[hautamak]

Hello,

Have had my cnc complete for months and had run some programs with it, mainly engraving. The thing is, Z-axis seems to loose steps, either up or down..
also slaved A-axis or X-axis (dual stepper X-axis) is missing steps causing the gantry do some twist. seems that Y-axis have not loose steps, or I dont only know it.

Here is what I have tried so far

1. Played with speed and acceleration of motors.
2. Changed to shielded cable from motion controller to stepper driver
3. playing with different kernel speeds and sherline pulse mode
4. checked mechanical jams

I wrote test code today for z axis and first it was slowly losing steps and dropping. Then I created another test code and it was slowly rising?????

First code moved Z-axis down as Y and X wasn't moving and Z-axis up with X and Y axis also moving
Second code moved Z-axis down with X and Y also moving and Z-axis up individually Y and X still.

Do you have any thoughts or suggestions ?

I have this chinese usb smoothstepper

https://i.pinimg.com/originals/d8/ba...5d57b06a3b.jpg

15A power supply 24v

nema 23 motors

tb6600 drivers

[philba]

what's your backlash like?

[ger21]

Drop the acceleration very low, and see if you still lose steps.

[hautamak]

Hi, there is some backlash, but if I left mill running that test code which slowly looses steps and rises, it will raise without limit...

I have already lowered it but will do some more testing with lower values.

[ger21]

What values are you using for accel and velocity?

Backlash has nothing to do with losing steps.

[hautamak]

What values are you using for accel and velocity?

Backlash has nothing to do with losing steps.

400mm/min velocity and acceleration 100mm's/sec
Z-axis have 2mm pitch leadscrew

[awerby]

24 volts is really low; it's surprising this works at all. Replace the power supply with one that puts out 36vdc and it will probably work better. Those drives are supposed to handle 42v, but that's probably something you don't want to push to the max.

[hautamak]

24 volts is really low; it's surprising this works at all. Replace the power supply with one that puts out 36vdc and it will probably work better. Those drives are supposed to handle 42v, but that's probably something you don't want to push to the max.

https://www.reprap-3d-printer.com/pr...-driver-tb6600

This site recommends 12-24volt supply. those drivers worked good with 12 volt also, so I don't think thats the problem

[awerby]

How much inductance do those motors have?

[joeavaerage]

Hi,
those TB6600 IC's as used in your driver have a long and checkered history. A Gecko 540 would improve your experience.

Craig

[hautamak]

How much inductance do those motors have?

https://www.ebay.com/itm/-German-Shi...r/272109517319

Inductance: 3.5 mH/phase

[pippin88]

Optimal voltage for stepper with 3.5mH inductance is around 60 volts.

Square root of inductance x 32 = optimal voltage

[hautamak]

Well well, back to this. Today run some test with 1/32 microstepping and every axis was losing steps, like 1mm. Then I set all drivers to 1/4 stepping and seems that I didn't loss any steps.. So far so good. Also ramped speed and acc up as I was not loosing steps and still seems that no lost steps. I'm still buying gecko at somepoint.
Worth mentioning that I swap mach3 from laptop to my gaming pc and run tests with it. problem seemed to be with microstepping and drivers or simply USB motionboard is not fast enough for 32 microstepping.

Hope this helps somebody.

[wmgeorge]

Its my understanding that the coarser setting like you did gives a stronger pulse. I think you found the solution.

[pippin88]

Some parallel ports are borderline on their voltage output and this can cause issues