Safety Standard For Grippers in Robotics

[JLand]

Specifically regarding de-energizing a machine that needs to be energized to avoid the introduction of additional risk, what you are talking about is a category 2 stop. Standards try their hardest to steer you away from using a category 2 stop after an e-stop button has been pushed, but with justification that a freewheel stop (category 0) and a controlled stop (category 1) cannot be used, a category 2 stop is appropriate.

You also hit upon an enabling function. This overrides the emergency stop and allows for safe, slow motion even with a safety function active. If you are going this route, you need a full safety system design. The inclusion of an enabling function without a even a risk assessment is dangerous to the operator and legally dangerous to your company. If you go with an enabling function, you really should reach out to a machinery safety engineering team.

 

[padees]

I'm a graduated ME but have been hired into a controls position at my company, I liked reading this

Use your ME skills and apply to controls. Define controls thru your ME Training and Skills.

Be precise on how the mechanics work to convey to others.

Never let go of your ME training and Skills, keep working/developing on each and integrate into each other.

 

[bill4807]

If your grabbing metal maybe introduce magnets. They hold on loss of air. Yes they are expensive.
Some companies want all energy stored - released, so blocked ceter valves are out, you may need to add pilot operated checks on open center valves with a means to release those slowly under a maintenance condition.

Like also mentioned above add a mechanical feature to trap the part. This may be a good idea anyways for peace of mind.

 

[mk42]

Use your ME skills and apply to controls. Define controls thru your ME Training and Skills.

Be precise on how the mechanics work to convey to others.

Never let go of your ME training and Skills, keep working/developing on each and integrate into each other.

I've always said that Controls is at least as much about understanding the process you're controlling as it is the automation itself. An ME puts you in a good position to understand motors and robots and heating and several other useful processes. ChemE is really helpful in the process industries. Not saying EE or mechatronics or something is a bad idea, but effective engineers can come from more backgrounds than you'd think. Best "engineer" I've known come from a poly sci background, of all things. Crotchety old coot, but knew his stuff.

 

[JesperMP]

Their safety team has taught them that an emergency stop is supposed to remove stored energy in the system to prevent extra hazard,

You have a dilemma then if removing the stored energy causes additional hazards, eh ?
I don't know what OSHA says, but I bet that an emergency stop must bring a machine to a halt according to the stop category. That may or may not require removing stored energy.
Definitely yes to not cause additional hazards.

Does the standard ISO 10218-2 (for robot integrators) provide some clue as how to deal with loads that are held by grippers ?
In the US it is also known as ANSI/RIA R15.06 (second half).

 

[Mandel314159]

You have a dilemma then if removing the stored energy causes additional hazards, eh ?
I don't know what OSHA says, but I bet that an emergency stop must bring a machine to a halt according to the stop category. That may or may not require removing stored energy.
Definitely yes to not cause additional hazards.

Does the standard ISO 10218-2 (for robot integrators) provide some clue as how to deal with loads that are held by grippers ?
In the US it is also known as ANSI/RIA R15.06 (second half).

I'll take a look at these! thank you

 

[Mandel314159]

Use your ME skills and apply to controls. Define controls thru your ME Training and Skills.

Be precise on how the mechanics work to convey to others.

Never let go of your ME training and Skills, keep working/developing on each and integrate into each other.

I've always said that Controls is at least as much about understanding the process you're controlling as it is the automation itself. An ME puts you in a good position to understand motors and robots and heating and several other useful processes. ChemE is really helpful in the process industries. Not saying EE or mechatronics or something is a bad idea, but effective engineers can come from more backgrounds than you'd think. Best "engineer" I've known come from a poly sci background, of all things. Crotchety old coot, but knew his stuff.

I intend to take these ideas to heart, ME has taught me a lot in my schooling and I definitely have a penchant for it, but all in all, it's been super useful so far in my early career. I really like dissecting mechanical processes and finding entry points for more and more sophisticated controlling techniques. So far this job I'm on with these grippers is the most complex one I've been a part of, a lot of servo calibration/characterization around the mechanical system, programming HMI, setting up data readouts, and obviously, programming a robot.
I'm a huge fan so far, although my hobbies are still mostly divergent from controls engineering altogether, my dream project is a walking robot like a robot-dog or something.

 

[mk42]

although my hobbies are still mostly divergent from controls engineering altogether

way off topic at this point, but that's the way i prefer my hobbies. If you relax by doing things just like work, I feel like your hobbies would stop being relaxing really fast. Then again, maybe you aren't doing hobbies for relaxation, maybe its for the satisfaction or something else. If you're stuck in a slog at work and nothing is going anywhere, I could see a personal controls project that you can actually FINISH being very satisfying.

 

[BOWILLY]

You must first determine if there is a safety issue. A safety issue causes harm or increases the risk of harm, to a person. A proper and thorough risk assessment should address this. If there is no safety issue, it then becomes a controls issue. Will the product be damaged? Will it damage whatever it strikes? How does the loss of the gripped part affect line restart?

 

[Mandel314159]

You must first determine if there is a safety issue. A safety issue causes harm or increases the risk of harm, to a person. A proper and thorough risk assessment should address this. If there is no safety issue, it then becomes a controls issue. Will the product be damaged? Will it damage whatever it strikes? How does the loss of the gripped part affect line restart?

can you expand a bit on what you mean by risk assessment? My work has of course our own internal way to assess risk, but I'm interested to hear more

 

[Jeev]

can you expand a bit on what you mean by risk assessment? My work has of course our own internal way to assess risk, but I'm interested to hear more

A risk assessment compliant to and following the guidelines of ISO 12100 and/or ANSI B11.TR3-2000.

In my part of the world this is AS 4024.1201 and AS 4024.1303.

 

[BOWILLY]

A risk assessment looks at a process from the human perspective. That "human" could be an operator, a maintenance person, a cleaner ; any number of people. One must assess the possible points where a person can be harmed, to what level of severity and with what level of likelihood. Having a documented risk assessment is a critical part of an employer's responsibility to an employee's health and safety. It also goes a long way in proving "due diligence" should an accident occur. Your local department of labor, or OSHA should be able to quote the relevant legislation.