Видео

Working on it!

So, there's a couple things there. Is it the best small shop robot? I would say "Depends on your budget". Under $20k, it's pretty decent but there are definitely other options as well. For ease of use, a cobot like a UR or Fanuc is probably easier and safer to set up albeit at a higher price point. A UR10 is probably the gold standard and can lift 25lbs but it will be more expensive. As far as pallets go, I love the Pierson system (I have one here) but it probably wouldn't be a system I'd try to load with a robot now that I've used it. It requires very precise alignment of the pins onto the base and sometimes mine is hard enough to align as a human let alone for a robot. There are pallet systems from Schunk and some of the others specifically designed to be easy for robots and automation systems to align. Those are probably worth looking at. A Tormach ZA-6 trying to load Pierson pallets would probably be a very challenging system to tune to be reliable.

​@@bigtchopshop I wasnt sure about the positional accuracy needed for the Pierson pallets so thank you. Thank you so much for the Schunk reference too, I'm going to look into those. Some times I search everywhere online for specific solutions and just can't find anything, I appreciate the help.

@@kellykramer2529Yes. It needs to be quite precise for the Pierson pallets to line up. The Schunk stuff isn't cheap, but you can see where they are going with it - basically, there are features on the zero point systems and pallets that are tapered and/or mechanically guide the pallet into the right alignment as the automation system places it. This allows you to have a little bit of slop on the loading and still be able to get the pallet loaded successfully. In my experience, this will save you a lot of headaches. The robot arms are very accurate, but it's a long swing, and there are variables loading and unloading and how you pick up the pallet, and how firm the surface is that the robot is bolted onto, and so forth. Schunk isn't the only one making such things for sure, but a good reference. For example - schunk.com/ca/en/workpiece-clamping-technology/quick-change-pallet-systems/automated-machine-loading/c/PUB_8474. Feel free to reach out via bigtchopshop.com if you want to discuss this in more detail.

Hey Luke! Good question. There is a master GCode file that calls two subroutines. The first subroutine is whatever the current machining program is and the second is a program that runs to clean the vise in between runs. That just keeps looping on the Haas control. There is a MCode in that script that tells the robot controller whenever the CNC cycle is done. It sets a relay high when called and waits for the robot to finish and then send a signal back. So, it’s pretty simple - just runs in a big loop like that. The robot code is set to increment through all the loading positions I specify and then it waits with a message after the last one so that I can swap in another set of material. Tried to keep it as simple as possible - not a lot of code actually.

Thanks! Yeah. I found it mostly got rid of the coolant, so that was great. Especially parts like that with a pocket. They get real messy if you just lift them out of the machine full of coolant!

The program always puts the table back to a known position before signalling the robot to run its cycle. I have to guarantee that for it to work.

Agreed. I'm working on firming up that base!

It’s alive! Check out the newest video. Finally up and running.

Great question. I do still have the Tormach ZA-6 as well. There were two main reasons I switched over to the UR. 1) Safety - Although the Tormach does stop on a collision, that threshold is high enough that if it collided with my machines it would cause damage. The UR can bump the machine and shut down safely without huge concerns. For a production machine, I was leery to cause any damage that might take me offline. 2) Free Drive / Programming - Being able to free drive the cobot for programming is night and day over having to explicitly jog the industrial robot using the controls. I’m going to create a video of that. That said, I still have plans for the ZA6 to try create a simple pallet changer. So that will be on my to-do list for when things are running smooth here.

Yes. I was running a wireless probe that was held in the ATC.

Totally agree. I am actually working on a system for loading pallets onto a zero point fixture next so that I can vary up the jobs and the workholding a bit. The challenge there is that there isn’t great off the shelf scheduling software for running different jobs on different pallets. I’m just trying to figure out how to manage all that.

Exactly! In my case, I got my hands on some UR3’s that a local company was selling off and I got to thinking of ways to put them to use. The 3kg payload on a UR3 is more than enough for most of the parts I make, but the reach not so much. This seemed like a worthwhile experiment and it worked well. I am just getting the vise ready to go into one of my TM’s and then I’ll make a video about the whole system. I found the UR3 was fine at opening the doors as well, so that was a nice bonus.

Thanks for the kind words! It’s great to hear that it resonates and you find the robot interesting. I wish we could get all the robot fans together for a hangout, but RUclips is the next best thing.

Thanks! Yeah, it was great to finally make some parts with it.

Check it out in the latest video - I have it tending my CNC. CNC Mill Automation With the Tormach ZA6 Robot and VersaBuilt Multigrip System - Episode 3 ruclips.net/video/447gZ9MdeJY/видео.html

Thanks! I've actually got it doing a lot more these days. I am just about to record the first run if it actually making parts and swapping stuff out on the mill. Stay tuned - hope to have that out in the next couple weeks.

If you're looking for high precision, I would say this approach probably wouldn't be the best. The data that you'll get from it is still pretty noisy, and good for a rough approximation, or finding curvature, etc. But, you're sort of limited by the resolution of your probe, and diameter of the tip. If you were going to try get a higher precision probing cycle, I'd probably start by trying to find a probe tip with a really small diameter, and then work from there. The probes themselves should all be reasonably accurate (they need to be for the machine probing to work).

Also, I need gcode that will probe a 3" x 22" x 1" area. That will cover my necks completely. I will need to make it smaller and larger depending on the neck length but you get the idea.

@@dunawayguitars7494 Yep. Have you considered 3D scanning as an alternate approach?

@@bigtchopshop I looked at hand scanners but they didn't give me the accuracy I wanted. I'm open to ideas though.

Thanks so much. Glad you liked it.

Thanks so much. Working on it!

You're very welcome. Thanks for watching!

Nothing off the shelf currently, but I see no reason you couldn't make or adapt any of the ones from other robots. The ZA6 is a pretty open architecture in terms of being able to add in your own components.

Thanks Al! Been a pleasure to work with your hardware as well. I'm hoping to get the next episodes up here soon, so will have more of the Tormach side and programming coming up.

Thanks Alan. You bet. Spoiler - it's getting there! I just need to cut the video together!

So do we!

Glad you enjoyed it

Thanks Jeff! This is definitely a step up, and a lot of fun.

Thank you! Working on installing the tending system right now!

You got it! Thanks for watching.

Thanks for checking it out Arthur!

Thanks Nick. Was fun to pull together. Learned a lot. Next one in the series will be even better!

You’re very welcome. Thanks for watching!

Great editing too. :)