[timbochung]

Hi,

I have finally connected my system (see pic)but am unable to make the motors move. Not sure what to do next...

- I've configured channels (0,1,2) to outputs (8,10,12) respectively (see pic); All other setting are identical otherwise; The channels have each been downloaded manually to the board

- In the step response screen, I can get the motors to do a tiny jog (<0.5sec) but unable to do any other movements otherwise (see pic)

- No success in using the jog function in KmotionCNC either.

Help!

Thanks,
Tim

P.S. I've uploaded my I/O map from a previous post

[TomKerekes]

Hi Tim,

Use the "Move" button instead of the "Step" button on the Step Response Screen for Steppers. A step will be too high of a frequency and the motors will just stall if the driver even sees the pulses.

I see the C10 board needs an Enable signal to be wired to +5V did you do this?

If you power cycle KFLOP so nothing is configured you could go to the Digital IO screen and set IO8 as an output and toggle it. You could then check if the motor steps and use a volt meter at various places to see where the signal toggles.

Regards

[timbochung]

Thanks for the reply Tom.
- The C10 ENABLE is powered up (took me awhile to figure that one out...)
- I did do a voltmeter reading to confirm a signal (around 5v). I went as far as the pins on the c10 board to make sure that my KFLOP DIO was outputting to the correct C10 pin
- KMotion settings look alright?
- I guess my next step is to examine the wiring between the c10 board and drivers and then between drivers and motors

Tim

[TomKerekes]

Hi Tim,

I isn't clear if you understood or performed the steps in the previous email.

When you toggle the Step Output in KFLOP's Digital IO Screen. Does the voltage on the pin go High (>3V) and Low (<0.4V)? Does the motor make one tiny Step?

Regards

[timbochung]

I did an I/O check where I would toggle a KFLOP DIO and then check if the right pin had voltage. I've traced/checked the reading at several points, ending at the c10 board
- I enabled a DIO
- Checked if the right pin on the board had voltage
- Checked if the right pin on the cable had voltage
- Checked if the right output on the C10 board

I did this check for all of the DIO I setup so I'm pretty sure everything between the KFLOP up to the output on the C10 is solid.

From the C10 board to drive and drive to motor, something might be fishy but I'm not sure where.

Since I'm pretty much a newbie to KFLOP I would like to confirm that with my current channel configuration in KMotion, and if all the hardware was wired properly, should I be able to press move or jog and see the motors move? Also, should the jog buttons in KMotion CNC be working as well?

Tim

[TomKerekes]

Hi Tim,

If the signal on the Step Input the C10 board is toggling but the motor is not physically stepping then the problem is further down the line.

You didn't tell us the voltage levels.

Are you sure the motor is not stepping? It will be a small amount. But after a few steps it should be noticable.

Are the motors energized? If so they will be hard to turn (relative to power off) and getting warm.

If Configured as Step/Dir and OuputChan0=8 and no Limit Switch options the Step Response Screen "Move" button should move the motor. Don't use KMotionCNC at this point.

Regards

[timbochung]

Hi Tom,

I did another voltage check and when I toggle DI/O #8 and check C10 pin #2 - the reading goes from 0V -> 4.84V.

The motors are powered up because I have observed the motors move on occasion.

This is so weird - just tried the move button on Channel 2 and got a motor to move. Changed channels to check the other axis - no go, then came back to Channel 2 and no go

Went back to the DIO page to check the voltage readings again and wasn't able to toggle the state of the DIO8. Had to unpower board to be able toggle again.

[timbochung]

So I cycled the power the KFLOP board to be able to toggle the DIO - this time I was able to get a motor to move (sorta). Movement is very strange and erratic.

https://www.dropbox.com/s/p567sq7sao...ment_sorta.wmv

[TomKerekes]

Hi Tim,

If you are getting 0 - 4.8V Toggle on the Step signal then the wiring is ok up to that point. You forgot to tell us if the motor steps when you toggle the bit.

The reason you can not toggle the IO from the digital IO screen after you try a move is because when you configure a Step/Direction Generator to drive the pins it takes over control of the pins. The simplest thing to do is as you discovered cycle power to return to the default state.

You didn't post your settings and I cant read the screen in your video, but it sounds like you are just trying to move way too fast with way too much acceleration so the motor stalls.

Try setting the Velocity to 1000 steps/second with acceleration of 4000 steps/sec^2 and Jerk of 1000000 steps/sec^3

Regards
TK

[timbochung]

Hi Tom,

Still no luck. I called my motor and driver vendor and had me do the test where I spin the motor shaft when it is unpowered and then try to do the same thing when powered. Since I wasn't able to move the shaft - he says that the motors are wired correctly. Verified my dip switch settings from other threads on the forum so they should be alright.

I've traced the signal from my C10 board all the way to the driver connection and I do see a change of 5V when I toggle my DIO #8 in Kmotion.

In theory, when I ask Kmotion to step or move - it should be sending a pulse to my C10 board which in turn will send a pulse of 5V to the driver. How fast is this pulse - would I be able to see it on my voltmeter?

BTW, I also tried the settings for the VAJ and still no success

Thanks,
Tim

[timbochung]

Also, do I need to upload or flash the KFLOP board each time I powerup? I have no scripts or programs whatsoever at the moment...

[TomKerekes]

Hi Tim,

What kind of Stepper Drives do you have and where is a link to their specifications?

But regardless, if you toggle the Step Pin and +5V and 0V is toggling on the input to the drive. The motor should make one single micro step each toggle cycle. Do you understand what one micro step is? It will be a very small amount. But if you do it a few times you should see and feel it. I have asked this before. I'm not clear on your answer.

Regards

[timbochung]

Hi Tom,

Below is a link to the stepper drives:
http://www.automationtechnologiesinc...s/KL-5056D.pdf

I have been toggling the PUL+ and the DIR+ inputs and have verified the voltage change of 5V at the drive inputs.

After toggling about 5 times, I haven't seen any movement whatsoever. How small is this microstep? I believe the drives have a 1.8deg resolution - does that mean for each pulse it should move 1.8deg?

Thanks,
Tim

[TomKerekes]

Hi Tim,

No. 1.8 degrees is a full step. Each Step pulse will only move one microstep. Those drives have programmable resolution. What have you selected? See page 9 of their manual. Depending on the resolution you have selected you may need to toggle 10 ~100 times to see 1 degree of rotation.

Those drives are incompatible with KFLOP because they require 5us of Direction Setup time. KFLOP only provides a maximum of 4us. Most drives require much less. But that is not the reason your motors are not moving. This problem can cause a single microstep in the wrong direction every time the direction reverses. I'm not sure what to tell you. We could debug your wiring and configuration and get the motor to move smoothly, but it may never work reliably anyway. Another User has these same drives and he is adding a simple RC+Schmitt trigger circuit to delay the pulses by 1us so it meets the drive specification. But that would be difficult for you. Do you have that level of electronics skill? If you wish to return the KFLOP you can certainly do that. I don't have a problem with helping you get the motors to move smoothly to see if there really is an issue with the direction change.

What are your thoughts?

Regards
TK

[timbochung]

Hi Tom,

Alot has happened in the past hour. Another user sent me a link to confirm my wiring and I realized that my wiring for the DIR+/DIR- & PUL+/PUL- was reversed. Fixed it and voila! Confirmed the microstepping with all three drives (toggle a DIO and the motors makes a tiny movement, can't really see it but can feel it/hear it). Also got my KL-8056D drive (Z-axis) working:
https://www.dropbox.com/s/0gke02kvo3...20KL-8056D.MOV

Unfortunately, still wasn't able to get my KL-5056D drives working and after reading your post - very confused/conflicted.

Below is a link for the KL-8056D driver:
http://www.kelinginc.net/KL-8056D.pdf

Are the control signals requirements the same as the KL-5056D driver? If so, wondering why I can't get it to move? I double checked that the wiring and settings are the same as the KL-8056D...

I'm not sure if I want to return the KFLOP just yet - especially after the effort I put into setting it up. After hearing all the good things about it I was really looking forward to seeing it in action.

Tim

[timbochung]

BTW - looking into this RC+Schmitt circuit. If I can't build. maybe I can purchase it?

[TomKerekes]

Hi Tim,

As we stated earlier please use the "Move" button not the "Step" button for Stepper Motors. A Step Motion is basically an infinite speed command which is guaranteed to just stall a stepper (and actually the pulses will be too fast for your Driver to see).

The Drive Inputs are the same so you must have a wiring or a configuration issue on the drive that doesn't work.

To troubleshoot do the bit toggle to check voltages and your wiring.

But by default KFLOP outputs a 2us pulses. These drives require a 2.5us minimum pulse. To change the KFLOP pulse length to max 3.78us run the C program shown below.


#include "KMotionDef.h"
main()
{
FPGA(STEP_PULSE_LENGTH_ADD) = 63 + 0x80; // set polarity and pulse length to 4us
}

Regards

[timbochung]

Hi Tom,

That C code snippet did the trick and and I've been able to get all the drives going. Finally!

Tim

[TomKerekes]

Hi Tim,

Yahoo!

You never answered my question on what microstepping you have the drives set at.

But now you might mark a motor shaft somehow. Then move the motor forward and backward several hundred times. The motor shaft should return to the same place. If the Direction is not always being recognized properly there will be a small drift in position.

Here is a program to move back and forth 100 times that might help:

#include "KMotionDef.h"

#define AXIS 0
main()
{
int i;
for (i=0;i<100;i++)
{
MoveRel(AXIS,1000);
while (!CheckDone(AXIS)) ;
MoveRel(AXIS,-1000);
while (!CheckDone(AXIS)) ;
}
}

[timbochung]

Right now it is at the default which means I don't know - I have run at 400, 800 and 1600. The vendor recommended 1600. For our experiment - which setting should you recommend?