Need help with my Automation

[Tbarber]

I've been machining for about 12 years, worked with lots of different machines mostly mills, 3-5 axis. But now I'm at at shop that I am trying to figure out the best way to run my parts lights out. Most of my production runs are 2,000 - 5,000 pieces orders ( the part is a small aluminum .250" x 1.5"). So I made a fixture to hold 14 in .375 bars of material and mighty bites. And it has been working great. But I don't know the best way to automate this. Either change my pallet style and grip it on the sides or the center or change away from the fixture process all together. Since I have so many small parts I feel like it's not worth having the robot pick up each part individually out of the fixture. Instead swap the fixture out for one that's already prepped... Sorry for the long post but it's just a lot going on.

Any suggestions would be amazing!

 

[Matt@RFR]

So much missing information.
- What machine?
- Do you already have the robot? Have you considered a spindle gripper instead?
- How many ops are the parts?
- Do you want to make one op at a time? Or have a complete part come out every cycle?
- Consider that every automated cycle that I'm aware of is single-part flow
- Attach a picture of the parts, or a representative model missing info that you can't give out
- What is your current cycle time?
- Any tight location tolerances to worry about?
- Do you have enough time on these parts that tool life is a known?
- How long do you want the machine to run after you leave? Do you have the space for that many blanks?
- Do you have enough space for that many completed parts?
- Do you have a way to alarm machine / robot if coolant goes too low? If shop air goes too low? Broken tools?
- Do you have spindle probes to check for in-location blanks / parts, blank size, part loaded correctly?

 

[Tbarber]

So much missing information.
- What machine?
- Do you already have the robot? Have you considered a spindle gripper instead?
- How many ops are the parts?
- Do you want to make one op at a time? Or have a complete part come out every cycle?
- Consider that every automated cycle that I'm aware of is single-part flow
- Attach a picture of the parts, or a representative model missing info that you can't give out
- What is your current cycle time?
- Any tight location tolerances to worry about?
- Do you have enough time on these parts that tool life is a known?
- How long do you want the machine to run after you leave? Do you have the space for that many blanks?
- Do you have enough space for that many completed parts?
- Do you have a way to alarm machine / robot if coolant goes too low? If shop air goes too low? Broken tools?
- Do you have spindle probes to check for in-location blanks / parts, blank size, part loaded correctly?

Yes sorry im running the parts in 2 ops, op 2 is being probed for location. I also do tool break detection after all drills and taps and tool life management. Coolant automatically refill. We are getting a UR10E and was thinking about going with Pierson pro pallets system and just having the robot swap in and out full pallets loaded with parts but I feel like there's a better option

 

[DavidScott]

If you still need a person to swap parts in fixtures that isn't very automated.

 

[Tbarber]

If you still need a person to swap parts in fixtures that isn't very automated.

If you still need a person to swap parts in fixtures that isn't very automated.

The fixture I'm running the parts on is running 4 rows of Op 1 And 4 rows of Op 2. So every cycle I'm getting ~64 completed parts off and just flipping the op one over to op 2,loading a new bar in op 1 and clamping. If I continued with this method with a robot I would have to make 3-5 of the same fixture and stage them all up.

So much missing information.
- What machine?
- Do you already have the robot? Have you considered a spindle gripper instead?
- How many ops are the parts?
- Do you want to make one op at a time? Or have a complete part come out every cycle?
- Consider that every automated cycle that I'm aware of is single-part flow
- Attach a picture of the parts, or a representative model missing info that you can't give out
- What is your current cycle time?
- Any tight location tolerances to worry about?
- Do you have enough time on these parts that tool life is a known?
- How long do you want the machine to run after you leave? Do you have the space for that many blanks?
- Do you have enough space for that many completed parts?
- Do you have a way to alarm machine / robot if coolant goes too low? If shop air goes too low? Broken tools?
- Do you have spindle probes to check for in-location blanks / parts, blank size, part loaded correctly?

The fixture I'm running the parts on is running 4 rows of Op 1 And 4 rows of Op 2. So every cycle I'm getting ~64 completed parts off and just flipping the op one over to op 2,loading a new bar in op 1 and clamping. If I continued with this method with a robot I would have to make 3-5 of the same fixture and stage them all up, so the robot would pick up the entire pallet and place it into the pneumatic pallet system but if anyone has any better ideas I would love to hear them... I just haven't seen anyone else automate parts. and as I said earlier The vice is setup as 3-4 rows of op 1 and op 2. Op 2parts are finished and removed and op 1 flipped over to
op 2 and new bar is added to OP 1.

 

[DavidScott]

I use wedge clamps, like Miteebite Uniforce but better, so I can hold 2 parts of your size with one clamp and ONE screw. With parts of that size my mill makes them faster than I can swap parts using pneumatic screwdrivers so there is no faster way to do it and use screws. This is why I say if you are still fixturing small parts manually that is a lot of labor that I would want to avoid if I had a robot, which I see as my next investment.

 

[Tbarber]

I use wedge clamps, like Miteebite Uniforce but better, so I can hold 2 parts of your size with one clamp and ONE screw. With parts of that size my mill makes them faster than I can swap parts using pneumatic screwdrivers so there is no faster way to do it and use screws. This is why I say if you are still fixturing small parts manually that is a lot of labor that I would want to avoid if I had a robot, which I see as my next investment.

Thanks for your input. Yea, the whole having to stage a 3 -5 fixtures before leaving every night is something I'd like to avoid. I feel like my only other option is to just use 2 numatic dual clamping vices instead of the fixture. I would just be getting a lot less parts out each cycle but hopefully less hassle and room for error.

 

[Overland]

Look at WheelieKings shop thread. I know it's several hundred posts, but he went through a big automation journey, with lots of vices, pallets and automation.
You may get some ideas, and benefit from some real life experiences.
Bob

 

[Overland]

Here's Wheelies thread. Several thousand posts.
Bob

B&A Precision

Figured I might as well start a thread to share some of my goofy ramblings, and the opening of the new shop. Ive just about got everything lined up. Its a small industrial suite. 1350sq-ft. 250amps of 3-phase. large door. small office. and really crappy concrete. Its in a good part of town to...

www.practicalmachinist.com

 

[Tbarber]

Here's Wheelies thread. Several thousand posts.
Bob

B&A Precision

Figured I might as well start a thread to share some of my goofy ramblings, and the opening of the new shop. Ive just about got everything lined up. Its a small industrial suite. 1350sq-ft. 250amps of 3-phase. large door. small office. and really crappy concrete. Its in a good part of town to...

www.practicalmachinist.com

http://Awesome thank you, I appreciate any tips I can get at the moment.

 

[Orange Vise]

We are getting a UR10E

We have a UR10e. Works well for individual part handling.

If you're looking at pallet handling, consider a 20-30kg cobot like the Yaskawa HC20/30. Weight adds up quick and 10kg really isn't enough. The heavier robots are only incrementally more expensive, e.g. 40% cost premium with 300% payload.

Pallet handling with a cobot is dead simple, as cobots are essentially crash proof. All of the automation I/O and actuation of ZPS/vises/autodoor can be handled by the robot, so integration into your machine is optional.

 

[Orange Vise]

Yea, the whole having to stage a 3 -5 fixtures before leaving every night is something I'd like to avoid.

You could hire a part timer or an intern to load/unload fixtures. It's straight brute force labor, very easy to train, and is cost effective.

We've done it all. Pallets/tombstones with varying degrees of complexity, part handling with a robot, part handling with a spindle gripper, and prismatic parts on a barfed+SY lathe.

Pallet handling is the easiest way to get into automation by a long shot because the loading routine is always the same, irrespective of part size/weight/shape/qty.

 

[Comatose]

That looks like a lathe part to me. Set it up and then let the bar feeder handle all 5000.