On Jul 11 2013 7:48 PM, Jon Elson wrote: > John Thornton wrote: >> How do you determine the right angle is reached? >> > A few samples will have to be bent to determine the effect of > springback. > There will be an encoder on the bending arm, I probably failed to > mention > that. So, it will have an encoder, but not a servo driving the axis. > So, > that is why this needs to be treated differently from a standard A > axis. > > A few samples will have to be bent to determine the effect of > springback. > > We'd like for the angle of the bend to be specified in the G-code, > somehow, but it could be a variable, it doesn't need to be like an > axis word. The value can then be tweaked until the bends end up > right.
I've operated a machine like that a few times (but not many). The person who taught me to use the machine did not tell me to let it normalize over a long period of time (like the 24 hours mentioned in one of the other emails), just a few seconds. What we did was to mark the final angle with a chalk mark, bend to the mark, let it spring back, get a sense for how far it sprung back, now bend again to the mark + dA_1/2, repeat for dA_N/(N+1). It usually only took a couple of bumps before I needed to measure against a large bevel gage... This would be automatable trivially and you can then set it aside to normalize if you need. As for using the motor/servo, you can try putting an encoder on the bending arm and see if you can control it well enough. My guess is that you will be OK if the gear reduction is high enough (which I suspect it will be at least 50:1 and more likely 200:1). I expect that you should be able to tune a PID to that, but you might not be running the motor as fast as you could otherwise. Oh, I almost forgot. I have a picture in my head of the encoder being able to measure the springback directly when you remove power from the motor, then you can just bump it. Also, if you take many measurements of the springback over your relaxation time, then you may be able to develop a predictive model of the material over man seconds/hours/days. It would take some tricky statistical analysis to make sure that the relaxation follows various assumptions of decay in the long tail (like exponential or Gaussian). Just some thoughts... EBo -- ------------------------------------------------------------------------------ See everything from the browser to the database with AppDynamics Get end-to-end visibility with application monitoring from AppDynamics Isolate bottlenecks and diagnose root cause in seconds. Start your free trial of AppDynamics Pro today! http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
