Datalogic Smart Vision Sensor Configuration Part 1 of 2 - from AutomationDirect

In this video series, we are going to be looking at the configuration options for the Datalogic Smart Vision Sensor. The first video will address setting up the camera with the web interface. In the second video, I will take you through a real-world example of using the camera at a high rate of speed. To view the second video in this series click here. The Datalogic smart vision sensor has two ways it can be set up. It has a yellow teach button on the top of the case, or you can connect the sensor to your PC via ethernet. While you can set this camera up with just the teach button, there are numerous settings on the camera that need to be tailored for each application. Also, the only way to see what the camera is seeing is to use the browser interface. I would strongly recommend for initial setup, that you connect the camera to your computer. Let’s return the vision sensor to factory default settings so that we are starting from a neutral point. To do this I will turn off the circuit breaker supplying power to the camera. With the camera powered down, I will then press and hold the yellow teach button and turn the breaker back on. Upon power up the LEDs will start flashing. I will then release the teach button and immediately press it again once. That beep denotes the sensor has been reset to factory defaults. We know that the vision sensor is in a neutral state ready to be taught because the first LED flashes steadily. It is very important that you do not continue to press the teach button after the LEDs start flashing at power up. If you hold the button longer than 4 seconds upon power up, the camera will roll back to the previous revision of firmware. If this does occur, you can find the latest version of firmware at AutomationDirect.com to get you back up and running in no time! Now that the vision sensor has been set to default, we are ready to teach the camera good and bad images. For our setup, I have put plastic clips onto a conveyor belt to represent parts. A good part is a yellow clip. A bad part is a white clip. It is important to note that this camera is not a color-sensing camera, but if there is enough contrast between parts, the camera can detect it. As a starting point, we will go in the software and change the IP address of the camera so that it resides on the same subnet as the rest of our network. The default IP address of these cameras is 192.168.3.100. The rest of the devices on my test setup are on the 192.168.27.XXX subnet, so we need to move the camera to that subnet. This video isn’t about setting your computer’s IP address, so I will assume you know how to do that. Once you have set your PC to the camera’s default subnet, you can go to your web browser and type in the camera’s default IP address of 192.168.3.100. If your camera’s ethernet cable is properly connected to your network or PC, it will pull up this landing page. You can then set the camera’s IP address to something compatible with your existing network. For our setup, we will use 192.168.27.100. I will leave the subnet mask alone as that is a good value, and I've left the optional Gateway address blank because I have no need for the vision sensor to communicate with anything outside of the current network. Once we hit the save button it will change the camera’s IP address and the camera will no longer be accessible. We then will need to change our PC's IP address back to the 192.168.27.XXX subnet. Once we have done this, we will now be able to access the camera at 192.168.27.100. To start our setup we will select the new job button. I will name this job Camera Test 1. There are 32 different save slots, called banks, in the camera. You can run and save up to 32 individual programs and switch between them using the browser interface. We will store our new program into Bank 0. The camera then brings up a live shot of the image it is currently seeing. We will move a good part in front of the camera. You can select the start automatic setup button to have the camera automatically evaluate the image and optimize the exposure settings. If you prefer to do this yourself or tweak the camera exposure and focus settings you can do this by twirling down the advanced image settings section. Focus distance changes the focal point distance from the face of the camera. Exposure time changes how long the camera captures the image. Because it is in micro-seconds and the camera requires 50 milliseconds to evaluate, you will not see significant increase or decrease in cycle times regardless of this setting. Finally, sensor gain changes how much exposure the camera adds to the image. We can then select next. We have now entered the good teach portion of the process. The camera can store a maximum of 5 images of good or no-good parts and 6 total images in any combination. For our sensor let’s do 2 images of good, and 2 images of no-good. We already have the first good part under the camera so let’s teach this image as good. The camera captures the image and places it under the good box to let us know that it will be one of the good images used for comparison moving forward. Let’s index the belt one position manually. We now have a second good part that we will teach the camera with. Again, the new image has been added to the good image database. Now let’s move onto teaching the camera the no-good parts. We will click next to move to the no-good teaching portion of the process. Let’s index the belt again to get a bad part in front of the camera. We will then teach the camera this as a no-good part. Let’s index again and teach the camera a second no-good image. Now we have given the camera a few examples of good and bad parts and are ready for a test. To finalize these images, we will select Train. We also need to account for the camera’s I/O settings. I have this camera connected to an input card from a P1000 PLC wired for sourcing or PNP. As we can see from I/O status on our digital input card, the PLC inputs are showing on when they should be off. To address this, we need to go to the I/O settings. In the output mode section, we can see that the camera can be set to NPN, PNP, push-pull active high, or push-pull active low. Since I have a PNP input card, I can select either NPN, or push-pull active low for the camera. If I had an NPN input card I could select PNP or push-pull active high. I also want to adjust the output hold time to the maximum setting of 45 milliseconds. This will give my PLC input card plenty of time to see the camera’s outputs coming on. At this point we would be ready for some cycling, which we will get into in the next video. As you can see, the browser interface for the camera is easy to use and intuitive. It makes setting up a vision application very straightforward, and if you have a good feel for your process and the potential outcomes of that process, you should have no issues setting up your own vision application. For further questions, access our award winning technical support here. To see more information on identification products from AutomationDirect, click here. To subscribe to our Youtube channel click here!