Hi Rob,

What kind of drives do you have?  What are the input requirements?  How are they wired?  What are your configuration settings?  Do the other axes work?

Regards

TK

Group: DynoMotion	Message: 9432	From: Tom Kerekes	Date: 4/30/2014
Subject: Re: Move and step on step response screen

Hi Rob, I Googled and found this: http://www.pushcorp.com/Manuals/6410-user-manual.pdf This manual states that a Direction change requires a 50us setup time before the first step pulse.  KFLOP only provides 4us so that could be a problem.  But that would not cause the motor not to move at all like you are describing.  It would cause a possible single microstep move in the wrong direction when the direction reverses. Pushing "Step" on the step response screen isn't really useful for non-servo type systems like yours.  It would cause a big jump at max step rate (2.5MHz) to the new position.  In fact at that rate and pulse size all the step pulses will overlap to form just one big pulse - which probably explains what you are seeing one step. It isn't really clear what steps you are performing. Output Channel 11 should be correct for Step/Dir Generator #3 in TTL mode.  But is that what you are entering in the Config/Flash Screen?  Or in the C Program.  When pushing the Move button all the settings in the KMotion Screens are written down to KFLOP which would overwrite whatever a C program might have configured. Regards TK   Group: DynoMotion Message: 9433 From: Rjreese29 Date: 4/30/2014 Subject: Re: Move and step on step response screen Read below my integrated responses. Thanks Tom!!! Rob On Apr 30, 2014, at 1:22 PM, Tom Kerekes <tk@...> wrote:   Hi Rob, I Googled and found this: http://www.pushcorp.com/Manuals/6410-user-manual.pdf This manual states that a Direction change requires a 50us setup time before the first step pulse.  KFLOP only provides 4us so that could be a problem.  But that would not cause the motor not to move at all like you are describing.  It would cause a possible single microstep move in the wrong direction when the direction reverses. This axis is has an encoder, but is being setup as openloop. Thanks for pointing that out.My vision was to close the loop, but not until I get it to move:). My idea was to close the loop, if the loop is closed, and I don't have high expectations of running it at extremely high response time/speeds, would it/may it still be ok?  Pushing "Step" on the step response screen isn't really useful for non-servo type systems like yours.  It would cause a big jump at max step rate (2.5MHz) to the new position.  In fact at that rate and pulse size all the step pulses will overlap to form just one big pulse - which probably explains what you are seeing one step. Thanks for clarifying:) makes sense. It isn't really clear what steps you are performing. Output Channel 11 should be correct for Step/Dir Generator #3 in TTL mode.  But is that what you are entering in the Config/Flash Screen?  Or in the C Program.  When pushing the Move button all the settings in the KMotion Screens are written down to KFLOP which would overwrite whatever a C program might have configured. I have downloaded these settings in the config/flash screen, changed the values in init.c and flashed (standalone). Flashing the standalone mode may have been pedantic, but I am trying to be consistent in the variety of places to change settings. Thanks for clarifying the priority of what gets overridden. Regards TK   Group: DynoMotion Message: 9434 From: Tom Kerekes Date: 4/30/2014 Subject: Re: Move and step on step response screen Hi Rob, Regarding Direction Setup Time: It shouldn't really make a difference if the system moves fast or slow.  The problem occurs when the system reverses and where the speed would be very slow regardless.  Closed loop could potentially correct the problem.  If a micro step is lost or made in the wrong direction more will be generated until the encoder eventually reads the correct position.  One problem with this is that the missed steps could accumulate over time to be a very big value.  So even though the axis would always be in the right place, the commanded position may be far off.  This will prevent you from relying on following error to detect a stall, crash, or fault.  If you have the correct setting in the Config/Flash Screen and push "Move" on the Step Response Screen then the Parameters should be downloaded and set properly in KFLOP.  So I can't see why it isn't working.  We don't recommend Flashing parameters and Programs into KFLOP as it becomes confusing what has been changed.  It is better to just start with a clean slate and set everything as required.  There is not usually any benefit to Flashing KFLOP when a PC will always be available to set things automatically.  Try Flashing "New Version" to revert everything in KFLOP to default. You might also test our hardware by just toggling the step signal using the Digital I/O Screen.  You must cycle power on KFLOP for that to work. If the pins have been configured to a Step/Dir Generator they will not toggle as GPIO.  Of course you must toggle many time before you will see much movement. Regards TK   Group: DynoMotion Message: 9437 From: Rjreese29 Date: 5/1/2014 Subject: Re: Move and step on step response screen I'll have read and digest the rest of your email later,.. Thanks. But wanted to let you know it is working. The move button was having an effect on my step led light this morning. When it actually started working is when I set the init.c from: FPGA(STEP_PULSE_LENGTH_ADD)=32; To FPGA(STEP_PULSE_LENGTH_ADD)=32 + 0x80; Everything is 'move' and 'step' as normally would be expected. Thanks Tom!! Rob On Apr 30, 2014, at 5:32 PM, Tom Kerekes <tk@...> wrote:   Hi Rob, Regarding Direction Setup Time: It shouldn't really make a difference if the system moves fast or slow.  The problem occurs when the system reverses and where the speed would be very slow regardless.  Closed loop could potentially correct the problem.  If a micro step is lost or made in the wrong direction more will be generated until the encoder eventually reads the correct position.  One problem with this is that the missed steps could accumulate over time to be a very big value.  So even though the axis would always be in the right place, the commanded position may be far off.  This will prevent you from relying on following error to detect a stall, crash, or fault.  If you have the correct setting in the Config/Flash Screen and push "Move" on the Step Response Screen then the Parameters should be downloaded and set properly in KFLOP.  So I can't see why it isn't working.  We don't recommend Flashing parameters and Programs into KFLOP as it becomes confusing what has been changed.  It is better to just start with a clean slate and set everything as required.  There is not usually any benefit to Flashing KFLOP when a PC will always be available to set things automatically.  Try Flashing "New Version" to revert everything in KFLOP to default. You might also test our hardware by just toggling the step signal using the Digital I/O Screen.  You must cycle power on KFLOP for that to work. If the pins have been configured to a Step/Dir Generator they will not toggle as GPIO.  Of course you must toggle many time before you will see much movement. Regards TK   Group: DynoMotion Message: 9438 From: Tom Kerekes Date: 5/1/2014 Subject: Re: Move and step on step response screen Hi Rob, Looks like that drive doesn't like negative going pulses.  Strange most drives prefer that. Thanks for letting us know. Regards TK