Strange behavior on Y axis

I'm running the latest version of Marlin on an Arduino Mega with a RAMPS 1.4 controller. The printer itself is home-brew.

I'm still in the early stages of learning how to use this equipment, although the frame and steppers were from a CNC mill that I later converted to a laser cutter, so I'm very familiar with G-code. I've put larger steppers on the X and Y axes, they are from Stepperonline, 17HS19-2004S1. They behave well in all the manual tests I've done (sending G1 codes through Pronterface) and have printed several small "calibration" objects flawlessly.

Now I'm trying to print a case for the LCD control unit, and I'm seeing the Y axis shift to the right bu about .5 mm intermittently when the Z axis increments. I've tried different current settings on the A4988 thinking that the Y stepper wasn't getting enough current when the X and Z axes were both moving at the same time, but it hasn't helped, and I can't duplicate the problem manually with G-codes. I've also tried a couple of different designs, but they all exhibit the same problem. The first layer prints OK, then the perimeter of the case is printed at layer 2, but the Y axis is shifted slightly to the right. If I let the print continue, it will print several layers at the new Y-offset, then shift th the right again.

Any ideas, please?

Walt

Sounds like you've tried the obvious.

Can you hear it missing steps when this happens? Does it happen at different print speeds? Can you print the object on a different portion of the bed to see if it behaves the same to ID a possible binding problem? Does it happen on different objects?

Most likely your Y axis is missing steps at the start of a travel move. So you need more motor torque or less acceleration.

Those motors are rated at 2.0A, which is almost double the current you can get from most plug-in drivers. So with RAMPS and plug-in drivers, you won't be able to run them close to maximum torque without the driver overheating.. That leaves reducing Y acceleration as your best option. Or switch to a board with integrated drivers that can be cooled properly.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
Most likely your Y axis is missing steps at the start of a travel move. So you need more motor torque or less acceleration.

Those motors are rated at 2.0A, which is almost double the current you can get from most plug-in drivers. So with RAMPS and plug-in drivers, you won't be able to run them close to maximum torque without the driver overheating.. That leaves reducing Y acceleration as your best option. Or switch to a board with integrated drivers that can be cooled properly.

Thanks both of you for your replies. No, I can't hear the skip but I'll listen more closely. I'll have to try to get the problem to happen on some smaller objects, since it takes over 10 minutes on this particular object before the problem first appears.

I thought the current rating on the motors was a maximum, not a rated draw. What you say about acceleration makes sense. I've read all sorts of different values for acceleration, without much of an explanation as to WHY the particular values were chosen. I've been homing in on acceleration and jerk as likely causes of the problem. I've already reduced acceleration from the Marlin default of 3000 to 500, and jerk from 20 to 4.

I've also read some confusing articles about setting the current on the A4988 controller, by measuring the voltage at the potentiometer wiper. Again, I don't know if the current you use in the calculation is based on the maximum current or the desired running current.

It looks like I'm in for a lot of experimentation!

Thanks for your input

Walt

Stepper motors in 3D printers are generally run at between 50% and 90% of their rated current. Too little current and you won't get very accurate positioning, due to insufficient torque and the effect of detent torque. Too much current and the motor will overheat (if the drivers don't overheat first).

The specified holding torque of a stepper motor is obtained when you run it with full stepping at its rated current. Once you turn on microstepping, the holding torque at full current reduces to 71% of the specified value, more or less independent of what microstepping you use. If you reduce the current below the rated current, the holding torque (and also the incremental torque per unit angular error) decrease in proportion to the current.

So it is best to choose stepper drivers than can drive the motors up to at least 80% of their rated current; or if you are set on using particular drivers, to choose stepper motors with a rated current no more than 25% higher than the current that your drivers can realistically supply. But don't choose low current high inductance motors, because then you run into the problem of needing a higher supply voltage to maintain current at higher speeds.

As for setting motor currents using a potentiometer and voltmeter, this was ok in the early days of 3D printing Z but IMO is long obsolete and shouldn't be inflicted on 3D printing novices. All the more advanced electronics with integrated drivers let you configure the motor currents in firmware, typically using a configuration file stored on the SD card.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].