Motor with too little torque

Hi RepRap community,

I recently acquired a Prusa I3 kit and started building it.

Everything went good so far until I got to the motor part. I configured everything and when I first tested, the motors just kept vibrating without doing any effective work. I tried removing the motor from the printer and see if it worked by itself and, to my surprise, it did. I'm guessing that I have some kind of problem with the torque of these motors even though I saw that they are somewhat used without problems by other Prusa users.

I'm attaching two videos that show the problem I'm having:

• Motor not working
• Motor working

For now I'm just testing the X axis to reduce the number of variables.
I'm using a pololu A4988 driver and a 42BYGHW811 motor.
Things I think are relevant:

• I'm using Marlin and I made a configuration change from
  

   #define DEFAULT_AXIS_STEPS_PER_UNIT {80, 80, (200.0*8/0.8)*1, 1178}

  to
  

   #define DEFAULT_AXIS_STEPS_PER_UNIT {53.33, 53.33, (200.0*8/0.8)*1, 1178} 

• Calibrated the voltage coming from the stepper driver to around 0.95V

I got to say that I blindly tested lots of other values with no luck. To be honest I don't fully understand all the electronics behind the motors and that's the main reason I'm stuck and feeling so frustrated.

Any help would be highly appreciated.
Thank you.

What is different between when you tested the motor on the printer and when you tested it off the printer? Cable? Load? Driver?

Hi,

The only difference is the load, it works fine if it has nothing to move and fails when the motor is plugged into the belt and has to move the extruder. That's the reason I'm thinking the problem is probably a low torque. Still sounds weird to me though, as this motor and driver models have been used for a long time in prusa printers as far as I know.

If I try moving the extruder by hand I can tell that it moves smoothly and the element that offers the more resistance is the very same motor that is not working fine. Is the first time I play with one of these so I can't really tell if the movement is smooth enough.

One thing that I haven't tried is to put some oil in the rods but, as I said, I don't think it will make such a difference because it feels like the motor itself is the one element making it harder to move.

If you can think of any other test to know where the problem is I'll be glad to try it.

Thanks for your help.

It could be that Marlin is trying to move or accelerate the motors too fast, either because you have it configured incorrectly, or you do not have the jumpers in your electronics board set for 16x microstepping.

---

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].

Hi again,

I double checked the jumpers and they look ok (I also tried removing them 1 by 1 to see if they were changing the behaviour of the motor and because they do I deduce that they are doing something).

I attached my marlin configuration, is the one I found in this repository with the modification I commented earlier. I understand there are more recent versions of Marlin but I couldn't make them work at all with my printer. Is there any definition in marlin I must know to give the motor more strength and/or reduce its velocity?

Thank you all for your time.

Do the other motors in the printer work? If so, swap in the motor in question and see if it works in another motor's position.

I just swapped it for the extruder one and I have the very same problem.
From this we can assume it's either a bad marlin configuration, a badly calibrated stepper driver or a problem with RAMPS 1.4 (or every motor has the same problem but I found that very improbable).

I'd appreciate if someone with a Prusa I3 (RAMPS 1.4 + A4988) post his Marlin configuration to start discarding issues.

Thanks.

How did you set VRef on the driver? It sounds like you are underdriving the motor . . .

- Tim

VRef is about 0.95V, that's the value I see working better when not attached to the belt.

I've already tried increasing it to 1.1V but the motor start skiping some steps and more than voltage that the stepper driver just decides that it's not worth it and stops.
To keep track:

• Stepper drivers are A4988 with Rs=0.1 ohm. Not sure if pololu, stepstick or any other.
• I'm using the three jumpers for each motor (for x16 microstepping).
• Motors are Wantai Model 42BYGHW811. 2.6A (I've read that motors with that intensity might not be as good as others), 1.8deg/step.
• Incoming V from the power supply is 12V

I included some photos, maybe I'm missing something obvious that I'm just incapable of seeing.

RAMPS 1.4
Motor

Thank you all for your time.

I am unclear what motors you are having problems with. In your first post (the topic) you use plural (motors) but later and in your photos you just talk about one motor. In short, I would like to understand if;

Is it correct that all other motors are working except for the X axis?

Is it correct that you tried to swap the working extruder motor to the x axis and in that position and loaded it does not work?

Further -

In your Marlin config file you set the steps per mm to 53.33 - This is wrong, it looks like you have a 20 tooth pulley so the correct setting for that is 80. This should however have made it easier, not harder for your stepper.

Try to move a working motor at the same time as you are trying to move the not working X axis. See if that makes the working motor stop also [1]

Have you tried swapping the driver (A4988 ) - So take the working extruder driver (A4988) and put it in the x-axis place and see if that works?

Are you sure that the extruder, that you conclude is working, is actually really working when loaded?

Try to leave the extruder connected as extruder but put it to drive the X and see if it moves when you ask it to "extrude" - Remember to adjust step size.

How are you testing the movement? Try to use this command G1 X5 F3 - That will make it go 5 mm in the X direction with a feed of of 3mm per minute. That is very slow. If it can't move with that, it can't move.

Try to block one of the working motors and see how that shakes when it is blocked. Now compare that to how your X is shaking when it is blocked. If the motor is getting full power then it will feel roughly the same. If it is a circuit problem it can (but not necessarily must) be different.


And thats it - That sounds like a lot of tests in all sorts of directions but each test have a purpose. For example the test in [1] shows us if the error is bringing down the entire system (then it is probably RAMPS or Arduino or power supply and not software).


Finally - A good advice - You are probably a little stressed about this right now, and it is in that state of mind that I myself start to do not so smart things. Like forgetting to turn off power before moving around cabling. Remember to remove both USB and 12V power when doing changes. I had to burn one Arduino before I learned that, not recommended.

Hi,

Sorry I didn't keep it clear from the beginning. When I first tested I just plugged in all the motors and tried to move them one by one with pronterface. Some of them just didn't move (Z axis) and others did but erratically. In subsequent tests I just focused in two of the motors (X and extrusor) to keep it simple debugging the problems. Both of them behaves similarly so I deducted the problem wasn't in the motors. I can still play around with the others.

I understand the proper steps per mm was 80, I just couldn't make it work while playing with the VRef, the motor was just (what I think it was) skipping too many steps and either changing directions non stop or not moving at all.

I tried what you suggested, moving two motors at the same time (X and Y) but I'm having some problems because they won't move at the same time, the motor X will try to move and when that's finished then the Y motor will try the same (with no luck for both).

I also swapped the stepper drivers a bunch of times with the same results.

The extruder is working just a bad as the others, sorry if I didn't make it clear, let me explain myself. I'm just trying basic axes movements, I didn't start playing with the extruder yet and what I'm trying to say when I mention the extruder motor is just that I replaced the motors for the one I was originally using for the extruding part.

In the beginning I tested several movements but in the end I'm just asking the motors to go home, I didn't know about sending them direct orders and my only tool for this has been pronterface. I tried sending G1 X5 F3 to both motors with different steppers and they do the same, I included a video of the results. I didn't try but I highly doubt they will be able to move the belt.

I know, I'm a bit frustrated with all of this and I already broke some A4988 by doing exactly what you said, maybe in the process I also damaged the whole Arduino or Ramps and that's why I'm having all these problems.
I'll do some more tests as the ones you suggested and if nothing works I'll think seriously about replacing the whole board.

Thank you very much for your time.

If the motors are really 2.6A as you wrote, and your drivers have .1 ohm sense resistors, the your correct VRef as calculated would be closer to 3 volts, not .95, so you are massively underdriving. I also keep asking how you are setting VRef, since I keep seeing comments about seeing voltage output, and with motors running, and none of that is correct or relevant when setting VRef . . . VRef on a Pololu is nothing more than measuring from the pot wiper to ground and setting a value . . . and it is derived from logic power, and is static . . . you don't need to have 12v or even a motor connected to make the setting . . . if you are measuring anything directly connected to the motor, you are measuring nonsense . . .

- Tim

Hi,

I understand what you say but my pololu will only get to 1.2V, more than that and they stop working. I'm measuring VRef with a cheap multimeter I got, I basically place the negative pin in ground and the positive one in the trimpot of the pololu, I took a picture for reference.

Hello,

Ok, your situation is a little different from what I first imagined.

Some important points which are not always 100% that way but which can generally be counted on:

* Digital electronics fails the same way as it works - As 1 or 0. Either it works or it does not work. If it is a integrated circuit error (Arduino or Driver) then the error will not be load dependent - It just would not run at all.

As such I do not think the Arduino is the problem.

* You motor should be able to drive just fine at Vref 1 V. If you can move you carriage with your little finger (applying force but without hurting yourself) then you should not have a problem at all. Even it will be able to move much more then that.


Based on the new information I will point to two potential issues:


1) Your power supply is messing things up for you. The easiest way to test this is to try a different power supply. If you do not have a different power supply handy you need to do the following:

a) Try to heat the hotend and bed and see if the voltage drops.
b) Try to put your voltmeter to AC voltage and measure your terminals (on the RAMPS). 0.5V AC would be expected. Several volts indicate we found the problem.

You say your power is staying at 12V but what I am looking for is leaking AC . Leaking AC is when the PSU is actually not working. It allows ripples from the AC to go into the DC and that crashes your drivers. If you put full load on the PSU (heatbed + hotend) then if that is the problem it will crash the PSU (the voltage drops or shutdown)


2) Your RAMPS board might be broken. In particular I am thinking a leaking capacitor. That can give all sorts of erratic behavior.

Edited 1 time(s). Last edit at 07/05/2015 03:08AM by LarsK.

Hi,

That are two things I didn't even consider, I'll try to do those tests and report later on this week. I have a fully functional power supply running my computer right now so that won't be a problem.

Thanks