Y axis shift

Hello, I just finished building a machine. Everything is square, belts right, bearings new, motors new etc.. I am using an acceleration of 2500 mm/s2. Not too high, but not low either. The machine should be able to handle this easily. The bed is under 1 kilo. X carriage under 350 grams.

I use 20T GT2 pulleys. The Y axis skips every now and then, I am not sure if the belt is slipping because the belts are tight. I reduced my accelerations to 2200 to try and solve the problem. I also touched my motors and although the motor is warm to the touch, it isnt hot. I brought my driver voltage down to 0.54V from 0.63V hoping this would keep the motor from skipping. It still skipped, this time the gap was large. Almost makes me want to say a belt tooth large, but I just cant see the belts slipping since they are plenty tight. The motor runs cooler now on the lower voltage, but honestly I think I should be able to run more power to the motors since I am using active cooling with heat sinks on my drivers and 2 amp NEMA 17 motors. I dont want to have to keep reducing my accelerations, the entire point of this machine was to run upwards of close to 3000 mm/s2 all day.

Edited 1 time(s). Last edit at 04/17/2016 03:04AM by dbeck002.

Hi DBeck,
There are a lot of reasons that it could miss steps. I pulled together the Shifted Layers Page to document many of them.

Y axis missed steps are often because of an imbalance in the relationship between driver current, mass, torque, speed, and acceleration.

When you lowered the voltage, you were actually lowering the stepper motor current. The fact that it skipped worse may therefore indicate that the current is not high enough.
It's OK for the stepper motors to get quite warm. You can always add a fan to them if you are concerned with the heat - it doesn't take much to cool them. You can also purchase a driver that can handle higher current. Note that the heat sink pad on the driver chip is on the bottom of the chip package - soldered to the PCB.

Since you are using active cooling and heat sinks on the drivers you should be able to turn that one up a bit more. More current = more torque for a stepper. Too much current and the driver will be too hot to keep a finger on for a couple seconds - and it could start doing thermal resets - that can last fractions of a second or longer, and could occur when the motor is stationary too - you may hear a clunk sound when it re-engages.

Hope that helps.

---

My printer: Raptosaur - Large Format Delta - [www.paulwanamaker.wordpress.com]
Can you answer questions about Calibration, Printing issues, Mechanics? Write it up and improve the Wiki!

Quote
Paul Wanamaker
Hi DBeck,
There are a lot of reasons that it could miss steps. I pulled together the Shifted Layers Page to document many of them.

Y axis missed steps are often because of an imbalance in the relationship between driver current, mass, torque, speed, and acceleration.

When you lowered the voltage, you were actually lowering the stepper motor current. The fact that it skipped worse may therefore indicate that the current is not high enough.
It's OK for the stepper motors to get quite warm. You can always add a fan to them if you are concerned with the heat - it doesn't take much to cool them. You can also purchase a driver that can handle higher current. Note that the heat sink pad on the driver chip is on the bottom of the chip package - soldered to the PCB.

Since you are using active cooling and heat sinks on the drivers you should be able to turn that one up a bit more. More current = more torque for a stepper. Too much current and the driver will be too hot to keep a finger on for a couple seconds - and it could start doing thermal resets - that can last fractions of a second or longer, and could occur when the motor is stationary too - you may hear a clunk sound when it re-engages.

Hope that helps.

Thank you very much for the thorough reply. I appreciate it.

I will boost my current back to where it was before. I am currently using the biggest muscles you can get, NEMA 17s rated for 2 amp, 1/16 step Black edition pololu drivers, thermal adhesive with heat sinks and a massive 90mm fan blowing directly over the ramps with a dedicated enclosure. Heat is not an issue for my ramps.

The motors can be turned up quite a bit, they are made for this. I was just scared that maybe they were too high, they really were not hot just warm. I think there is more left in them. But the shifted Y axis phase is eluding me.

That's a pretty high acceleration for a moving bed design - I run 800. You might want to start low and work your way up once you have the machine running reliably. Same thing with XY jerk. Personally I hate the way the printer behaves with high acceleration and jerk settings, I'd rather wait a little longer and have things run smoothly, but I understand everyone has different priorities.

Quote
JamesK
That's a pretty high acceleration for a moving bed design - I run 800. You might want to start low and work your way up once you have the machine running reliably. Same thing with XY jerk. Personally I hate the way the printer behaves with high acceleration and jerk settings, I'd rather wait a little longer and have things run smoothly, but I understand everyone has different priorities.

I totally understand. I would run 800 but I have time constraints for my prints and every minute counts. My goal was to run 3000. I designed this machine to be low mass, albiet the nature of a moving bed makes high accels difficult but I had no choice, it is also a very compact machine. I really think I can boost motor current and this should greatly help. I will give that a try.

Yes, make sure the driver Vref is set correctly for the current your motor is designed for, and make sure that the axis moves freely. But at the end of the day, you have little alternative to building a system and then testing it's limits to see if it meets your criteria, and the way to do that is to start well within the limits and then increase the parameters one at a time to see how it behaves.