KL-6050 Stepper Driver Frustration

[tjp912]

I have a G0704 that's been converted to CNC and am having quite a bit of trouble with the KL-6050 drivers in use. It took me some time to track this down, but basically the system is perfectly quiet and behaved (I'm speaking from an electrical point of view looking at the DIR and PULSE signals on an oscilloscope) until I power up the motor drivers. There's all sorts of oscillation present in the system once these drivers are plugged in at the oscillation appears to have a fundamental frequency of somewhere around 12MHz. The Y and the Z axis don't seem to care too much, but the X axis actual direction of travel is basically random relative to the commanded direction. If I command the bed to go right, it goes left 50% of the time and vice-versa. I know I'm not the only one that's using these drivers, but based on what I'm seeing from these EMI generators, I'm not sure how anyone is using them successfully.

Does anyone have any ideas for how to make this work or should I just cut bait and switch to a better driver (and I'm not sure what that might be)?

I have tried various different 48V power supplies for the motors including a high end lab supply, so I know it's not the 48V supply that's injecting the 12MHz; it has to be some harmonic of whatever their DSP update rate is within the driver.

Thanks for any thoughts.

Tom

Similar Threads:

• Need Help!- stepper driver using h-bridge to analog servo driver
• Need Help!- Replacing Syil Stepper Driver With Other driver. No Luck getting motion
• Need Help!- 6 leads stepper motor connect to TB6560AHQ stepper driver board
• as anybody bought the keling 6050 driver here
• STEPPER DRIVER FAQ {Stepper/driver amp per phase match.

[jalessi]

Tjp912,

What are you using for a breakout board?

Welcome to the Zone,

Jeff...

[tjp912]

The C10 break out from cnc4pc. It seems to behave just fine from what I see on the scope.

[jalessi]

Tom,

The C10 is very good, most likely not the issue.

Have you tried reducing the supply voltage down to 36vdc?

Jeff...

[tjp912]

I didn't even think about that...are these motor drivers sensitive to supply voltage? I'll go try it out but I'm curious if there's a reason why this might be worth a try.

[jalessi]

Tom,

Voltage might however there are several issues that may be the culprit.

Jeff...

- - - Updated - - -

Tom,

How are you providing 5vdc to the C10?

Jeff...

[tjp912]

Tom,

Voltage might however there are several issues that may be the culprit.

Jeff...

- - - Updated - - -

Tom,

How are you providing 5vdc to the C10?

Jeff...

The lowest I can set the supply to is 40V and that had no effect on the problem.

5V is presently supplied by a bench top lab supply...very clean.

[jalessi]

Does the C10 have screw terminals or the new spring clip connectors?

Is the C10 on metal standoffs?

Have you tried moving the offending drive to a different axis connection on the C10?

[jalessi]

Is it possible you may have a loose coil wire on the offending stepper motor?

[jalessi]

Do you have limit switches or a touch probe connected to the C10? They sometimes act like antenna's.

[tjp912]

Screw terminals...

Metal stand-offs, but attached to a piece of acrylic...

I moved the connections to the C10 and the problem followed the driver. I switched motors and the problem followed the driver. I replaced the driver and the problem was still there.

[tjp912]

Seems unlikely since the problem appears on 2 different motors with the same driver. I'm starting to wonder if there is a physical distance problem between the driver and the C10 where the driver is just too close to the C10 and there's cross-talk somehow.

- - - Updated - - -

No limit switches yet. Just the motor drivers.

[jalessi]

Are you using shielded wires anywhere? Is it possible there is a ground loop?

Is the power wire provided to each driver individual or daisy chained?

[jalessi]

Are Mach3 step and directions pins configured active low?

[jalessi]

I have seen metal standoffs short out a portion of the C10 board on one pad.

[jalessi]

Tom,

I have to leave for the evening, check back with you tomorrow.

Jeff...

[tjp912]

Wires are not shielded since there's no place to tie off a shield, but are fairly short (8-12 inches). I've reviewed the interconnect a few times and don't believe there's any sort of a ground loop.

Power is provided to each driver separately from a separate stud on the power supply.

[tjp912]

They were...I switched the logic level around and now they're not. This made no difference in the observables.

I'll check out the possible shorted pad idea on the C10. I've also ordered new driver ICs for the C10 in the off chance that the ICs were faulty, but given the fact that the C10 is doing exactly what it's supposed to do until the drivers are powered-up, this doesn't seem likely.

[jalessi]

Tom,

If you remove all connections from the 6050 drives and only connect just power (vdc supply) to one at a time without the C10 connected is the noise present on the scope?

Jeff...

[109jb]

I have KL-6050 stepper drivers on my G0704 and have had no problems. I use a 48V computer server power supply, so I can tell you that 48V should not be a problem. I have a different breakout board, but my breakout board is about 3 inches away from the drivers and the wires are just long enough to reach between them.

Since you are having problems with the X-axis and the Y and Z seem to operate correctly, have you tried any of the following:

1. Test the offending driver on the Y or Z axis.
2. Swap the signal wires from the C10 from the X to another axis
3. Hook the driver to the pins on the C10 for A axis and then change pin assignments in software.

From your first post, I couldn't tell if this is a new build, or if this was a previously working setup.