London LabVIEW User Group

cancel
Showing results for 
Search instead for 
Did you mean: 

Small and cheap platform for NI LabVIEW

Introduction

Hi there,

My name is Kevin Ross and I am a System Engineer for Austin Consultants. Over the last few months I have been working on a project for our sister company Green Running, which is an energy monitoring and reporting company.

The project I have been working on is to find a small, cheap platform for National Instruments (NI) LabVIEW. In this blog, I explore some options that will hopefully give you some insight into what is available at the moment.

LabVIEW Requirements

First things first, understanding what the minimum requirements are to run NI LabVIEW. You can find out more about this on the National Instruments website.

NI LabVIEW has been designed to run on the three main Operating Systems (OS), Windows, Mac OS X, and Linux. Depending on how you want to use LabVIEW will depend on the requirements. If you wish to use the development environment to develop code, then you require more from your OS compared to if you just want to install the run-time engine and run your built LabVIEW application.

OK, so we know LabVIEW will work on the three big OS’s, now we need to pick one. Windows and Mac OS X have licensing costs associated with them, so that rules them out as we want to keep the cost down. We are then left with Linux which is an open source language (open source meaning free).

Link: What Linux Distributions Do National Instruments Driver and Software Support?

Linux LogoLinux

Linux is a very diverse OS and has many different “flavours”.  All of these flavours are based on the Linux Kernel which are the brains, in essence it is the OS. The “flavours” of Linux such as, Ubuntu, Debian, Fedora, Red Hat and OpenSUSE are known as distributions. These distributions have a different look and feel when they sit on top of the Kernel.   As a user friendly interface that makes calls to the Linux Kernel, each distribution has its own unique implementation of calls to the Kernel.

In order for us to run LabVIEW on a Linux system we require a specific Kernel version and also a specific distribution that NI support.  NI sell a specific version of LabVIEW that is designed for Linux, unfortunately there is no cheap way of getting hold of LabVIEW, the Linux license cost just as much as a windows LabVIEW licence.  If you want to go cheap then program using a text based language.

Link: LabVIEW Professional Development System

OpenSUSEOpenSUSE Logo

The version I used in my project was LabVIEW Professional Development System (Linux) and OpenSUSE 13.1.  OpenSUSE 13.1 (KDE Desktop) has a very similar feel to Windows 7 and earlier (ie the good Windows) which makes using Linux intuitive. Before I used Linux I thought it was this black art that is all command line driven, and only the hard-core techys used. Yes, it still has that element, but all of the things you can do via command line you can do via a graphical interface.

You can find out more about OpenSUSE by visiting the OpenSUSE website.

The Platform

We have established an OS, now we need a platform that can run LabVIEW and Linux. Finding a platform that supports Linux is not difficult, due to the open source nature of Linux. There are many platforms which support Linux PCs, routers and mobile phone to list just a few. The tricky bit is to find cheap platforms that support LabVIEW. Recently, NI have released cRIO 903x which runs an NI Real Time (RT) version of Linux.  This is great, as it combines community driven Linux with LabVIEW and an embedded rugged form factor with an FPGA.  cRIOs don’t come cheap, but you do get a lot for your money.
Beagle Board LogoArduino LogoRaspberry Pi LogoSomething like a Raspberry PI, Arduino or Beagle would be a perfect platform as they are cheap to purchase and small. Unfortunately, they cannot run LabVIEW as LabVIEW requires an x86 processor architecture to run. National Instruments have a toolkit for the Arduino platform: LabVIEW Interface for Arduino. Also, recently, there is a way to write LabVIEW code and compile it to an Arduino.  We will talk more about Arduino and LabVIEW later on; note this compiler was not available at the time of my project.

Raspberry Pi

Raspberry Pi

Arduino Uno

Arduino Uno

Beagle Board

Beagle Board

 

 

 

 

 

 

The next best thing is a small PC. Small PCs are taking off with the Raspberry PI being the size of a credit card able to run Linux on an ARM processor. However I needed a small PC with an x86 architecture processor. I did what all the greats do… and began Googling. My search turned up a few embedded PCs, these were the type of “Kiosk” PC that are embedded in stands at trade shows, with I5 processors SSDs 4GB RAM, they ranged in price from £300 – £1000 which is pretty steep and miles away from the £15 of Raspberry PI or Arduino. I wanted to get the cost down as much as possible, finding the two top contenders to be the Intel Minnow Board and the Gizmo board.

Intel Minnow Board

Intel Minnow Board

Gizmo Board

Gizmo Board

Gizmo Board

Gizmo Board

 

 

 

 

 

 

Both boards cost around the $200 dollars (£135 ish) a piece, have a number of peripherals such as USB, MicroSD, Ethernet, SATA. They are both pretty small. Granted they are not as cheap as a Raspberry PI or Arduino but what they do have is fundamental to LabVIEW, an x86 processor.  The Minnow has an Intel Atom and the Gizmo has an AMD Dual Core processor, making them good candidates. Because there was not much between them, I chose the Gizmo.

When the Gizmo arrived it had a preinstalled OS.  I installed OpenSUSE 13.1  64 bit and NI LabVIEW for Linux 2014. National Instruments provide a step by step guide on how to install LabVIEW for Linux.

Link: Linux LabVIEW Installation Guide

Once installed, I was able to open LabVIEW and create VI’s as I wanted, just like I normally do on Windows.  However, as the Gizmo is no Intel I7, 8GB RAM SSD beast like my laptop, I found it a bit laggy. To get around this, I was able to develop LabVIEW projects on my Windows machine and then load them on the Gizmo (Linux Machine) and create a Linux application that would then be able to run standalone. I then took a look at the driver support and installed DAQmx and VISA – again the versions for Linux. Next, I connected an NI USB DAQ device (NI 6218) to the Gizmo and was able to communicate just fine. I then connected a Telegesis USB Zigbee Dongle, that appeared as a COM port on my Windows machine, exactly the same thing happens under Linux.

Now that I have this small, relatively cheap platform that is based on Linux that is able to run and build LabVIEW applications all I need is a project….

Arduino

I said I would revisit the Arduino again, I feel this needed its own section as I really like the Arduino platform. In my spare time I enjoy programming and have worked on a few hobbyist Arduino projects.

How can we use an Arduino with LabVIEW? Up until now the answer has been with the LabVIEW Interface for Arduino Tool Kit, which recently has been replaced by LINX.

This allows you to write a LabVIEW project running on a PC to connect to an Arduino and communicate to the Arduino from the LabVIEW VI, however the Arduino must be connected to the PC. The Arduino has some C code that you must compile down to the Arduino that interprets the messages the LabVIEW VI send it over the USB link.  Not very embedded as it requires the USB connection.

This year (2015 March) an NI Alliance Partner TSxperts have created a compiler that takes LabVIEW code and compiles it to Arduino code that can then be deployed down to the Arduino. This is a great idea, having the ability to purchase a low cost board and utilise the power of LabVIEW.

Personally I haven’t had a chance to get my hands on it yet but I am very excited to try this out for myself.

Conclusion

Hopefully this post has given you some insight into the options to run LabVIEW on a small and cheap PC. If you have any questions or would like to discuss anything please feel free to contact me, tweet the Austin Consultants team or leave a comment below. You can also visit my personal blog.

The post Small and cheap platform for NI LabVIEW appeared first on Austin Consultants.

Message 1 of 1
(5,094 Views)