gepace 0.2.0

GE Pace library

GE Pace library

This library is used to control basic features of a GE Pressure Automated Calibration Equipment (Pace) models 1000, 5000 and 6000.

It is composed of a core library, an optional simulator and an optional tango device server.

It has been tested with the Pace 5000 model, but should work with other models.

It can be used with either the ETH or the serial line connection (read below on the recommended way to setup a serial line connection)

Installation

From within your favorite python environment type:

$ pip install gepace

Library

The core of the gepace library consists of Pace object. To create a Pace object you need to pass a communication object.

The communication object can be any object that supports a simple API consisting of two methods (either the sync or async version is supported):

• write_readline(buff: bytes) -> bytes or

  async write_readline(buff: bytes) -> bytes

• write(buff: bytes) -> None or

  async write(buff: bytes) -> None

A library that supports this API is sockio (gepace comes pre-installed so you don't have to worry about installing it).

This library includes both async and sync versions of the TCP object. It also supports a set of features like reconnection and timeout handling.

Here is how to connect to a GE Pace controller:

import asyncio

from sockio.aio import TCP
from gepace import Pace


async def main():
    tcp = TCP("192.168.1.123", 5000)  # use host name or IP
    pace = Pace(tcp)

    idn = await pace.idn()
    name = await pace.name()
    print("Connected to {} ({})".format(idn, name))

    # channel access:
    temp_A = await pace['A'].temperature()
    unit = await pace['A'].unit()
    print("Channel A temperature: {}{}".format(temp_A, unit))

    # loop access:
    source_1 = await pace[1].source()
    print("Loop 1 source: {}".format(source_1))

    # activate control
    await pace.control(True)

    # hardware only accepts queries every 100ms. Yo can, however,
    # group queries in single request:
    async with pace as group:
        pace.idn()
        pace.control()
        pace['A'].temperature()
    idn, ctrl, temp_A = group.replies


asyncio.run(main())

Serial line

To access a serial line based Pace device it is strongly recommended you spawn a serial to tcp bridge using ser2net or socat

Assuming your device is connected to /dev/ttyS0 and the baudrate is set to 19200, here is how you could use socat to expose your device on the machine port 5000:

socat -v TCP-LISTEN:5000,reuseaddr,fork file:/dev/ttyS0,rawer,b19200,cs8,eol=10,icanon=1

It might be worth considering starting socat or ser2net as a service using supervisor or circus.

Simulator

A Pace simulator is provided.

Before using it, make sure everything is installed with:

$ pip install pace[simulator]

The sinstruments engine is used.

To start a simulator you need to write a YAML config file where you define how many devices you want to simulate and which properties they hold.

The following example exports 2 hardware devices. The first is a minimal configuration using default values and the second defines some initial values explicitly:

# config.yml

devices:
- class: Pace
  package: gepace.simulator
  transports:
  - type: tcp
    url: :5000

To start the simulator type:

$ sinstruments-server -c ./config.yml --log-level=DEBUG
2020-05-14 16:02:35,004 INFO  simulator: Bootstraping server
2020-05-14 16:02:35,004 INFO  simulator: no backdoor declared
2020-05-14 16:02:35,004 INFO  simulator: Creating device Pace ('Pace')
2020-05-14 16:02:35,080 INFO  simulator.Pace[('', 5000)]: listening on ('', 5000) (newline='\n') (baudrate=None)

(To see the full list of options type sinstruments-server --help)

You can access it as you would a real hardware:

$ nc localhost 5000
*IDN?
GE,Pace5000,204683,1.01A

or using the library:

$ python
>>> from sockio.sio import TCP   # use synchronous socket in the CLI!
>>> from cryocon import CryoCon
>>> cryo = CryoCon(TCP('localhost', 5000))
>>> print(cryo.idn())
Cryo-con,24C,204683,1.01A

Tango server

A tango device server is also provided.

Make sure everything is installed with:

$ pip install cryocon[tango-ds]

Register a cryocon tango server in the tango database:

$ tangoctl server add -s CryoConTempController/test -d CryoConTempController test/cryocon/1
$ tangoctl device property write -d test/cryocon/1 -p address -v "tcp://192.168.123:5000"

(the above example uses tangoctl. You would need to install it with pip install tangoctl before using it. You are free to use any other tango tool like fandango or Jive)

Launch the server with:

$ CryoConTempController test

TODO

• Add on_connection_made callback to initialize controller with:
  • unit=K
  • cache IDN, fw revision, hw revision
  • should we cache system:name? and input:name? in theory in could be modified directly with the hardware front panel