Comparing robots is driving me *!#${

[steevke]

Hello robot guys,

Some advise is needed: we (small company) are planning to buy our very first robot; plan is to use it for 3D laser welding. We will only weld for ourselves, and we don't need offline programming etc.

We've had quotations from Motoman, Fanuc and Kuka.

Pricing is all more or less comparable to each other once we add the cost of the laser & the optics, so we decided not to take the small price difference in the robot into account, and aim for the best robot for our application.

Kuka claims that their KR30 HA robot is the best solution for us, because of higher path repeatability compared to their competitors.

So my question: how accurate are the Motoman (HP20D + DX100) or the Fanuc M-20iA + R30iA compared to the KUKA? We've never worked with robots before, so it's difficult to separate the marketing talk from the reality.

Is the KUKA's path repeatability really so much better compared to other brands?

Any tips welcome.

Thanks

Steven

[rzapo]

motoman is 0,06 mm. but you must take also other advantage in calculation like the fiability, suport, maintanace, easy to use. the dx100 is the best from Motoman and have grate facility. cal the motoman distributor from your area to explain all. In welding field Motoman have a name. Where are you located?

[botix]

Motoman is comparable in terms of repeatability as rzapo says, and Fanuc is looser at 0.08.
Question is how accurate do you really need it to be. There's not that much difference between them.

A KR30 is a whole lot bigger physically than any of the other two robots, they are probably offering it because Kuka's next size down is 16Kg.

If you have the robot moving delicate and expensive stuff around then look at collision detection. This varies greatly between robot brands and frankly in the case of Kuka just plain sucks. I find that Motoman is better and Fanuc the best if you do it right.

In terms of ease of teaching with pendant Motoman, Fanuc and ABB all beat Kuka in my opinion.

Try to think of any extra features/software/capabilities that you may need to add down the road and check prices for that too.

[TygerDawg]

Never before used bots, but making your first purchase for 3D laser welding.   Hmmmmmm.....

See the discussion concerning accuracy / repeatability / path accuracy in this thread:
 
http://www.robot-forum.com/robotforum/fanuc_robot_forum/robot_repeatability-t1550.0.html

You really should spend the money to spend some time in each vendor's application lab playing with the arm to get a feel for how each one programs, moves, etc.  That might be an eyebrow-raising experience that would help you make a decision.

And after that, if you haven't done it already, you should study up on laser safety.  Eyes don't get second chances with lasers.

[steevke]

Never before used bots, but making your first purchase for 3D laser welding.   Hmmmmmm.....

And after that, if you haven't done it already, you should study up on laser safety.  Eyes don't get second chances with lasers.

Yeah, well holding the optics with our hands would be a tad difficult, so we'll take our chances with the robot  We've test welded at a local laser research facility and that has gone remarkably well.

Regarding safety: the laser will be in a light tight double walled cell and we'll be sure to hang those fancy Laservision protector D-L8 glasses on the door. I'll even put a special label on the door: "Warning: don't look at the laser beam with your remaining eye" for the foolhardy.

But kidding aside, your reply in the Fanuc forum to local frames of reference seems to confirm my feeling that for our application we don't really need the path accuracy of the Kuka, as long as we provide plenty of teached points on the way for the robot to follow.

Thanks.

Steven

[Robo Guru]

I think you are getting accuracy and repeatability mixed up.  Take for example a MOtoman robot at around 0.06 mm.  That means the robot will come to a taught point within 0.06mm.  That is pretty freakin good in my opinion. 

Accuracy really only becomes a factor once you start using offline software.  Since you are not using offline software you should be more concerned with repeatability.  Any major industrial robot that is on the market should have a decent repeatability. (ie. Fanuc, Motoman, ABB, etc...)

In my opinion I would buy from the company that gives you the best local support.

RoboGuru

[rvrklr]

I am not sure as to your question of which is more repeatable and has tighter movement tolerances.  I Have dealt with Fanuc, and Motoman.  The pros to Motoman are that their tech support is amazing, their application guys, the same.  The con however is that I am on a first name basis with the motoman tech support, due to all the problems and incomplete literature that comes with the robots.  The fanuc Robots we have are easy to understand, and their programming and teach pendant layout could not be simpler.  I would recommend fanuc over all others.  The Fanuc vision is seamless and really the best vision package on the market.  3D vision is cutting edge, and fanuc does it well.

[TylerRobertson]

I think you are getting accuracy and repeatability mixed up.  Take for example a MOtoman robot at around 0.06 mm.  That means the robot will come to a taught point within 0.06mm.  That is pretty freakin good in my opinion. 

Accuracy really only becomes a factor once you start using offline software.  Since you are not using offline software you should be more concerned with repeatability.  Any major industrial robot that is on the market should have a decent repeatability. (ie. Fanuc, Motoman, ABB, etc...)


RoboGuru

I'm with this - if you're not concerned with any sort of offline programming (ie: Shifting points in space by typing in arithmetic, or writing programs in text form, etc.) then accuracy is not an issue. When you teach programs you're managing the accuracy - the repeatability is the key, and most modern robots should be able to handle a decent repeatability... It will come down to what robot you like to use - if it's flexibile (if your demands change), what the local support/parts availability is like ..