Robot cells : wash / deburr

In the STUTTGART Parts2clean exhibition , a lot of WASHER manufacturers were claiming their ability to supply robot applications for the wash and deburr process.

The application of robots in the wash/deburr is not new. Late in the 1980’s AGULLO was one of the first companies in the world to look for the application of these technologies. In this time the robots were mainly hydraulics or pneumatics and the main robot suppliers were the Americans UNIMATION ( Polar-articulated) and PRABB( cartesian) .

 To look closely to this applications I remember that I visited these two companies in USA , and both were proclives to decline any warranty of the robot working in the wet and dirt conditions of a washing machine.

Why? : The conditions inside a wash chamber are very inconvenient for the robot : water splash, possible direct contact with high pressure jet( 150 to 800 bars), hot ambient ( 50 to 70ºC), high humidity air , vapours , condensations ,  solids dirt , mechanical chock due to the chips removed from the part that can crash against the robot arms as bullets…

For the part gripper at the end of the robot wrist the same problems that for the robot : water , humidity, condensations, temperature, mechanical bullets… and additionally : problems for the electric switches and wiring , and pneumatic hoses ,controlling the open/close positions of the gripper. 

So de-couraging was the result of these visit to USA that in the AGULLO company it was decided to design a specific robot to respond to the difficult conditions of the wash/deburr operations. The Market was requesting machines able to wash different parts, in the same machine, and at low production and the robot could help for that. The classic washing machines had then wash box fixtures with pin point jets adapted to each part to be washed . If the parts to be washed were different , then it was necessary to change the wash boxes of the machine at each time that the new part arrived to the machine.

For certain machines like crankshaft washers , or cylinder head washers , it was possible to create one fixture that was carrying he wash boxes and then the operator could change the complete package for each new part. With the cylinder block washers the wash fixture becomes heavy and requested a lot of time and hand labour to be changed. A robot (despite his high price) could simplify the work with only a wash robot program to be stetted for each part.  

The result in 1983 was the design of a Cartesian robot , five axis ,with all their mechanics and servos on the opposite end of the arms( patented) .The payload was 350Kg and a prototype for validation was build in the Barcelona Agullo factory. Some of the customers robot experienced people visited the robot and gave some feedback. The most important  was coming from a Mercedes robot engineer suggesting to switch from the hydraulic servos( American Moog) of the prototype to the electric drive system as it was the future for the robots at least in Europe. The robotic electronic command was also another concern . Too much electronic companies offered robotic commands in the shows but they didn’t have experience in the field. By other side the customers didn’t like to train their people to each robotic command of the suppliers .

Another concern was also to consider if instead to grip the part to be washed to the robot arm , if it would be faster and agile to move the jet with the gantry robot and leave the part static. Then instead to need a 300Kg payload to move a cylinder block to wash and his gripper , we would need only a 50 or 70 Kg payload for the high pressure applications. Finally the decision was to re-convert the design to a Cartesian gantry robot with electric drive and servos with CNC controls ( Siemens or Allan Bradley the more popular then in the field ) and to move the jet. Totally a four axis gantry robot CNC controlled , working against the part to be washed presented in a fixture , or an in line transfer or a rotary table transfer. The fixture could have a supplementary CNC axis giving a total of five axis , enough for the wash/deburr process.

With this gantry robot located in the roof of the machine , with folded bellows to close the gap on the roof , all the mechanical an electrics of the gantry were outside the machine well protected against any water splash , or condensations. No mechanics neither electrics were on the wet area of the machine but all was outside the machine in the roof for easy access.The CNC command was controlling the gantry robot and the machine itself.

This AGULLO gantry robot was build for more than 17 years , around 100 units , and a derivative of it is still supplied by Dürr Ecoclean on some Rotary transfer washers for cylinder blocks and cylinder heads applications. There is one of these gantry robot 14 years old that is still working in his washer.

Why I have explained all these past experiences?  Because the alternative to the gantry robot using the articulated robots in the machines was difficult years ago. The first that I know were used the Smart-COMAU(Fiat) for the Borg Warner USA compressor applications but with the articulated robot outside the machine , vertical, on the floor, and with a complete vertical circular bellow wall to protect the robot. This bellow last normally no more than one year in normal conditions but with accidents or bad manual movements it has to be replaced very often. The english manufacturer CERA was also trying to use the same robot layout but with ABB robot.

In France Renault developed also one small vertical Cartesian robot that was used by the french company Brochot on the machines supplied to Renault to clean / deburr gear boxes but with many problems on the belows and mechanics, because the complete robot was inside the machine.

Fanuc was also marketing his small articulated robots for small wash applications ( first in japan and later in USA), but with frequent disassembling for maintenance.

Then ABB pushed also his robots . They were applied in the high pressure jet cutting located in the roof of the water-jet cabinets , and this application inspired to certain washer manufacturers to use the articulated robots inside the washing machines. 

 

Manufacturers like STIC_Hafroy in France and ICOM in Italy , Eurowide in England used the ABB in the roof with a big textile wrap ( moving the water jet).

Today the articulated robot suppliers have improved their robots and now they start to propose a better water protection to the end wrist and to the robot body . (As is the case of ABB and in a second place Fanuc,may be we could also add Staubli if his last robot runs well). In consequence there are more and more manufacturers that they propose “robotic washers” using standard robots of the market.

A lot of robot  suppliers assure IP67 protection for the robot ,but maybe we have to remember that IP67 protection in the robot is not enough for a wash deburr application . You would have to go to the “electrics standards” and see what means “IP67″ . The first digit”6″ means solids protection :”power dirt protection” .The second digit (“7″) is a water protection but “7″ means protection to the water inmersion without pressure for 30 minutes¡¡ Inside a wash/deburr cell , there is high pressure water ¡¡

Some robot suppliers claim higher protection :”IP68″ , but again please note that the digit ”8″ means “harder conditions (to be defined by the supplier) of immersion than the IP67″ but still immersion , not protection to the high pressure jets.

What means all that ?, that the washer manufacturer who is proposing to a client a “robot cell” with an articulated robot of the market needs to add some additional engineering to his machine for additional protection of the robot . If not, the client is taking a big risk ordering a machine to this washer  manufacturer. 

But there is still a confusion using the term “robotic washer”. What is better to move the jet with the robot or the part against the wash/deburr jet?. We will see in a next post…