My homemade bar puller for my CNC lathe.

[rgray]

I made this "bar puller" from a scrap piece of 1" exhaust tubing.
I still push the "go" button for every part. I haven't gotten brave enough to program it to continuously work.
Mostly because it's not super reliable. It's not spring steel so after some parts I had to adjust the diameter when it failed to pull the bar the entire distance.
I'm no programer so it took me some doing to get this to run.

Initially I programed a G1 for the puller to go to the bar as I thought that way I could control speed. That doesn't work as the spindle is stopped so tool is also stopped.

Then I fought the lathe just stopping at some point and I guess it was it didn't have a rotational speed command. This lathe has speed commands under turret 2 program so confusing sometimes. I put speed commands everywhere trying to get it to work and that rev right before the first bar pull seemed to be what made it cooperate. Odd as later it pulls the bar without that same "rev" beforehand.

I had to add a chamfer to the cut off at the end as the bar puller would move in and run into the bar and push it into the collet. (only did that once that was enough).

The view isn't great. my waterproof Gopro housing deteriorated in a matter of months inside the lathe cabinet. Guess it couldn't handle the coolant. So video was shot through the door window.

 

[JimDawson]

Is that one just spring loaded ''jaws'', or does it work off of turret coolant pressure? You can't feed in a tool with the spindle at zero speed?

I bought an Easy-Puller for mine, seems to work well.

 

[rgray]

Is that one just spring loaded ''jaws'', or does it work off of turret coolant pressure? You can't feed in a tool with the spindle at zero speed?

It's exhaust tubing from an old race snowmobile I had in the 90's.
Karl T said months ago (in your cnc thread) just build it from scrap. So that's what I did.

Two cuts across the end and squeezed down to have some grip on 1" bar stock.

The part has a chamfer so it just pushes on. Had to add the chamfer for the cut off piece to make that work.

Goes on not rotating. But when writing the program the lathe would just stop. No alarm no clue. So adding the speed command made the program continue on, the rev then M5 to stop spindle and then it continues on.

I'll have to mess with it and see if I can make it work without the rev up. Seems silly, but I'm happy it works, I spent all day yesterday figuring it out.

 

[rgray]

You can't feed in a tool with the spindle at zero speed?

Yes it will rapid a tool to position with the spindle stopped. I have had it do that before and stop and do nothing until I pressed the coolant button, then continue on and stop again waiting for me to release the coolant button.
Somehow when hitting the emergency stop button it can get the relay activation backwards or command backwards.
I learned this by chance. Lathe would stop, I'd look at the screen and M8 was next line, so I pushed the coolant button and off it goes (surprise).

Once the machine has been completely shut down it goes back to normal, but as a new user with my first cnc I spent some time figuring that one out.

Thought something like that was my stoppage when commanding the bar puller, but I don't think it was. I'm really just stumbling along here.

 

[JimDawson]

That is a bit strange. Sorry, I can't help you with this one, I've never seen a Mori before.

 

[rgray]

Sorry, I can't help you with this one, I've never seen a Mori before.

Fanuc 6TC from 1983. That probably doesn't help either, but I thought I'd throw it out there.

 

[JimDawson]

Fanuc 6TC from 1983. That probably doesn't help either, but I thought I'd throw it out there.

Nope, doesn't help much. That's a real antique.

 

[rgray]

I'm always amazed that it still works!!!

 

[rgray]

Little update. I swore the turret was moving away to fast outrunning the collet closing and pulling extra length bar out.
I slowed the video down and could see that that wasn't happening. Not in the piece in the video anyway.
Thing is that by chance was the last of the bar, so it was light.
After the video I put another 6 foot bar in and started again. The quick movement of the axis (start/stop) causes the bar to slide farther into the puller
making a longer part, and leaving a to long starting length for the facing op of the next cycle.(no sure of this???)

I fixed that with a few G4 dwell lines giving the collet time to clamp before the puller pulls away.
Not entirely sure my theory on that length variance is correct. Originally thought the puller was pulling more length as the turret speed out ran the collet closing, but then it should have been better with the full lenght bar as it would be heavier and harder to move, but it was then worse.
Leading me to think the inertia sliding it deeper into the puller. Sounds a little far fetched. I don't know.

Somehow the dwell make the length come out right very consistently (so far). It is a pullback collet so maybe the dwell makes it's clamping more consistent. Bar got pulled out .315 to far after first part was machined (a little much for the facing) I hit the E-stop, added the dwell codes and it has been good.

 

[JimDawson]

I have seen inertia change the position of the bar. If the puller does not grip the bar tight, that could be a problem.

You might slow down the pull cycle, I think mine is at 50 IPM. I also use a positioning stop in the turret which pushes the bar to where I want it before closing the collet. I final position this at 20 IPM so the inertia is not a factor.