Skip to content

Latest commit

 

History

History
571 lines (484 loc) · 18.5 KB

Documentation.md

File metadata and controls

571 lines (484 loc) · 18.5 KB
Raw

pylogix API

Everything with pylogix is done via the PLC module, all the properties and modules will be discussed beow. Whether you are reading, writing, pulling the tag list, etc. To make things easy for the user, pylogix will automatically make the necessary connection to the PLC when you call any of the methods.

Properties:

  • IPAddress (required)
  • ProcessorSlot (optional, default=0)
  • Micro800 (optional, default=False)
  • Route (optional, default=None)
  • ConnectionSize (optional, default=4002)
  • SocketTimeout (optional, default=5.0)

Methods:

There are a few options for creating an instance of PLC(), how you do it is a matter of style I suppose. My preferred method is using contexts, or with statements, but is up to you. Some options are:

comm = PLC()
comm.IPAddress = "192.168.1.10"
# do stuff
comm.Close()

or

comm = PLC("192.168.1.10")
# do stuff
comm.Close()

or

with PLC() as comm:
    comm.IPAddress = "192.168.1.10"
    # do stuff

my preferred

with PLC("192.168.1.10") as comm:
    # do stuff

Again, how you do it is up to you, each method works. There are some things to consider:

  1. Don't call Close() unnecessarily. If the PLC no longer sees a request, it will eventually flush the connection, after about 90 seconds. So if you are reading/writing more often than 90 seconds, don't call Close(), just keep reading, call Close() when your program exits

  2. With statements will automatically call Close() when the indent returns. It's a common issue where people write their with statement within a loop. Doing this will open and close the connection with each iteration of the loop. Instead, write the loop inside the with statement, that way, the driver is declared first, then the loop performs the actions.

  3. When using with threads, be sure to create an instance for each thread, as opposed to sharing the instance between threads

NO:

while True:
    with PLC("192.168.1.10") as comm:
        # no no

YES:

with PLC("192.168.1.10") as comm:
    while True:
        # ahhh, much better

All pylogix methods will return the data in as the Response class, which has 3 members: TagName, Value and Status. Not all members apply to every method, so there are occasions where TagName, for example, will be None. Like when getting the PLC time, TagName doesn't apply. Value's type varies depending which method is being used. For example, Value can be a single value when reading a single tag, or can be a list of values when reading a list of tags. Or Value can be the lgx_device type, when requesting device properties or discovering devices on the network

ret = comm.Read("MyTag")
print(ret.TagName, ret.Value, ret.Status)

IPAddress Straight forward, this is the PLC's IP Address, as a string. Or, for ControlLogix, the Ethernet module you will be accessing the PLC at.

comm.IPAddress = "172.17.130.11"

ProcessorSlot Integer for which slot the processor is in. By default, the value is 0, since it is most common for a processor to be in slot 0. For CompactLogix and Micro8xx, this is always true. For ControlLogix, the processor can be in any slot. In fact, you can have multiple processors in one chassis.

comm.ProcessorSlot = 4 # connect to controller in slot 4

Micro800 True/False property, False by default. The Micro820/850's support Ethernet I/P, but use a slightly different path, so the driver needs to know this up front. Set to True when accessing a Micro800 PLC's.

comm.Micro800 = True

Keep in mind, these PLC's are pretty feature limited, so not all pylogix methods will work with them.

Route Route allows you to bridge to other controllers through a ControlLogix backplane or 5380 controller in dual IP mode. Routes are always in pairs of tuples. An example:

comm.Route = [(1,4), (2, '10.10.10.9')]

NOTE: Routing has always been a feature of ControlLogix, so the value of going out of an Ethernet port has always been 2. That is, until the 5380 CompactLogix and dual IP mode came along. You use a value of 3 or 4, depending on which port you are going out of.

ConnectionSize pylogix, as it currently is, will make an attempt to connect at the maximum possible connection size, which is 4002 bytes. If unsuccessful, it will attempt at the next common max size, which is 508 bytes. The user can specify a connection size anywhere in between. The best practice is to leave this value at default. The main reason for making this configurable is a history problem. pylogix made this configurable originally. Later, the concept of default to the max size, then fall back another size was implemented. Making it configurable made sure that there was no compatibility issues.

There is no known benefit to reducing the connection size. In fact, you will get the best performance by leaving it at the max. There seems to be no latency issue with always using the larger packet size.

The early controllers and Ethernet modules supported connection sizes of 508 bytes. At around v18, Rockwell implemented connection sizes of 4002 bytes.

SocketTimeout If pylogix cannot connect to a PLC or loses its connection to the PLC, the default timeout is 5 seconds (5.0). If this time is too long, it can be lowered. Just be sure to not set it lower than the time it takes the PLC to reply to prevent false timeouts. PLC's typically respond in a few milliseconds, but that is not guaranteed.

Read

Read allows you to pull values from the PLC using tag names. You can perform simple reads using single tag names, or bundle reads using lists of tags names. Read is only currently capable of handling the fundamental data types (BOOL, SINT, INT, DINT, LINT, REAL, STRING). While you can read members of UDT's, it must be at the fundamental data type level. This method will currently return the raw bytes of the UDT values, which you will have to parse.

While it is necessary for pylogix to know the data type of the tag being read, to make it simple for the user, pylogix will discover the data type the very first time a tag is accessed. The data type is saved in a dict KnownTags so that this only has to happen once per new tag. This does cause some extra overhead and can impact performance a bit, especially if you are just reading a large number of unique tags, then closing. To help, you can provide the data type up front, which will skip the discovery of the data type.

NOTE: if you want to access program scoped tags, use the following syntax

Program:ProgramName.TagName

Single tag reads

To read a single tag, provide a tag name and a single instance of the response class will be returned.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.Read("MyDint")
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDint 8675309 Success

Read an array

To read an array, provide a tag name and the number of elements you want to read. Value in the response will be a list of the values you requested.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.Read("MyDintArray[0]", 10)
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDintArray[0] [42, 43, 44, 45, 46, 47, 48, 49, 50, 51] Success

Read a list of tags

The best way to improve performance is to read tags in a list. Reading lists of tags will take advantage of the mulit-service request, packing many request into a single packet. When reading lists, a list of the Response class will be returned.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    tags = ["MyDint", "MyString", "MyInt"]
    ret = comm.Read(tags)

    # print exactly how the data is returned
    print("returned data", ret)

    # print each item returned
    for r in ret:
        print(r.TagName, r.Value, r.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
returned data [Response(TagName=MyDint, Value=8675309, Status=Success), Response(TagName=MyString, Value=I am a string, Status=Success), Response(TagName=MyInt, Value=90, Status=Success)]
MyDint 8675309 Success
MyString I am a string Success
MyInt 90 Success

Skip the data type discovery

While I prefer to keep things simple and let pylogix get the data type for me, you can bypass this feature to get a little better performance. The data type discovery only happens once per tag, so if you are reading in a loop, this doesn't have much benefit. But of you are reading a large list of tags, it can speed things up.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.Read("MyDint", datatype=0xc4)
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDint 8675309 Success

Write

Use Write() to write values to PLC tags. You can write a value to a single tag, a list of values to an array tag or write a list of values to a list of tags. Write will return the Response class, which is mainly useful for the status to see if your write was successful or not.

Write a single value

To write a single value to a single tag, pass a tag name and a value. Be mindful of the data type, for integers (SINT/INT/DINT) use the int type for the value, for REAL, use the float type and for strings, use the str type.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.Write("MyDint", 10)
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDint 10 Success

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.Write("MyString", "I am a string")
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyString I am a string Success

Write list of values

You can write a list of values to an array by passing a list of values. You don't have to specify the length, pylogix will simply use the number of values in the list.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    values = [123, 456, 789]
    ret = comm.Read("MyDintArray[0]", values)
    print(ret.TagName, ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDintArray[0] [123, 456, 789] Success

Write multiple tags at once

Similar to Read, you can write multiple tags in one request. Pylogix will use the multi-service request and pack the requests into the minimum number of packets. You make a list, where each write is a tuple containing the tag name and the value.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    request = [("MyDint", 10), ("MyInt", 3), ("MyString", "hello world")]
    ret = comm.Write(request)
    for r in ret:
        print(r.TagName, r.Value, r.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
MyDint 10 Success
MyInt 3 Success
MyString hello world Success

GetTagList

Retreives the controllers tag list, including program scoped tags (default). Returns the Response class, where the Value will be a list of Tag class. pylogix also saves this list internally in TagList. Along with the tag list, pylogix also retrieves the UDT definitions, which are stored as a dict in UDT.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    tags = comm.GetTagList()
    for t in tags.Value:
        print("Tag:", t.TagName, t.DataType)

result:

pylogix@pylogix-kde:~$ python3 example.py
Tag: Program:MainProgram
Tag: MyDint DINT
Tag: MyDintArray DINT
Tag: MyString STRING
Tag: MyInt INT
Tag: MyUDT Pylogix

GetProgramsList

Retrieves only a list of the program names. This will automatically call GetTagList in order to get the list of program names. Only a list of the program names will be returned. This can be useful if you want to only get a list of a particular programs tag list.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    programs = comm.GetProgramsList()
    print(programs.Value)

result:

pylogix@pylogix-kde:~$ python3 example.py
['Program:WidgetProgram', 'Program:ThingyProgram']

GetProgramTagList

Retrieves a list of a particular program. Requires a program name to be provided. Returns the Response class where the Value will be a list of Tag class.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    program_tags = comm.GetProgramTagList("Program:WidgetProgram")
    for t in program_tags.Value:
        print("Tag:", t.TagName, t.DataType)

result:

pylogix@pylogix-kde:~$ python3 example.py
('Tag:', 'Program:WidgetProgram.WidgetDint', 'DINT')

GetPLCTime

Reads the PLC clock, returns the Response class, by default, Value will be the datetime class. Optionally, if you set raw=True, the raw microseconds will be returned.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    time = comm.GetPLCTime()
    print("PLC Time:", time.Value)

    raw_time = comm.GetPLCTime(True)
    print("Raw Time:", raw_time.Value)

result:

pylogix@pylogix-kde:~$ python3 example.py
PLC Time: 2021-04-20 15:41:22.964380
Raw Time: 1618933282970171

SetPLCTime

Synchronizes the PLC clock with your computers time. This is similar to what happens when you ae online with a controller and click Set Date, Time and Zone from Workstation, in Controller Properties of RSLogix5000 or Studio5000 Logix Designer. Returns the Response class, which is mainly useful for the status.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    ret = comm.SetPLCTime()
    print(ret.Value, ret.Status)

result:

pylogix@pylogix-kde:~$ python3 example.py
1618958684216474 Success

Discover

Sends a broadcast request out on the network, which all Ethernet I/P devices listen for and respond with basic information about themselves. All Ethernet I/P devices are required to support this feature. This is what RSLinx uses in its Ethernet I/P driver to discover devices on the network. Returns the Response class, Value will be the Device class.

NOTE: Because all Ethernet I/P devices are designed to respond to this, many think that pylogix will be able to communicate with 3rd party devices in some meaningful way. The CIP objects targeted by pylogix are Rockwell specific objects, not part of the ODVA spec. The CIP spec allows for vendor specific object. While there may be devices out there that support the same objects, I only have access to Rockwell PLC's, so I have no way to support them. I can offer no support for them.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    devices = comm.Discover()
    for device in devices.Value:
        print(device.ProductName, device.Revision)

result:

pylogix@pylogix-kde:~$ python3 example.py
1769-L30ER/A LOGIX5330ER 30.11
5069-L310ER/A 32.12

GetModuleProperties

Requests properties of a specific module. Requires a slot to be specified. This method is useful for querying devices that are in a chassis. Like local I/O in a CompactLogix chassis, or even modules in a Point I/O chassis. Returns the Response class, where Value is the Device class.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    device = comm.GetModuleProperties(3).Value
    print(device.ProductName, device.Revision)

result:

pylogix@pylogix-kde:~$ python3 example.py
1734-IB8 8 PT 24VDC SINK IN 3.31

GetDeviceProperties

Similar to GetModuleProperties, this queries a device at an IP address. This is useful for querying things that are not part of a chassis, like PowerFlex drives, or maybe a barcode reader that supports Ethernet I/P. Returns the Response class, where Value is the Device class.

Example 

from pylogix import PLC
with PLC("192.168.1.9") as comm:
    device = comm.GetDeviceProperties().Value
    print(device.ProductName, device.Revision)

result:

pylogix@pylogix-kde:~$ python3 example.py
1769-L30ER/A LOGIX5330ER 30.11

Additional information

When reading/writing, pylogix keeps a dict called KnownTags, this is used to store the tag name and data type for the purpose of not having to request this each time a tag is read or written to. There are users who read a list of 1000 or more unique tag names, retrieving the data type essentially doubles the time that it would take to read the values. While the performance has increased drastically in this case by utilizing the mult-service request to retrieve the data type. Still, there ae users that want maximum performance, so Read/Write allows providing the data type up front. Doing this adds the tag name and data type to the KnownTags dict up front, then skips the initial exchange. You can see the atomic data type values by printing the CIPTypes dict:

print(comm.CIPTypes)

When requesting the PLC's tag list, there is also a dict that is saved of the UDT definitions called UDT. This will be the Tag type, which contains a lot of properties. After reading the tag list, you can print this dict:

print(comm.UDT)