[TomKerekes]

KMotion/KFlop Test Version 4.33l Available

Includes new KMotionCNC Fiducial Alignment App. See Description and Video:

https://youtu.be/a_p38zktIZI


KFLOP Step/Direction Generators now have 5.7us of Direction Setup Time for compatibility with Leadshine drives and others.
CW/CCW Pulse Mode is also available. See updated document here:
Step and Direction Setup

LabView Sample now Included

https://youtu.be/dPZIdLSd_4M


Moving out of Soft/Hard Limits issues fixed

KMotionCNC M105-M119 bug fixed

KStep Step-by-Step Tutorial Added

New C Examples

Many other bug fixes and additions

For details see
http://dynomotion.com/Software/KMoti...%20Changes.pdf


Download here:
http://dynomotion.com/Software/KMotion433l.exe


Please report any issues.

Regards,
Dynomotion, Inc.

[mk00]

I have issue with 4.33l firmware. One of my steppers twitches on stop and lost steps, on accleration all looks good. All parameters and initialization are the same as with 4.32 firmware, but with 4.32 all is good. I tried different speed, acceleration and pulse time. I use MoveRel command. Driver is Leadshine M542.

[mk00]

addition: with Jog and Move comands all is good, MoveRelAtVel is bad.

[mk00]

no, problem with small movements

[mk00]

with large amount of jerk, like 10^8, works ok
but anyway that operates in 4.32 not working in 4.33l

[TomKerekes]

Hi mk00,

Sounds like clearly we have a bug. We did make some changes in V4.33i that are regarding the unused "Trip States" for Independent motion that does require all states (constant Acceleration, constant Velocity). And others have reported something that sounds similar.

Do you have a specific case that fails? Velocity, Acceleration, jerk, Distance?

Regards

[mk00]

Do you have a specific case that fails? Velocity, Acceleration, jerk, Distance?

Hi, Tom

This occurs during deceleration with small movements, like 800 steps with 10000 velosity and 10^5 acceleration, 10^6 jerk. Now I set 5*10^4 acceleration and 10^8 jerk - looks ok.

[TomKerekes]

Hi mk00,

You are correct we definitely introduced a bug in V4.33l in the KFLOP Firmware. Sorry about that. When making moves (Move, MoveRel, MoveAtVel, MoveRelAtVel) with different numbers of Trip States (some that achieve max V or A and some that don't) the last Trip State may not be terminated and a Trip State from a previous move may erroneously added in. Here is an example bad motion:

Attachment 284726

Here is a new patched firmware for KFLOP that should be compatible with V4.33l and correct the problem:
http://dynomotion.com/Software/Patch...l/DSPKFLOP.out

Copy the file to the \DSP_KFLOP directory, run KMotion.exe, Flash New Version, then cycle power on KFLOP.

Please let us know if it corrects the problem.

Regards

[mk00]

No, the behavior remains the same. Version 4.33m. This occurs during deceleration. On your graph axis in not marked, is this speed or position. This is my full init for this axis:

ch3->InputMode=NO_INPUT_MODE;
ch3->OutputMode=STEP_DIR_MODE;
ch3->Vel=10000;
ch3->Accel=1e+005;
ch3->Jerk=1e+006;
ch3->P=0;
ch3->I=0.01;
ch3->D=0;
ch3->FFAccel=0;
ch3->FFVel=0;
ch3->MaxI=200;
ch3->MaxErr=1e+006;
ch3->MaxOutput=200;
ch3->DeadBandGain=1;
ch3->DeadBandRange=0;
ch3->InputChan0=3;
ch3->InputChan1=0;
ch3->OutputChan0=3;
ch3->OutputChan1=0;
ch3->MasterAxis=-1;
ch3->LimitSwitchOptions=0x100;
ch3->LimitSwitchNegBit=0;
ch3->LimitSwitchPosBit=0;
ch3->SoftLimitPos=1e+009;
ch3->SoftLimitNeg=-1e+009;
ch3->InputGain0=1;
ch3->InputGain1=1;
ch3->InputOffset0=0;
ch3->InputOffset1=0;
ch3->OutputGain=1;
ch3->OutputOffset=0;
ch3->SlaveGain=1;
ch3->BacklashMode=BACKLASH_OFF;
ch3->BacklashAmount=0;
ch3->BacklashRate=0;
ch3->invDistPerCycle=1;
ch3->Lead=0;
ch3->MaxFollowingError=1000000000;
ch3->StepperAmplitude=20;

ch3->iir[0].B0=1;
ch3->iir[0].B1=0;
ch3->iir[0].B2=0;
ch3->iir[0].A1=0;
ch3->iir[0].A2=0;

ch3->iir[1].B0=1;
ch3->iir[1].B1=0;
ch3->iir[1].B2=0;
ch3->iir[1].A1=0;
ch3->iir[1].A2=0;

ch3->iir[2].B0=0.000769;
ch3->iir[2].B1=0.001538;
ch3->iir[2].B2=0.000769;
ch3->iir[2].A1=1.92081;
ch3->iir[2].A2=-0.923885;

[TomKerekes]

Hi mk00,

I'm not able to see a problem with the new code. The previous plot was of position.

Attached is the code to perform the move and capture the data to a file. It actually does a small movement first and then captures data for a large movement which caused the problem before the fix. Also an Excel Spreadsheet to plot it.

Could you provide an example motion that shows the error and capture a plot of the trajectory?

Thanks

[mk00]

Here the result. Motion abrupt at 6200 steps (beginning of deceleration?). Also I noticed that with 4.33m problem occurs with any movement, not only with small movements.

[TomKerekes]

Hi mk00,

Unfortunately that doesn't help. That looks to me like perfect motion all the way until the data capture ends at 4000 points or 0.72 seconds. What size move were you commanding? 8000 steps? Please increase N to around 8000 points to capture the deceleration. Is the physical motion jerky or something or are you thinking there is a problem based on the plot?

Thanks

[mk00]

I run your code without modification, yes it is short capture. Here is correct capture - like your first graph.

[TomKerekes]

Hi mk00,

That plot is indeed bad. But when I run the test it is always good. Are you sure you Flashed the new Firmware? That is exactly what I would expect with the old Firmware. You must cycle power on KFLOP after Flashing the new Firmware to be running the new Firmware. Please type "Version" on the Console Screen to verify what Version is running.

Otherwise I've attached the capture program with your parameters included please run that to see what you get. I've included the plot I get.

Attachment 285444

Regards

[mk00]

In the console is displayed 4.33m. May be it is wrong firmware file? Or revision of the board is important? I`ll run program tomorrow.

[TomKerekes]

Hi mk00,

My apologies. You are correct that is basically an unchanged file.


Please try this Firmware which should report as V4.33n
http://www.dynomotion.com/Software/P...l/DSPKFLOP.out

Regards

[mk00]

With 4.33n all is good, thank you Tom.

[TomKerekes]

Hi mk00,

Thanks for the feedback and your patience.

Regards

[slimneill]

How do we get this to work?

I installed 4.33l exe file and the console shows that version.

Then i copied and pasted the 4.33n file from post 16 above,into the Kmotion4.33/.../DSP_Kflop folder and replace the one that is in there.

Then i flashed the drive with this 4.33n file. It tells me to reboot the kflop. I turn power off then back on and i get a message with a red X circle and a message that it didn't work.

Is there a 4.33n exe file? Am i missing something?

[TomKerekes]

Hi slimneill,

Those should be the correct steps. There isn't yet a PC side Version greater than 4.33l. I don't know why it wouldn't work.

What message do you get that says it didn't work?

Did you try restarting KMotion.exe?

Does KFLOP boot? Does it communicate and can it report the Version as 4.33n?

Regards