What's new
What's new

Our Fanuc Industrial Robot Loader with Haas VF-2SS Video

wrustle

Titanium
Joined
Jun 8, 2006
Location
Massachusetts
Had some time to take a quick 4k video of our Fanuc Robot with Haas VF-2SS.



We had the system installed last August and have become pretty comfortable with the whole system now.

After running a variety of different sized parts I have to say the only real weakness we have a concern with are the robot grippers. They are just very basic in their design and I wish they would grab onto the parts with a little more force or contact surface than they currently do.

The grippers are basically just four 5/16-18 threaded studs that screw into the plate that's mounted on the sides of the pneumatic gripper assembly. The "gripper" parts of those studs is just a smooth turned straight diameter.

This makes the contact surface of the gripper on the sides of the blanks very small, and quite easy to move the part in the grippers when closed onto the part. This can be a very big concern when moving the parts around held by the robot. You really have to be aware of the acceleration and deceleration when programming the movements, and especially for any rotational moves if using a two gripper system (one piece out, one piece in), versus a one gripper (one piece out-drop off-get new one-load in vise).

We have had instances of parts only being partially clamped in the vise due to blank movement upon entry to the vise, and subsequently ripped out in very short order since we are always using high speed machining toolpaths. There is a special tool (plunger) that seats the blank in the vise after being loaded, but if the part is not seated flat and the serrated jaws close on it, the jaws close on the radius of the extruded corner and blow material out the bottom of the blank so when the plunger goes to seat it, it is now sitting on a big chunk of aluminum on the bottom and once again the serrated jaws are only clamping on a partial surface. Once that endmill or face mill comes in contact with that surface sticking up, it's ripped out in short order!

Fortunately, there is always someone close by in the shop to go over and stop the cycle, but for this reason only, we have not been able to run lights out yet until we come with a new design and make some better grippers for our applications.

For now though, "WALL-E" (as he's been affectionately named) has made a significant impact in our shop. He runs all day long, through breaks, and lunch, and chews through the workload without a single complaint! :D

The parts you will see in the video run for approx. 60 minutes before someone needs to go over and remove the finished parts from the loading fixture (a new design we came up with and made in house) and load with new blanks which takes only a couple minutes to do.

Then WALL-E goes back to work all by himself for the next hour.

Life is good! ;)

Later,
Russ
 
Last edited:
You need a sensor in your vise, around $200 last I heard. If the part isn't where it should be the machine won't try to machine it.
 
The grippers are basically just four 5/16-18 threaded studs that screw into the plate that's mounted on the sides of the pneumatic gripper assembly. The "gripper" parts of those studs is just a smooth turned straight diameter.

This makes the contact surface of the gripper on the sides of the blanks very small, and quite easy to move the part in the grippers when closed onto the part. This can be a very big concern when moving the parts around held by the robot. You really have to be aware of the acceleration and deceleration when programming the movements, and especially for any rotational moves if using a two gripper system (one piece out, one piece in), versus a one gripper (one piece out-drop off-get new one-load in vise).

We have had instances of parts only being partially clamped in the vise due to blank movement upon entry to the vise, and subsequently ripped out in very short order since we are always using high speed machining toolpaths. There is a special tool (plunger) that seats the blank in the vise after being loaded, but if the part is not seated flat and the serrated jaws close on it, the jaws close on the radius of the extruded corner and blow material out the bottom of the blank so when the plunger goes to seat it, it is now sitting on a big chunk of aluminum on the bottom and once again the serrated jaws are only clamping on a partial surface. Once that endmill or face mill comes in contact with that surface sticking up, it's ripped out in short order!

Fortunately, there is always someone close by in the shop to go over and stop the cycle, but for this reason only, we have not been able to run lights out yet until we come with a new design and make some better grippers for our applications.
Not an uncommon issue and not isolated to Haas' system.

Easiest solution in your case is to probe the workpiece, which IMO should be done in every automated milling application anyway. Z-probe the four corners of the blank and you'll know right away if one corner is sitting high. If it is, jump to M99 and move on to the next workpiece. You be running lights out starting Monday.

Glad to see the automation is starting to pay off for you guys.
 
Last edited:
Not an uncommon issue and not isolated to Haas' system.

Easiest solution in your case is to probe the workpiece, which IMO should be done in every automated milling application anyway. Z-probe the four corners of the blank and you'll know right away if one corner is sitting high. If it is, jump to M99 and move on to the next workpiece. You be running lights out starting Monday.

Glad to see the automation is starting to pay off for you guys.


Great advice! Thank you! Will talk with my son out on the floor about this.

The automation is paying off. It is however a double edged sword. On one hand the throughput on the shop floor has increased quite a bit, but on the downside, it was at the expense of our co-op student who started a year ago February and stayed on full time in May. The robot eventually pushed him out the door the end of November. Despite the normal workload for 3 guys out on the floor, with the robot becoming more and more productive, there just wasn't enough work at a consistent level to keep four guys on the payroll. He was a very good employee as well, that's what hurt the most. On the upside I did talk with him a few weeks after he was let go and he had already found another job starting the beginning of January.

Later,
Russ
 
Great video! How does the co-bot and the vise communicate? Specifically, what tells the vise to close? Is the cart setup a standard thing? We are looking into a ro/co-bot and I had concerns about tipping over if the unit isn't bolted fast to the floor....very cool stuff!!
 
Is this a Haas component add on you're talking about? I've never heard of that, but sounds like it would be beneficial.
No, I think it was something BrotherFrank spoke of, and perhaps Lott or Comotose used them. It was just a small proximity? sensor that you would nest in a pocket in your vise jaw, or anywhere else you wanted to confirm a part position.
 
That's awesome and looks like you guys did a great job!

Besides a part sensor, after the part is picked out of the vise, I would also add a robot stroke with air blast before the stock is placed so the jaws are cleaned.
 
Like orange vise said just check with a probe. No extra hardware required to interface with the machine or robot, no extra wires or hoses in the danger zone. 9810 cycle with an m1 can be used to check if a part is present, or just set tolerances for size/position using the appropriate circle or rectangular routines.
 
Russ
Ya said the grippers threaded in but smooth where it grips?........Buy some really long set screws or threaded rod..............the exposed threads would give it more grip??..................
 
could have kept your apprentice guy and he could have learned something and he would compete only against other humans in the future.but spend lots $ to buy a robot and spend time to train 'it' and off you go. replacing any human with a robot is the immoral shitshow of the decade.
 
Fun fun!

I wonder if making oring grooves in the grippers would make it hold more firmly
even the tiniest shoulder on the end of the gripper would, I think prevent movement. At the cost of needing the blank to be up off of the surface of the table, which might be more complicated than it is worth

Detecting mis aligned parts isn't really fixing the problem
 
could have kept your apprentice guy and he could have learned something and he would compete only against other humans in the future.but spend lots $ to buy a robot and spend time to train 'it' and off you go. replacing any human with a robot is the immoral shitshow of the decade.

What a load of shit.

A business is not a charity. This country already has the highest labor costs in the world.

A part loading robot is not a job killer. It is just another level of automation. CNC's are automation.

Automation makes our lives better. Workers and owners.

Go find a turret lathe to run. Idiot.
 
Great video! How does the co-bot and the vise communicate? Specifically, what tells the vise to close? Is the cart setup a standard thing? We are looking into a ro/co-bot and I had concerns about tipping over if the unit isn't bolted fast to the floor....very cool stuff!!

The robot and machining center are all integrated and setup as a package deal from Haas. How it communicates is beyond me, all I know is it works great! The base on the robot is very heavy, and comes in as one unit. It has rollers underneath, but once positioned it sits up on leveling pads. The loading fixture was a design I came up with and machined in house and mounted on the robot table. It suits our purpose for now as we do mostly small parts, the majority being less than 6" x 6", so we can get quite a few on the fixture plate, and if need be, blanks can be stacked 3 levels tall.

Later,
Russ
 
What a load of shit.

A business is not a charity. This country already has the highest labor costs in the world.

A part loading robot is not a job killer. It is just another level of automation. CNC's are automation.

Automation makes our lives better. Workers and owners.

Go find a turret lathe to run. Idiot.
I'm part of an automation company and hear that crap once in a while. Evolve or die is all I say to them.
 
Russ
Ya said the grippers threaded in but smooth where it grips?........Buy some really long set screws or threaded rod..............the exposed threads would give it more grip??..................
Fun fun!

I wonder if making oring grooves in the grippers would make it hold more firmly
even the tiniest shoulder on the end of the gripper would, I think prevent movement. At the cost of needing the blank to be up off of the surface of the table, which might be more complicated than it is worth

Detecting mis aligned parts isn't really fixing the problem

David, we had actually thought of the threaded rod scenario you suggested, and I agree with you about giving it more grip. We'll have to give that a try for the "pick up" side of the process. It may be the simplest solution of all!

gustafson, thanks for the "o"-ring suggestion. It would not mark up the parts at all when picking up finished parts out of the vise.

Later,
Russ
 








 
Back
Top