LabVIEW for Dummiees
Section 3

Back to the Section 2

Attribute Nodes

Making Works of Art

Initalize Array A[1000]=0; Array B[1000]=0; number=0;
While (ON)
{
GET point(x,y);
if(Start)
   {
    A[number]=x; B[number]=y;
    number=number+1;
    //Insert the cursor values into the arrays   
   }
else
   {if((NOT Start) AND (NOT Pause))
      {
       A[1000]=0; B[1000]=0;
       number=0;
       //reinitalize to default for mistakes
      }
    if((NOT Start) AND Pause)
	//just pass data through}
WAIT 100 ms;
}

Well I could go through each of the Attribute Nodes, but I would be typing until the next millennium. But just use these facts that any control or indicator can have them, just choose Create >> Attribute Node and they can control or read the properties, or attributes of the Front Panel Icon. The names are fairly descriptive and can alter the graphing properties, foreground and background color, sizes of icons and text, to make an icon visible or blink. Play around with them. If anyone completely confuses you, I will be happy to add it.

Creating Sub-VIs

Imagine that you have just written a very complicated program. You feel really proud and run it, but it doesn't work quite right. When you look over the wire diagram you realize you missed an important complicated calculation. &#@*!! There is no way to fit it in the diagram, and if you tried, probably more mistakes would be created. Or you need a piece of code for several programs you are developing, and you don't feel like rewriting it every time. In text based programs, you could just type or copy the code in anywhere. LabVIEW uses sub-VIs which are sub-routines for functions you may use many times. In fact most of the functions from file on down are sub-VIs. All sub-VIs are squares, but not all squares are sub-VIs. An easy way of determining this is by double clicking on the node. If a new VI opens, it is a sub-VI. Sub-VIs are completely self sufficient programs, i.e. they can run on their own. The controls are the inputs and the indicators are the outputs.
How to make a sub-VI:

1. Write your subroutine with the controls as the inputs and the indicators as the outputs
2. Be sure to label the controls and indicators with descriptive and appropriate labels
3. The little icon in the upper right hand corner , is where to make the node.
4. By right clicking on the icon and choosing Edit Icon..., you can put a short description of what the sub-VI does. This is what you will see on the diagram when you use it. If you are an artist, color it. When you are finished, click on OK.
5. Make sure that you are on the front panel. Right Click again and choose Show Terminals.
6. The square is automatically divided for the number of indicators and controls. If you don't like it, you can choose a new pattern, add or delete terminals, or rotate the terminals. Usually LabVIEW has a good idea of what to give. Just be warned that the maximum number of terminals is 28. I never needed that many, but if you do, there are two ways around it. Use more than one sub-VI by breaking the subroutine in half, or by bundling the elements (See Clusters).
7. Get the wire spool. Try to keep the inputs on the left and the outputs on the right. Click on the terminal, then click on the indicator or control, or vise versa. The terminal will appear in the color of the wire. Keep repeating for the rest of the terminals, but not everyone has to be used. Also, a sub-VI need not have any outputs, or any inputs.
8. If a connection is a vital necessity, choose This connection is.... Useful if others will be using the sub-VI.
9. SAVE IT
10. Go to your main program, and by using the functions pallet's very bottom icon (Select a VI...), choose your sub-VI. By passing the wire spool over the sub-VI you will see the connectors appear. Now you can connect. You can also insert or replace a sub-VI.

Libraries

User Libraries

Data Acquisition

Before I tell you about how to program for Data Acquisition, you need to know how to connect the hardware. The data acquisition (DAQ) device we have here is a PCI-MIO-16E-1. The configuration we used is (Under the NI-DAQ Configuration Utilities
<-10.0V - +10.0V><Referenced Single Ended><Bipolar><None><Disabled>
If you have no idea how to use it, here is how the first electronic nose team did it. When the DAQ board came we unwrapped it carefully. There is a board you have to insert into the computer. We put it into a slot where it fit (I'm not a computer engineer). Turned the computer on and it worked! Afterwards we waited for a month for the cable and the DAQ connection board. The company said we didn't specify it when we ordered. Make sure all the software is installed. Find a friend and a multimeter. Just in case you can't find a multimeter, you can use a LED (light emitting diode). Be sure to attach a resistor or it will burn out! Go to the NI-DAQ for Windows, and view the NI-DAQ Test Panels.

Find the DIO0 and DGND pins (on my board 52 and 53), place your two multimeter probes on the "screws". You should register +5 V. If it is negative, you have the probes backwards. Once you have reached the test panels, go to the Digital I/O where there is a long row of circles. Move your mouse to the second row left hand side where there should be a clear circle above the 0. CLICK! If you did this right, the voltage should reduce to ZERO. Find the other Digital I/O pins on the DAQ board and click on the corresponding circles. This can get exciting when it works for the first time. By connecting a MOSFET to the board (not the DAQ board, your own circuit of course), you can operate a solenoid switch.

Find the DAC0OUT and the AOGND pins (on my board 22 and 55), and place the multimeter on the "screws". You should register 0 V. Go to Analog Output on the Test Panels. Output a DC Voltage anywhere between -10 V and 10 V. The multimeter should register the voltage being put out. If you are using a LED, the intensity will change.

Find the ACH0 and AIGND pins (on my board 68 and 67), get a 1.5 V battery (if you can, a calculator battery is fine), and two wires. Make sure the battery has some life yet! Go to the Analog Input Panel. The chart should be nearly constant, there is always some noise. If there is some drift, don't worry, you just need to connect it to a ground. Connect your battery terminals to the two pins and voltage should jump to +1.5 Volts, or whatever the battery voltage is. If it's negative, your backwards. Use the Multimeter to compair the two voltages. The agreement is usually very good.

If you need more specific info on the technical specifications see the user's manual that comes with the DAQ board. Now that you are familiar with the location of the pins and what they do, I will explain what the electronic nose group used the DAQ board for.

• Analog Input was used to detect voltage changes in the electrochemical sensors due to the reaction caused by the gas. Due to the relativly slow response of the sensors, 100 points were averaged over one second intervals. Also op-amps and potientiostats were used to amplify and filter the signals. If you want to know more about Electrochemical Sensors, See this site
• The Digital Trigger was used for precision control of solenoid switches. The use of a MOSFET (Metal Oxide Silicon Field Effect Transistor) used the voltage difference to turn a switch on and off.
• Analog Output was used to output a constant voltage between 0 and +2 Volts to change the temperature of a filament

1. AI Sample Channels.vi - You will need the device number which can be determined when you open the NI-DAQ Test Panels. The channel numbers which are written in a string comma delimited. The high and low limit don't have to be wired, they use the values set down in the NI-DAQ Configuration Utility (-10 V to +10 V). The output (Samples) is a 1-D array. Connect it to a Build Array to make a 2-D array which defines the values as different plots. Finally connect this to a Waveform Chart. Make sure to have all this in a While Loop.
2. AO Update Channel.vi - Three simple things
   a) Device - same as above
   b) Channel Number - a string 0 or 1
   c) Value - How many volts between -10 V and 10 V.
   That's all there is too it.
3. Run_me.llb/Getting Started Digital I/O.vi - This is what I used for the control of the switches. But I had to alter it. Remember to save it under a different name!! The program uses a cluster to control the ON/OFF. Just run it to see if it works. Look at the diagram to see what is happening.
   A cluster of 8 Boolean Switches moves to the Cluster to Array which moves to Boolean Array to Number and so on.
   You could just use the cluster as a manual control, but it becomes more difficult if you want to control from a file. I deleted the Custer and Cluster to Array, and made a control going to the Boolean Array to Number. Now enlarge the array to 8 elements and try it. It is as good as the original. Try it to make sure. Now reconnect the terminals, so you can have your sub-VI. I didn't bother altering the rest. In your control file you need to have interger numbers. If you are just going to be using one switch, use 0 and 1 for on and off. If you need to control more switches, refer to Boolean Section to see how LabVIEW converts from number to boolean array. Connect the control to a Number to Boolean Array. You may be thinking, why didn't I just use the number as the control since it will be converted back anyway? Well, this way you can see the boolean control, and will know which switch is on.
   When you connect the program in your main diagram, you may have a problem. The first time we tried it, it kept switching on and off between each iteration. I made the "Port Configure" a control in the main program and made "configure the port for reading or writing" the default. I just had to hide it behind a graph, and it worked fine. Sometimes if it isn't working right try anything.

As a good exercise, I suggest you look at a recent issue of Scientific American in The Amateur Scientist section. There are some inexpensive and interesting projects where a DAQ system could be useful.

Analysis

Other Stuff

Clusters

1. Bundle - Used to combine any combination of your needed wire types. Just enlarge and attach as needed. It is a good idea to have a different descriptive label name for each of the items. If you don't have them coming from a direct control or constant, Create Indicator and name it. As a starting point the Source Element doesn't need to be wired. To replace an element in the cluster, wire your original cluster to the Source Element, make sure the Bundle size is the same as the original, and only add an input to the element you want changed.
2. Unbundle - Quite simply, it undoes the bundle operation. All you have to do is make sure that the size is the same, and then just wire them to indicators or the rest of the diagram.
3. Bundle By Name - If you have a large bundle and just want to replace one or two of the elements in a cluster, you can tell the VI to replace only the element with a particular name. This is why I told you to label the inputs before! Wire the cluster, then Select Item >> "Whatever", and wire your new value. This is analogous to the Replace Array.
4. Unbundle By Name - Can you guess? Same thing as above, just the item will be an output.
5. Cluster to Array & Array to Cluster - They do what they sound like. However, it only works for elements individually clustered, that is only individual numeric and Boolean.
6. Here is an example of the bundling operations for clusters. Reproduce them and try your own.

List & Ring

Advanced

Application Control

Communication

Instrument I/O

Acknowledgments