Experimental Physics and
Industrial Control System

Folks,

 

measComp R3-1 is available.  This is from the release notes:

 

**********************

  - Added support for the USB-1608G on Linux. Thanks to David Dudley for this.

 

  - Added support for the USB-3100 series of devices on both Windows and Linux.

    These have 4, 8 or 16 channels of analog output, plus 8 digital I/O and one counter.

 

  - Added support for configurable analog output ranges in drvMultiFunction.cpp.

    The USB-3100 models are the first supported modules to require this.

 

  - Added new OPI screens for 4, 8, and 16 channels of analog output and

    analog output configuration.  Added Range to analog output configuration screens.

 

  - Bug fix for measCompDiscover for Ethernet devices when UNIQUE_ID was an IP address

    rather than a MAC address, and the device was on the local subnet.

    It was creating a duplicate entry in the device table.  This was harmless, but resulted duplicate values being printed with measCompShowDevices.

 

  - Bug fix for the TC-32 in the digital input function.  This could cause stack corruption and crash.

 

  - Fixed OPI screen errors for TC-32.

 

  - Added 4 missing analog input channels to E1608_settings.req for autosave.

 

  - Added Warren Jasper's low-level Linux drivers for the USB-3100 and the USB-1608G.

**********************

 

The next release of measComp will be R4-0.  It is a major change on Linux.  Previously it used low-level drivers from Warren Jasper, because Measurement Computing did not provide an SDK for Linux.  Measurement Computing now provides their Universal Library SDK for Linux.  It is open-source and is on Github: https://github.com/mccdaq/uldaq

It is similar to the UL library on Windows, but has different function names and syntax.  Switching to this library has allowed me to remove 30,000 lines of code from measComp, and we now have a vendor supported library.

 

I have converted the EPICS drivers to use UL for Linux, and this version is now the master branch on Github.  The previous version has been moved to the WarrenJasperDrivers branch.

 

I have found 2 minor issues that I am working with Measurement Computing to resolve:

-          With the USB-2408-2A0 when the Data Rate is 20 Hz or less there are occasional glitches in the temperature readings.  This is due to the Cold Junction Compensation being read incorrectly.

-          With the USB-CTR08 MCS scans with dwell times less than 0.001 second do not work.

 

I encourage people to test the master branch to see if there are any other issues before I release R4-0.

 

Thanks,

Mark