Work Stops for Robot tended lathe?

[alek95]

Hi Everyone,

I don't have any experience in robot automated lathe production.

At my company we make thousands of the same parts every month. We have multiple lathes running 2 shifts per day. All loaded by hand.

These parts are round bars that are 24-48" long. Only the ends get machined.

Currently we bump the part against the turret to set the location.

The overall length of the part isn't super critical, but we need it to be about +/- .030

Can a robot load this accurately? Would you need a probe to update the work coordinate every time? Any recommendations are appreciated.

 

[Ox]

Bar feeder with pass through sub-spindle.


-------------

Think Snow Eh!
Ox

 

[alek95]

I don't think it would work because most of the material is tube, not solid.

Tube is a pain in the ass on a bar feeder because chips go down the bore and jam the bar feeder.

 

[Comatose]

Sure. Load the bar into the chuck, leave the jaws open, position your turret. Bump the bar into it with the robot. Close your vise.

Most robots have a reduced torque mode that will trip when they run into something.

 

[EmGo]

These parts are round bars that are 24-48" long. Only the ends get machined.

Double-end facing and centering machine, customized. I don't know if Fraser-Giddings & Lewis still makes those, Hey did also but if not, there's at least two or three places in china making similar. And robots work well with them. I'll snag a few photos if you like.

 

[alek95]

Double-end facing and centering machine, customized. I don't know if Fraser-Giddings & Lewis still makes those, Hey did also but if not, there's at least two or three places in china making similar. And robots work well with them. I'll snag a few photos if you like.

Very interesting. Please show me if you have photos.

 

[alek95]

Sure. Load the bar into the chuck, leave the jaws open, position your turret. Bump the bar into it with the robot. Close your vise.

Most robots have a reduced torque mode that will trip when they run into something.

Interesting. I assume that this would all be done with the door open, aka some type of override?

 

[EmGo]

Very interesting. Please show me if you have photos.

Somewhere. Okay, we're goin' in. If you don't hear from me for a week, it's cuz I got distracted.

 

[YoDoug]

I have used these in the past for work stop for long parts in a lathe that was robot loaded. Worked very well. Either have the robot push part against the stop or put a spring stop in the turret and push it into place with the machine.

Precision Machine Shop Minneapolis - DC Products - Accu-Stop

Precision Machine Shop. The Accu-Stop spindle stop provides a solid, precise stopping surface for work pieces. Contact us today at 763.428.3100.

accustop.net

 

[EmGo]

Very interesting. Please show me if you have photos.

Found videos. Now all I have to do is defeat this piece of shit browser and those assholes at google with their effing "security" garbage.

If you read in the paper that someone broke into google headquarters and took out about thirty people with an AK ? That'll be me.

 

[Digital Factory]

0.030" positioning accuracy is easy for a robot. The thing to watch for is that the parts don't slip in the gripper fingers.

If the machine has a sub, it doesn't matter. The machine can reposition the workpiece once it's in.

I don't think it would work because most of the material is tube, not solid.

Tube is a pain in the ass on a bar feeder because chips go down the bore and jam the bar feeder.

You could build a simple retractable air blast that mounts behind the drawtube.

Use a single acting air cylinder like this: https://www.mcmaster.com/6498K104/ (mounting brackets are available) and mount a nozzle to the end (bent copper tube). A single air source actuates the cylinder and blows air. Spring retract when the air turns off. The whole thing can be activated with two M-codes for on/off. You could even tap into an existing M-code chuck air blast.

You'll still need a robot to unload though. That said, robot unloads are significantly easier to implement than robot loads. Consider combining all of the above: barfeeder load, air blast during cycle, robot (cobot) unload. If you were running a single part number, this could be considered overkill, but as soon as you start running parts of different diameters and lengths, that notion goes out the window. Barfeeders adapt very well to different part sizes and are more or less fail safe. Robots can as well, but they require significantly more legwork to get there.

 

[EmGo]

Very interesting. Please show me if you have photos.

Searching the junkpi ... err, archive, takes a while, sorry.

If the parts you are describing are like what you are describing, doing this in a lathe is a loser, honest. They developed facing and centering machines for exactly this purpose and I can tell you from experience, going to a specialized end-working machine over trying to do this in a lathe is a joy and relief. The big names were G&L (they bought Frazier "endomatic" ?) but these tend to be large and heavy-duty, for like 6" bar 4' long and so on. Another is the #3 Hey, slightly different system, maybe more suitable for what you describe but also long gone and maybe hard to find. Both very automated, no human intervention required. I had a Seneca Falls single-ender that was perfect for my small, short parts (roughly 2-3" x 8-10", in general). Never seen another one of those.

Here's an idea of a few currently available - general idea

https://files.catbox.moe/lqsq30.mp4

robot loading, a pickanick mr ranger sir

https://files.catbox.moe/4bi58p.mp4

another overview but showing the two ends don't need to be the same. There's also at least one model with a turret arrangement like a robodrill on each end, so you can have quite a few operations per end, lots of variety out there

https://files.catbox.moe/gtq19p.mp4

Anyway, this type of thing is just way better for end-working or prepping shafts, instead of trying to do it in a regular lathe with stops or any of that other crap.

 

[alek95]

But can it do a 1 1/4 -12 tap?

That's a pretty big tap. And it needs to go to a depth of 3.25 inches.

 

[EmGo]

But can it do a 1 1/4 -12 tap?

That's a pretty big tap. And it needs to go to a depth of 3.25 inches.

be still my heart

https://files.catbox.moe/nuxeu0.mp4

A G&L with 15 hp heads probably wouldn't mind either. It's nice to have nc control tho, I think.

 

[DDoug]

"These parts are round bars that are 24-48" long. Only the ends get machined."

I don't think it would work because most of the material is tube, not solid.

Tube is a pain in the ass on a bar feeder because chips go down the bore and jam the bar feeder.

You said "BARS" and now you change your tune.
Enough.

 

[michiganbuck]

Robut lathe operators that hold a cell phone often should be fired..ops umplugged.

 

[nihilistic]

Just plug the end of the tubes and use the bar feeder. I'm sure one could find some little plastic reusable plug.

 

[Ox]

Just plug the end of the tubes and use the bar feeder. I'm sure one could find some little plastic reusable plug.

If this was an ongoing job, I would find a way to run it in a feeder, and highly likely w/o plugs.
No idea how they are getting birds nests down in there?
They must be dooing some inside boring, but maybe a tooling change could change nests to chips?

Pass through sub-spindle with parts coming out the back.
Twin spindle and you could run both ends at once.


----------------------

Think Snow Eh!
Ox

 

[Strostkovy]

Just plug the end of the tubes and use the bar feeder. I'm sure one could find some little plastic reusable plug.

You can get a four pack of silicone butt plugs on amazon for cheap and they'll plug just about any size hole.

 

[EmGo]

If this was an ongoing job, I would find a way to run it in a feeder, and highly likely w/o plugs.

And I still say I'd get a double end-working machine and kick your ass on price, quality, speed, and cost to manufacture. Likely by a factor of two or more