Robot Cell Automatic Cycle Time Recording - Robot Cell

[Syncro5]

Hello all,

I'm after some advice...
I'm a process engineer in a welding / assembly shop with over a thousand welding and handling robots. Currently there is limited automatic cycle time recording function in the jig & robot cells which means every time we launch a new product, there are hundreds of engineering hours spent videoing, analysing and reporting on the process achievement.

I’m carrying out an engineering efficiency activity and one of the biggest improvements will be to automate the cycle time recording during the new product trial phase, and output the history to a data-base. It’s a huge job involving software and robot program updates across the shop, as well as some PC engineering to output the data in a usable format. All in all it's likely to take a year or two to complete.

Before I start to pull the team together to initiate this change, I’m trying to gather some info on what other companies are doing for automatic cycle time recording. I feel as though we’re a bit behind the times – can anyone confirm that?

Can anyone give me an example of automated data-base cycle time recording where they work, and the industry they’re in?

Thanks,

Joe.

 

[IRstuff]

So, you either have a timed welding program, or a machine log with operations and timestamps? In either case, you have the time histories of each operation, don't you?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Syncro5]

Hello IRstuff,

Sorry I missed some important detail out of the original post. I've updated it now.

Yes the robot controllers log a history of one or two cycles of any given program, which can be taken from the robot pendant. There are software timers in the line software (PLC) and, in some instances, the timers manifest themselves as a "station cycle" on the line's main operator PC - again the history is limited to one or two units. The line software timers rarely get used because they aren't set in a manner which negates wait times such as "Exit Full" or faults on the line. One of the main bodies of work is to change the software and robot programs (adding additional outputs to start / stop software timers) so that we can record the important parts of the cycle and ignore waits. This task needs to be done on all stations individually as no two are the same.

I missed off the original post that I'm looking specifically for database recording, I think this is the only way to get the engineering time saving I have in mind.

 

[FACS]

I'll try to make this brief:

Each robot movement is recorded on the pendant. To get each movement out, you will need to add I/O to each PLC in the robot controller.

Each bit you want to extract needs an output.
Each bit you want to read needs to go into I/O on the main PLC running the system.
The main PLC will report the information to your MIS system, and you will need to buy software tags to transfer these bits.
Then you can do what you want with the information.

So every robot controller will need expanded I/O
Every process PLC will need expanded I/O
Every process PLC will need software tags for every bit of information
MIS then needs to figure out what type of output you want, as in reports, data logging....

The amount of everything listed above depends on how many part or process recipes you have, and how many features in that process do you want to capture?

250 software tags cost $3,300.00. One tag transfers one bit of information so think about how many individual features you need to capture.

So this is just the robot. The other parts of the system that are not using the robots will have to be addressed in the same way.