G540 Z drive loosing steps?

[Tonytony]

Hi Mariss/Marcus,

I upgraded my X2 mini mill CNC from Xylotex to Gecko 540, since then I have noticed my Z axis loosing steps. I begun troubleshooting my mini mill for mechanical issues, I also replaced my parallel cable and finally run a couple of tests on my Gecko.
My step motors have a dual shaft so a installed a copper cable and a piece of masking tape to serve as an indicator of steps, then I moved my Z axis 3 inches up and 3 inches down, my indicator showed a different position that the original.
Then I interchanged Z for X on my Gecko and changed the pins on Mach, so now original Gecko X axis drives my Z axis and my original Z axis drives my X axis. I repeated the same test and did not loose steps on Z (I tried many times), I run the same test on my X axis (using original Z axis driver on Gecko) and had the same issue I had on my Z axis.
The problem follows the original Z axis driver on my Gecko 540, it seems to me I need to replace the Gecko 250 driver, please help.

[joeybagadonuts]

Tony,

Please keep us posted, Speed Custom has been plagued by a Z axis issue with his G540 for several months.

JoeyB

[Tonytony]

Joeybangadonuts,

I do not have an extra G250 to replace my Z drive on my G540, I want to hear form Mariss or Marcus before I send my G540 or just my G250 or further troubleshooting instructions. I am confident this is a hardware issue, but they may have a fix I can do from here.

[Mariss Freimanis]

Originally Posted by Tonytony Hi Mariss/Marcus, I upgraded my X2 mini mill CNC from Xylotex to Gecko 540, since then I have noticed my Z axis loosing steps. I begun troubleshooting my mini mill for mechanical issues, I also replaced my parallel cable and finally run a couple of tests on my Gecko. My step motors have a dual shaft so a installed a copper cable and a piece of masking tape to serve as an indicator of steps, then I moved my Z axis 3 inches up and 3 inches down, my indicator showed a different position that the original. Then I interchanged Z for X on my Gecko and changed the pins on Mach, so now original Gecko X axis drives my Z axis and my original Z axis drives my X axis. I repeated the same test and did not loose steps on Z (I tried many times), I run the same test on my X axis (using original Z axis driver on Gecko) and had the same issue I had on my Z axis. The problem follows the original Z axis driver on my Gecko 540, it seems to me I need to replace the Gecko 250 driver, please help.

What CNC program are you using?

Mariss

[Tonytony]

Mach3 R3.042.020

[phungi]

There are several other examples of this problem occurring with the g540, It's not just speedcustoms ya donut!

Try running mach3 in Sherline mode, it has worked for the "others"

Mariss strongly discouraged people from using software to mask hardware problems, I can't see the harm. 1/2 pulse may be slower, but perhaps you can run 45000kz or 75000 to compensate?

[Tonytony]

Mariss,

I understand your skepticism about my problem; I understand you may think I do not know what I am doing. In fact I am no expert on CNC, just an enthusiast like others in this forum. However, I am very comfortable with electronics as I am an electronic engineer with many years repairing all kinds of equipment since 1988.
I hoped you can see the troubleshooting steps followed by me trying to solve this issue and come to the same conclusions. I did not assume my Gecko was faulty, I concluded by the results that either G540 motherboard or Z drive G250 was faulty.
I just finished another test, this time I wanted to find if the motherboard or the G250 is causing this issue. Before I run the gcode, I positioned my step indicator (a piece of wire firmly attached to the second shaft on the motor) to align with a mark in a masking tape attached to the motor by jogging the axis on Mach3 and zeroing the coordinate on Mach3. After running the gcode my step indicator will either be aligned back to the masking tape mark or not indicating lost steps. This is what I did:

1.- Restored pin settings on Mach3.
2.- Connected back Z motor to Z driver and X motor to X driver.
3.- Tested both axis with the following gcode:

For Z axis:

Z 0
Z 3
Z 0

For X axis:

X0
X3
X0

4.- Verified Z has the same issue (loosing steps) and X has no issues (works perfect).
5.- Turned power off, disconnected G540, opened.
6.- Swapped Z drive G250 and X drive G250, closed G540, connected to my mill.
7.- Turned power on.

Now Z axis works perfect and X axis looses steps, the problem followed the original Z drive G250 don’t you think?

[Mariss Freimanis]

Please return the faulty G250 or, if you like, the entire G540 for evaluation.

You might want to check if step is set 'active high' in ports and pins, step pulse width is set to '2 microseconds' in motor tuning.

Mariss

[SpeedsCustom]

I had the same problems, I had problems with my Y-axis that was causing more serious problems the past month and half. I swapped the Y-axis G250 with a spare G250 and it solved my y-axis problems. I still have a z-axis issue, but mine seems to gain steps, I only say gain becuase to does not work properly cutting down, then rapids and I gain steps. I swapped the z-axis board with a different G250 and for the first day it worked fine! Now a week later, I am seeing z-axis problems still, small scale but still not working correctly.


I have spoken to Mariss and am sending the Gecko540 for full evaluation.

-Jason

[Tonytony]

Mariss,

I sent you my G540 for evaluation, thank you for your help.

[ihavenofish]

i have the same issue here with my Y axis. below a certain speed the motor simply does move. on the Y it was about 2.5" per minute. i swapped the Y and A axis drivers and now its fine.

the Z axis also has an issue below a certain speed, but its much lower at under 1ipm (211.66 steps per second on my machine). oddly i put the Z axis on the "old" Y board and it preforms slightly better. the Z uses a different motor type. i wonder if the boards are sensitive to certain types of motors. i should also note im using a 19v power supply, which is near the lower limit for the g540. all motors are drawing the full 3.5 amps (in theory anyway).

for me this is only an issue on low speed contouring where one axis is moving very slowly. on higher speed work there are no lost steps.

sherline half step mode works flawlessly and for the moment doesnt seem to have any ill effect on speed acceleration or accuracy. i dont know if theres any disadvantage to this mode other than perhaps more noisy motors.

[Mariss Freimanis]

The 19V power supply may be the problem. If it is an unregulated supply, your meter may say 19V but the voltage at the ripple trough may be considerably less. Try a 24VDC supply and see if the problems go away.

Mariss