Aligning Takeout Robot To Molding Machine?

Davis In SC · Nov 22, 2015

I was sort of involved with the installation of some Star part removal robots, that I will be building some EOA tooling for. These are 3 axis robots, that will pick parts out of molds, and take them to a degating station, then place them in a box, table or conveyer.
My question is: Should these robots have the travel(s) indicated so they are parallel and square to the travels of the molding machine? They mount on a small pedestal at the front, and the sloppy holes on 2 planes allow a lot of error. When I mentioned dialing the travels in, the dealer and install tech rolled their eyes and tried to ridicule me.. Seems to make sense to do things right from the start, but I am from the Old School.

Tonytn36 · Nov 23, 2015

I would guess that what matters is how the parts come out of the mold. If they are ejected completely free of the mold via ejector pins, it's probably not a big deal. If the robot has to extract parts from the mold, then I would imagine that being square to the mold would be a requirement for a clean extraction.

capocoreyollo · Nov 23, 2015

In my opinion, it's never a bad thing to be square and parallel, especially in automation. We have a robot that pulls the runners from the mold after the parts are ejected, brings them out of the machine and places them in a chopper. All set up nice and square, parallel,level etc. She runs 24 hours without a hitch. Would she without the setup? Maybe. But knowing that it's setup right is peace of mind in my book.

Toolroomguy · Nov 23, 2015

Davis In SC said:
I was sort of involved with the installation of some Star part removal robots, that I will be building some EOA tooling for. These are 3 axis robots, that will pick parts out of molds, and take them to a degating station, then place them in a box, table or conveyer.
My question is: Should these robots have the travel(s) indicated so they are parallel and square to the travels of the molding machine? They mount on a small pedestal at the front, and the sloppy holes on 2 planes allow a lot of error. When I mentioned dialing the travels in, the dealer and install tech rolled their eyes and tried to ridicule me.. Seems to make sense to do things right from the start, but I am from the Old School.

We have 23 presses, all with robots. The xyx 3 axis robots are aligned with a three foot aluminum bar attached to a eoat coupler. The robot mount bolts are tightened after bringing the bar close to the platen and using a scale to measure parallelism..
The eoat suspensions take care of the slight misalignment.

Davis In SC · Nov 23, 2015

Tonytn36 said:
I would guess that what matters is how the parts come out of the mold. If they are ejected completely free of the mold via ejector pins, it's probably not a big deal. If the robot has to extract parts from the mold, then I would imagine that being square to the mold would be a requirement for a clean extraction.

That is part of my point... The Press/robot will run lots of different jobs, no telling what may be next. Since they will be storing lots of programs, seem smart to get it right from the start. If the alignment becomes an issue later, there will be a lot of programming to do over. I really think the dealer/installer do not know enough about machine geometry to really understand alignment, so it is just easier to ridicule me.. I am just a dumbass toolmaker..

Tonytn36 · Nov 24, 2015

Davis In SC said:
That is part of my point... The Press/robot will run lots of different jobs, no telling what may be next. Since they will be storing lots of programs, seem smart to get it right from the start. If the alignment becomes an issue later, there will be a lot of programming to do over. I really think the dealer/installer do not know enough about machine geometry to really understand alignment, so it is just easier to ridicule me.. I am just a dumbass toolmaker..

If it was a 6-axis robot, it wouldn't make much - if any - difference. But being 3 axis..........

Toolroomguy · Nov 24, 2015

Davis In SC said:
If the alignment becomes an issue later, there will be a lot of programming to do over. I really think the dealer/installer do not know enough about machine geometry to really understand alignment, so it is just easier to ridicule me.. I am just a dumbass toolmaker..

As a toolmaker, you understand nothing is truly flat, items came be brought closer and closer to flatness.
Since nothing is perfectly flat, nothing can be perfectly parallel. Accuracy costs money.

Take a look at the robots mount. What provisions do you see for micrometer like adjustments?

How close do you intend to get the robots axis to the machine axis?

How much time will you spend?

What do you gain?

A robot is not sent to a numerical xyz position, it is brought to where you want it and taught the position.
After a crash , you will have to reteach the positions, it is not a big deal. Drive the eoat manually to where you want it and press the button.

Miguels244 · Nov 24, 2015

I worked for CBW, alignment is very important when setting up the machine.
If you don't get it right you will be forever fussing with it.
Get it solid and you'll save yourself endless pain.

Toolroomguy · Nov 24, 2015

Miguels244 said:
I worked for CBW, alignment is very important when setting up the machine.
If you don't get it right you will be forever fussing with it.
Get it solid and you'll save yourself endless pain.

Could you expand on "forever fussing with it" and your meaning of "solid"?

Our Yushin robot z axis arms are a spindly aluminum extrusion driven by a toothed belt.
Watching them zip into the molding area,they stop and oscillate as the ejectors stroke.
The bellows suction cups and springs in the suspensions take care of misalignment and vibration.

The high speed robots are hard to follow in real time, they move so fast.
They move in,grab the parts, and get out faster than the parts would fall from gravity.

We mold about $16,000,000 worth of parts a year,I would like to know about the things we are doing wrong.

Miguels244 · Nov 24, 2015

Toolroomguy said:
Could you expand on "forever fussing with it" and your meaning of "solid"?

Our Yushin robot z axis arms are a spindly aluminum extrusion driven by a toothed belt.
Watching them zip into the molding area,they stop and oscillate as the ejectors stroke.
The bellows suction cups and springs in the suspensions take care of misalignment and vibration.

The high speed robots are hard to follow in real time, they move so fast.
They move in,grab the parts, and get out faster than the parts would fall from gravity.

We mold about $16,000,000 worth of parts a year,I would like to know about the things we are doing wrong.

What I'm saying is that the more variability you can take out of a process the better.
CBWs bots move so fast they can walk away from the molding machine...that said the installation techs take the alignment seriously.
As well as the anchoring process.
Maybe they are over doing it, but the controls guys can rely on the machine being in the right place all the time, and compliance only goes so far.

capocoreyollo · Nov 24, 2015

Davis In SC said:
That is part of my point... The Press/robot will run lots of different jobs, no telling what may be next. Since they will be storing lots of programs, seem smart to get it right from the start. If the alignment becomes an issue later, there will be a lot of programming to do over. I really think the dealer/installer do not know enough about machine geometry to really understand alignment, so it is just easier to ridicule me.. I am just a dumbass toolmaker..

Typical dumbass toolmaker.. :rolleyes5:

I agree buddy, those guys simply don't have the understanding of machinery and geometry that a toolmaker such as yourself would. More work for an installation guy that doesn't want to be there and is probably not getting paid by the hour, but for the job itself? Yeah right! You had to see the guy that came into to put on our robot...Go with your gut and get that sucker mounted right! Especially if your going to be running god knows what types of parts, your bound to run into to some problems with it here and there when some crazy parts start coming thru the machine. Taking a few variables out of the trouble shooting equation from the start could never possibly hurt. Taking a little time to get it set up right now, to save a lot of time in fussing around with it when those challenges do arise in the future, any good engineer/manager would see those benefits. Good luck!

Corey

Toolroomguy · Nov 24, 2015

Miguels244 said:
What I'm saying is that the more variability you can take out of a process the better.
CBWs bots move so fast they can walk away from the molding machine...that said the installation techs take the alignment seriously.
As well as the anchoring process.
Maybe they are over doing it, but the controls guys can rely on the machine being in the right place all the time, and compliance only goes so far.

So you never actually worked with robot equipped injection molding machines on a 24/6 basis, like I do?
I was wondering what that "constant fussing with" you were describing was because I don't see it.

Davis In SC · Nov 24, 2015

capocoreyollo said:
Typical dumbass toolmaker.. I agree buddy, those guys simply don't have the understanding of machinery and geometry that a toolmaker such as yourself would. More work for an installation guy that doesn't want to be there and is probably not getting paid by the hour, but for the job itself? Yeah right! You had to see the guy that came into to put on our robot...Go with your gut and get that sucker mounted right! Especially if your going to be running god knows what types of parts, your bound to run into to some problems with it here and there when some crazy parts start coming thru the machine. Taking a few variables out of the trouble shooting equation from the start could never possibly hurt. Taking a little time to get it set up right now, to save a lot of time in fussing around with it when those challenges do arise in the future, any good engineer/manager would see those benefits. Good luck!

Corey

Seems it would have only taken a few minutes to run an indicator along a tie bar and/or the platen. I even offered them an indicator and a way to mount it..

Miguels244 · Nov 24, 2015

Toolroomguy said:
So you never actually worked with robot equipped injection molding machines on a 24/6 basis, like I do?
I was wondering what that "constant fussing with" you were describing was because I don't see it.

Hey, seems to work for you, more power to you.
For my part having set up lots of production equipment, I'll take some time on the front end rather than run the significant risk of fighting the tooling just to end up spending the time on the tail end anyway.

Toolroomguy · Nov 24, 2015

Davis In SC said:
Seems it would have only taken a few minutes to run an indicator along a tie bar and/or the platen. I even offered them an indicator and a way to mount it..

You are right.
OK, say you found the robot travel was .025" off of being parallel to the platen face.
Are you going to put a shim under the mount?
Take the mount apart and grind an angle on it?

The problem is the first time someone loads the wrong program,doesn't change the eoat for the new job, misroutes a waterline, forgets to remove an eyebolt,etc, the robot will crash. The thinwall extrusions that make up the robot arms will get tweeked by a few thousands, and the pick positions will all have to be retaught.

Toolroomguy · Nov 24, 2015

Miguels244 said:
Hey, seems to work for you, more power to you.
For my part having set up lots of production equipment, I'll take some time on the front end rather than run the significant risk of fighting the tooling just to end up spending the time on the tail end anyway.

Thank you for recognizing the advice from a guy who has a theory about aligning robots to presses might not be equivalent to the observations of a guy who is living the dream day after day.

Fanuc had a video of our 1000 ton on their website years ago. The robot would pick the part, drop it into a station, the robot would do an automated eoat swap, clip the fan gate, swap eoat again, pick the part and place it on a conveyer, all with the molding cycle.

Davis In SC · Nov 24, 2015

Toolroomguy said:
You are right.
OK, say you found the robot travel was .025" off of being parallel to the platen face.
Are you going to put a shim under the mount?
Take the mount apart and grind an angle on it?

The problem is the first time someone loads the wrong program,doesn't change the eoat for the new job, misroutes a waterline, forgets to remove an eyebolt,etc, the robot will crash. The thinwall extrusions that make up the robot arms will get tweeked by a few thousands, and the pick positions will all have to be retaught.

I was mainly concerned with X and Y, which should both be correct, if one is dialed in,, Or at least close.. There is a pedestal mount with top and bottom fasteners, so there is a possibility of the travels being several degrees (possibly) off..

Toolroomguy · Nov 25, 2015

Davis In SC said:
I was mainly concerned with X and Y, which should both be correct, if one is dialed in,, Or at least close.. There is a pedestal mount with top and bottom fasteners, so there is a possibility of the travels being several degrees (possibly) off..

You should redo the installation if you are degrees off, I agree.
Using a three foot bar,as we do, and adjusting to be within 1/64, gives an max error of .024 degrees, or a misalignment of 1.4 minutes.

We pin secondary operation equipment to the floor with air cylinder operated pins built into the base of the equipment.
Most mold changes are plug and play, with no "constant fussing" of robot positions.

Miguels244 · Nov 25, 2015

Toolroomguy said:
You should redo the installation if you are degrees off, I agree.
Using a three foot bar,as we do, and adjusting to be within 1/64, gives an max error of .024 degrees, or a misalignment of 1.4 minutes.

We pin secondary operation equipment to the floor with air cylinder operated pins built into the base of the equipment.
Most mold changes are plug and play, with no "constant fussing" of robot positions.

This is in line with my experience.

Davis In SC · Nov 25, 2015

Toolroomguy said:
You should redo the installation if you are degrees off, I agree.
Using a three foot bar,as we do, and adjusting to be within 1/64, gives an max error of .024 degrees, or a misalignment of 1.4 minutes.

We pin secondary operation equipment to the floor with air cylinder operated pins built into the base of the equipment.
Most mold changes are plug and play, with no "constant fussing" of robot positions.

The bar test sounds good to me...

Aligning Takeout Robot To Molding Machine?

Diamond

Diamond

Aluminum

Cast Iron

Diamond

Diamond

Cast Iron

Diamond

Cast Iron

Diamond

Aluminum

Cast Iron

Diamond

Diamond

Cast Iron

Cast Iron

Diamond

Cast Iron

Diamond

Diamond

Similar threads