--- In DynoMotion@yahoogroups.com, "az@..." <az@...> wrote:
>
> I'm sorry I meant Inversedistance at "1" in my first question ( I wrote 50 as in 50 is one rev). The other numbers in my question relating to Inverse Distance are not what I meant either. I'll try the first step and get back to you.
>
> Thank You.
> AZ
>
>
>
>

Group: DynoMotion	Message: 7328	From: Tom Kerekes	Date: 4/26/2013
Subject: Re: Closed loop Stepper Snap Amp

Hi AZ, Attach a plot of a move on the Step Response Screen showing that the Red and Blue Plots are tracking each other with small errors (without any closed loop feedback - PID Gains zero).  Then Add some feedback gains.  Usually a small value of I gain.  The following errors should be smaller.  The process is the same as with Step/Dir Drives.  See: http://www.dynomotion.com/Help/ClosedLoopStep/ClosedLoopStepper.htm Regards TK Group: DynoMotion Message: 7383 From: az9633@ymail.com Date: 4/30/2013 Subject: Re: Closed loop Stepper Snap Amp Tom, I ran this and the red and blue plots were right on top of each other. I found that I had to zero the encoder count on the Axis screen to make the plots correspond to each other. With the I gain at zero the plots were the same as when I had up to I gain at 0.1. My main reason to use the feedback is to detect loss of position and to either re-position or indicate an error so we don't end up with a bad part. My machine is a punching machine so I'm not concerned with an accurate path I just want accurate end points. My motor has very High KE because it was designed to run with a 170 V bus and I have a 60 volt bus now with the Snap Amp. So I find that at speed the max current I can get into the motor is about 0.75 amp where at standstill I can get 2 amps. It looks like at speed there will be little correction that can be done by the gains to effect change in the following error because of the high Ke. Now that I have the axis set up this way My KCNC screen position is showing encoder counts and they don't preset when I execute the "EnableDest" command at the end of homing. How do I get my position displaying the encoder and the command to both read in inches? Also, when I hit the limit switches I have to initialize to allow motion off the switch. Can I set it up to just allow motion off the switches without re initializing? Thks Tom. AZ --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote: > > Hi AZ, > > Attach a plot of a move on the Step Response Screen showing that the Red and Blue Plots are tracking each other with small errors (without any closed loop feedback - PID Gains zero).  Then Add some feedback gains.  Usually a small value of I gain.  The following errors should be smaller.  The process is the same as with Step/Dir Drives.  See: > > http://www.dynomotion.com/Help/ClosedLoopStep/ClosedLoopStepper.htm > > Regards > TK > > > > > ________________________________ > From: "az9633@..." <az@...> > To: DynoMotion@yahoogroups.com > Sent: Friday, April 26, 2013 11:48 AM > Subject: [DynoMotion] Re: Closed loop Stepper Snap Amp > > > >   > I set up the stepper as you described. It worked. The encoder position displayed on the Axis screen is repeatable when I ran the axes from the test move screen in CL microstep mode. > > I found out that the encoders are actually 200 line per rev. > > What's next? > > AZ > > --- In DynoMotion@yahoogroups.com, "az@" <az@> wrote: > > > > I'm sorry I meant Inversedistance at "1" in my first question ( I wrote 50 as in 50 is one rev). The other numbers in my question relating to Inverse Distance are not what I meant either. I'll try the first step and get back to you. > > > > Thank You. > > AZ > > > > > > > > Group: DynoMotion Message: 7388 From: Tom Kerekes Date: 4/30/2013 Subject: Re: Closed loop Stepper Snap Amp Hi AZ, There should always be some difference between the Red and Blue Plots (Commanded and Actual positions).  Maybe you just didn't zoom in enough.  Or Plot the error to more easily show the difference.  Adding 0.1 I Gain should have made a significant difference.  Maybe your Filters are all set to Zero instead of One.  Or you didn't have CL Microstep selected? Setting a small Max Following Error will throw an error, disable the axes, and stop the Interpreter if the position is ever out more than that setting.  But another thing you could do is create an MCode to check that the position is within an even tighter tolerance after the motion completes. Actually I would expect that it is very possible to make corrections to the following error while in motion and while stopped. Position errors are relatively small and should only require small command changes. I don't understand why the Encoder wouldn't read in inches after homing.  Is it just offset by some distance?  If you move 1 inch does the Encoder Reading change by 1 inch?  At the Home switch your Home Program should set the Encoder Position with something like: ch0->Position = 0; Then enable the Axis with a destination that is the same as the Encoder Position with: EnableAxisDest(0,ch0->Position); When a Limit Switch is hit the axis is disabled.  In order to move the axis again it must be somehow enabled.  What is the problem with re-initializing?  There are ways to re-initialize without losing the Encoder position if that is the problem. HTH Regards TK Group: DynoMotion Message: 7392 From: az9633@ymail.com Date: 5/1/2013 Subject: Re: Closed loop Stepper Snap Amp Hi Tom : Thanks for the suggestions. How do I zoom the test plots? Can I use closed loop stepper with SnapAmp if I use an InverseDistance parameter other than "1" as long as i match the encoder input gain value to the same units? I'm just trying to scale everything to inches. How do I re- initialize without losing current position if I hit the limits? I would also like to recover from excess error without re homing by using the encoder position. Is the prcess similar as when recovering from a limit hit? Regards, AZ --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@...> wrote: > > Hi AZ, > > There should always be some difference between the Red and Blue Plots (Commanded and Actual positions).  Maybe you just didn't zoom in enough.  Or Plot the error to more easily show the difference.  Adding 0.1 I Gain should have made a significant difference.  Maybe your Filters are all set to Zero instead of One.  Or you didn't have CL Microstep selected? > > Setting a small Max Following Error will throw an error, disable the axes, and stop the Interpreter if the position is ever out more than that setting.  But another thing you could do is create an MCode to check that the position is within an even tighter tolerance after the motion completes. > > Actually I would expect that it is very possible to make corrections to the following error while in motion and while stopped. Position errors are relatively small and should only require small command changes. > > I don't understand why the Encoder wouldn't read in inches after homing.  Is it just offset by some distance?  If you move 1 inch does the Encoder Reading change by 1 inch?  At the Home switch your Home Program should set the Encoder Position with something like: > > ch0->Position = 0; > > Then enable the Axis with a destination that is the same as the Encoder Position with: > > EnableAxisDest(0,ch0->Position); > > When a Limit Switch is hit the axis is disabled.  In order to move the axis again it must be somehow enabled.  What is the problem with re-initializing?  There are ways to re-initialize without losing the Encoder position if that is the problem. > > > HTH > Regards > TK > > > > ________________________________ > From: "az9633@..." <az@...> > To: DynoMotion@yahoogroups.com > Sent: Tuesday, April 30, 2013 3:28 AM > Subject: [DynoMotion] Re: Closed loop Stepper Snap Amp > > > >   > Tom, > > I ran this and the red and blue plots were right on top of each other. I found that I had to zero the encoder count on the Axis screen to make the plots correspond to each other. With the I gain at zero the plots were the same as when I had up to I gain at 0.1. > > My main reason to use the feedback is to detect loss of position and to either re-position or indicate an error so we don't end up with a bad part. My machine is a punching machine so I'm not concerned with an accurate path I just want accurate end points. > > My motor has very High KE because it was designed to run with a 170 V bus and I have a 60 volt bus now with the Snap Amp. So I find that at speed the max current I can get into the motor is about 0.75 amp where at standstill I can get 2 amps. It looks like at speed there will be little correction that can be done by the gains to effect change in the following error because of the high Ke. > > Now that I have the axis set up this way My KCNC screen position is showing encoder counts and they don't preset when I execute the "EnableDest" command at the end of homing. How do I get my position displaying the encoder and the command to both read in inches? > > Also, when I hit the limit switches I have to initialize to allow motion off the switch. Can I set it up to just allow motion off the switches without re initializing? > > Thks Tom. > > AZ > > --- In DynoMotion@yahoogroups.com, Tom Kerekes <tk@> wrote: > > > > Hi AZ, > > > > Attach a plot of a move on the Step Response Screen showing that the Red and Blue Plots are tracking each other with small errors (without any closed loop feedback - PID Gains zero).  Then Add some feedback gains.  Usually a small value of I gain.  The following errors should be smaller.  The process is the same as with Step/Dir Drives.  See: > > > > http://www.dynomotion.com/Help/ClosedLoopStep/ClosedLoopStepper.htm > > > > Regards > > TK > > > > > > > > > > ________________________________ > > From: "az9633@" <az@> > > To: DynoMotion@yahoogroups.com > > Sent: Friday, April 26, 2013 11:48 AM > > Subject: [DynoMotion] Re: Closed loop Stepper Snap Amp > > > > > > > >   > > I set up the stepper as you described. It worked. The encoder position displayed on the Axis screen is repeatable when I ran the axes from the test move screen in CL microstep mode. > > > > I found out that the encoders are actually 200 line per rev. > > > > What's next? > > > > AZ > > > > --- In DynoMotion@yahoogroups.com, "az@" <az@> wrote: > > > > > > I'm sorry I meant Inversedistance at "1" in my first question ( I wrote 50 as in 50 is one rev). The other numbers in my question relating to Inverse Distance are not what I meant either. I'll try the first step and get back to you. > > > > > > Thank You. > > > AZ > > > > > > > > > > > > Group: DynoMotion Message: 7404 From: Tom Kerekes Date: 5/1/2013 Subject: Re: Closed loop Stepper Snap Amp Hi AZ, To Zoom in on the Step Response Screen just Left-Click Drag with the mouse.  Right-Click to Zoom out Yes there is a "CL Micro" mode where SnapAmp can drive a Stepper Closed Loop.  There really isn't any need to scale to Inches.  That can be done in the Application like Mach3 or KMotionCNC.  I prefer to leave invDistPerCycle at 1.0 and operate in "cycles" units (50 per rev).  Then scale the encoder to match. To re-initialize without losing encoder position simply do not change the encoder position in the Initialization. Yes as long as the encoders are working correctly the position should be tracked and maintained.   Regards TK