_TSDB_Insert.sTableName := ‘ServoMeasurement’;
QueryOption_TSDB_Insert.sDataType := ‘DUT_DATA’;
QueryOption_TSDB_Insert.pSymbol := ADR(Data);
QueryOption_TSDB_Insert.cbSymbol := SIZEOF(Data);
QueryOption_TSDB_Insert.nDataCount := 100;
QueryOption_TSDB_Insert.nStartTimestamp := F_GetSystemTime();
QueryOption_TSDB_Insert.nCycleTime := 10000; // (in 100 ns)
State := E_DbLogState.writing;

E_DbLogState.writing:
IF fbNoSqlQueryEvt.Execute(
hDBID:=dbid
,iNoSQLQueryBuilder:=fbNoSQLQueryBuilder_TimeSeriesDB
)
THEN
IF fbNoSqlQueryEvt.bError THEN
TcResult := fbNoSqlQueryEvt.ipTcResult;
State := E_DbLogState.error;
ELSE
State := E_DbLogState.idle;
END_IF
END_IF

E_DbLogState.error:
IF TcResult.RequestEventText(
nLangId:=1033
,sResult:=sErrorMessage
,nResultSize:=SIZEOF(sErrorMessage)
)
THEN
TcResult.Send(nTimeStamp:=F_GetSystemTime());
State := E_DbLogState.idle;
bError := TRUE;
END_IF
END_CASE

E_DbLogState.init:

This is the Step when TwinCAT sends the Query to the time series database.

The sTableName is the InfluxDB _Measurement, the sDataType must match the structure of the project by TwinCAT. pSymbol is the data to be sent to InfluxDB.

E_DbLogState.init:
fbNoSQLQueryBuilder_TimeSeriesDB.pQueryOptions := ADR(QueryOption_TSDB_Insert);
fbNoSQLQueryBuilder_TimeSeriesDB.cbQueryOptions := SIZEOF(QueryOption_TSDB_Insert);

QueryOption_TSDB_Insert.sTableName := ‘ServoMeasurement’;
QueryOption_TSDB_Insert.sDataType := ‘DUT_DATA’;
QueryOption_TSDB_Insert.pSymbol := ADR(Data);
QueryOption_TSDB_Insert.cbSymbol := SIZEOF(Data);
QueryOption_TSDB_Insert.nDataCount := 100;
QueryOption_TSDB_Insert.nStartTimestamp := F_GetSystemTime();
QueryOption_TSDB_Insert.nCycleTime := 10000; // (in 100 ns)
State := E_DbLogState.writing;

E_DbLogState.writing

This is a program that sends a Query to InfluxDB.

E_DbLogState.writing:
IF fbNoSqlQueryEvt.Execute(
hDBID:=dbid
,iNoSQLQueryBuilder:=fbNoSQLQueryBuilder_TimeSeriesDB
)
THEN
IF fbNoSqlQueryEvt.bError THEN
TcResult := fbNoSqlQueryEvt.ipTcResult;
State := E_DbLogState.error;
ELSE
State := E_DbLogState.idle;
END_IF
END_IF