The next subsections describe the architecture for the data logger app. A Producer/Consumer pattern will be used in conjunction with a couple of State Machines for the data acquisition. The next section describes the reasoning.

Design Decisions

The requirements for the responsive user interface can easily be handled by a P/C pattern. Usually, when user interface responsiveness is required, it’s better to use the event structure in a P/C loop rather than a State Machine that requires polling for user interface changes. But how should the data acquisition be handled?

Alternatively, the DAQ could happen inside the consumer part of the P/C loops, perhaps via a Queued State Machine kicked off by a timeout event in the producer part of the P/C loops.

But the requirement that pushes towards the parallel loop(s) approach is the need to allow for acquisition from more than one device, each with possibly different acquisition rates. By having separate parallel loop(s) for the DAQ, each operating at their own loop rates independently of the consumer loop, the consumer loop can run when required without getting held off or holding off either the user actions or the DAQ.

More Code

AE for Moving Data between DAQ

The ‘Data Value Manager AE.vi’, previously discussed in the November newsletter,will be able to manage the transfer of data from the DAQ loops to the consumer loop. Each DAQ loop will push its channel data into the elements of the vector held in this AE. The consumer loop will retrieve the entire data vector from this AE for graphing and writing to the result file.

Queue Data Type for the Producer/Consumer Framework

A really good way to move data between the producer and consumer loops is to use a queue configured for a cluster containing an enum and a variant. The enum indicates the type of data to be consumed by the case frame in the consumer loop and the variant contains the data.

Main VI

An initial layout of the P/C loop, this queue, and some code comments are starting to be captured in the ‘DataLoggerExampleforCLD Main.vi’ in the project file, the link to which is contained in the next section.

Source Code for the Project

Continuing with the release of the source code to date as we implement to code for this app, a ZIP file for the entire LabVIEW project can be found here. These VIs need LabVIEW 2009 or above.

Summary

The Data Logger app development continues by making the decision to implement the app with a producer/consumer loop and parallel loops for the data acquisition. Next month, the app development will continue by continuing with the architecture and providing details about the data acquisition code.