วีดีโอ

That's great to hear. Thanks for the feedback! :)

Great question. Most PLC inputs have some debounce filtering built-in, so it's rarely an issue. I can't think of a time in the last 10 years I've had to debounce in logic. That said, if you do find a case that needs debounce in the code, you'd have to do that one at a time for each input using timers.

Hi, the first go-to for reflectivity issues is usually a polarized filter. Machine vision usually does best with diffuse reflection of light. The shininess you're talking about is a specular reflection. Specular reflection maintains polarization while diffuse does not. So a polarized filter on the light source at 0 deg and another on the camera at 90 deg will mostly remove the shiny reflections from the image. Good luck!

Sorry for the slow reply. This video is available on our website. courses.breen-machine.com/courses/virtual-machines

I'm glad you enjoyed it! :) Thanks for the feedback. I'll try to get some more videos up soon.

@ 🙏🏼

I'm not sure what you mean. A one shot, by definition, stays on for one scan of the program. Ever part of the program will see the bit high exactly once before it goes low again. If you're talking about the pushbutton one shots (ie PassPB_OS), only one branch of the rung is active at a time, depending on the state of the one shot. That's how we want it to work.

Sorry for my slow response. :) The easiest and most reliable way to measure width is to setup the image so it can see both edges you're measuring between. If that's not possible, the camera will need to be able to infer something about the other edge. For example, if the other edge is always against a fixed guide, you know where it is. In that case, take a picture of a max size part and min size part, measure the edge location in pixels, then interpolate between those values to get an actual measurement. The interpolation will be linear (y = mx + b) and can be done in the camera or the controller (PLC, PC, or whatever is receiving the measurement). I usually trigger from the controller or based on a sensor that tells the system when the part is in place. If that's not convenient for the 1 photo per second requirement, you can also setup a trigger event in the spreadsheet using the Timer function. (support.cognex.com/docs/is_592/web/EN/ise/Content/Reference/Timer.htm)

Thanks for the feedback! :) It keeps me motivated to make more.

Thanks for watching! I think this is a little more than I can get into through TH-cam comments. The first thing to do is always check the manual. If that doesn't get you what you're looking for, I like to reach out to the supplier (wherever you bought the PLC and servo) because they usually have technical resources that are more personable than the manufacturer. If that doesn't work, then call the manufacturer tech support. And of course, if you'd like help from a friendly integrator, contact info is on our website: www.breen-machine.com/contact-us/ Good luck!

Excellent point, and well stated. Thanks!

I haven't tried connecting a camera to a C-More before. Usually, I see Cognex cameras being connected to Cognex Vision View displays or another web capable device.

It's a fair question, and I think I'd be asking it in your shoes. In my experience, most people in industry trying to do basic PLC programming (maintenance, tech school students, even controls engineers) don't have a compsci background. These videos are for them. It's been on my list for a while to make a quickstart PLC programming tutorial for compsci people... whenever I get some time. :)

Not true

Haha, I trust you. People on the internet are always very honest about the things they hate. ;) What do you prefer?

Changing the JOB on practically every Cognex camera takes a very long time (up to 10 seconds). The profinet communication frame itself is exceptionally complex and consists of numerous conditions. The cameras from the IS2800 series are a complete misunderstanding altogether. The software freezes, is not well-developed, and Cognex panels were not compatible (and probably still aren't). I definitely prefer using Keyence products (IV, IV2, IV3, and IX cameras depending on the needs). Due to all these issues with Cognex, I have developed an aversion :) @@breen-machine

Hi @rickrodriguez2477. Unfortunately, I don't have distribution rights for PCIS.exe, so I can't give it to you. You may be able to get it from your local Rockwell distributor. If there's something else I can do to help, let me know: www.breen-machine.com/contact-us/

The 2800 series is new and uses Vision Suite. I don't think it'll work with Explorer

Yes, most industrial cameras are designed for robot applications, and Cognex is a common choice. If you want to connect the camera directly to the robot, check with the robot manufacturer. They usually have one or two brands they work with. Otherwise, you can use a PLC as an intermediary. In this case, you'd send the results of an inspection to the PLC, then the PLC would load the appropriate registers in the robot, probably all over Ethernet.

Thanks for that. :)

The short story is, it depends. What kind of dimensional measurements? What precision? How big are the aluminum extrusions? Generally speaking, the larger the part, the less precision you'll get from a vision system. For example, if your vision sensor is 1000 pixels wide and your field of view (FOV) is 1m, each pixel is 1mm. If your FOV is 1cm, each pixel is 0.01mm. There are a few other details, like lens quality, etc, but the size of the pixels is usually the most important factor in measurement precision. This means if you have a large part and want high measurement precision, you'll need a very high resolution camera, and at some point it's not practical. Perspective and lighting can also be a challenge with measurement applications. If your parts are thin, a backlight works well. If your parts have crisp edges, you may also be able to top light with good results no matter how thick the part. If your part is round, you may need to use more specialized lighting like a collimated light source. If your part is rectangular with rounded edges, a 3D vision system or 2D laser profiler might be your best bet.

Good thought. I've seen that done before, but I personally don't like having more than one Step variable. I think it adds complexity and confuses the maintenance guys. I'd rather just make sure each sequence step is waiting for the appropriate result before moving to the next step. For example, if step 2 is extending a cylinder, it shouldn't move to step 3 until we see the cylinder switch shows it's been extended. It's a tradeoff either way. :)