Stepper motors vibration with RUMBA

Hello,

I'm using Nema 17 (SY42STH47-1684A) steppers, A4988 drivers with 16 microstepping on my Delta 3DR build. When steppers move or they are not moving but "loaded" they vibrate and transmit it to the extruder so all my printed parts have not full linear and regular walls.
I think it's a problem with my setup but I can't find why... I have a RUMBA board with jumpers to select the microstepping mode, I have all of them inserted as I read this is the correct setup for 16 microstepping. I also have tweaked the drivers pot to 0,15V because the stepper motors are less noisy that way and because, if not (0,5V as people suggests), they are too noisy compared with other Reprap printers I own and because I think they are not operating fine at all (probably because the vibration origin).

Do you know what I'm doing wrong? do you know from where this vibration is produced? is there any other configuration I should add to the RUMBA board to configure it correctly with Nema 17 and A4988?
Thanks

I've sometimes found motors vibrate when one of the coils is back to front.

eg: Say your wires are the following:

Red, Green, Blue, Yellow.

You may find that switching the Red/Green may improve things, or switching the Blue/Yellow.

Note: This will switch the direction the motors spin, so you need to compensate for that in firmware.

If it's not that, then it looks like you might have some bad motors. I've also had similar things happen if the motor shaft bearings get damaged (eg: the motor gets accidentally dropped on the drive shaft, if you've put a flat on the shaft yourself and not protected the bearings from metal dust, etc), where it then runs quite rough and vibrates.

Unless you feel really confident opening a stepper motor and changing the bearing over, which is not really a simple task, then replacement is the only option.

PS: If motors aren't packed right, so that they can't move and can't be dropped on their shafts, this sort of damage can happen in transit. No one in the postal/shipping services cares what's in your box, and they can't be expected to know either. Packing motors properly is one of the things I see a lot of suppliers fail at, as they don't expect the postal/shipping services to be as brutal as they can be, nor to they think that things like stepper motors can be so fragile as they can be.

Hi Cefiar,

Thanks for your comments. I have tried switching wires and they vibrate the same. There is a possibility that you are right about that they may be dropped while in transit to delivery, but if we don't consider it I have some other tests:

• if I lower the driver pots the vibration gets lowered
• the lower the speed the more it vibrates
• and finally, I have tested that it vibrate considerably more when performing curves than lines

I have the feeling that it's not due to a semiwroken steppers but a bad setup of RUMBA, drivers or even firmware...
Any other ideas?

Just an idea, maybe this is what you mean by vibration or noise: [hydraraptor.blogspot.ro] ?

Hi NoobMan,

I have seen the posts and the noise is different, it's not patterned it's just the normal sound but multiplied by 4, hehehe. It's much more appreciable when I rise the drivers to 0,5V that is the recommended voltage.
Other thing that I notice is that when steppers change direction they emit a different noise and much more appreciable, like if they where hitting something physically...

Then maybe values of Rsense, the sense resistors, are different between the drivers under comparison, that would make the relations vref to current different and then 0.5v has different meanings for each. Also you can check the PFD if adjust if its different.

NoobMan,

Thanks again for your help, could you be more specific? I don't know what do you mean by Rsense and PFD... are them in the RUMBA board, drivers or any other component? sorry but I don't know what are them...

All stepper drivers have some resistors, sinking the current from motor coils into ground. Because the coil current passes through them, then a voltage appears across the sense resistors, which the driver takes as input and this is how the drivers knows exactly how much current is in the coil. Hence the name of sense resistors. There are two of them, one for each coil, and have very small values. The a4988 datasheet mentions I_TripMAX = V REF / ( 8 * Rs), and if Rs=0.05ohms, that gives some result, which differs for example if Rs is 0.10 instead. About pfd sry i was thinking of something else, for this driver can only play with Rosc and on the driver boards you can try short across Rosc resistor to ground and see if you get an improvement - on the pololu site schematic that is labeled R4, and also from that schematic image i think check that R2 does not exist (not populated) because only either one of the resistors should be there.

Thanks again, I'll try it out. As far as I understand you think the vibration/stuttering is due to drivers, doen't it?

Yes that is what was going for, but you have to keep in mind, i am rather guessing, its hard both to describe or try to understand issues on forum, sometimes descriptions depend just on how one understands label words e.g. attributes like "vibration" which sort of have different meaning to everybody. The steppers work at high frequency, so basically their input varies widely and regardless how much the inductance tends to level things it can not do it perfectly so the coil fields also vibrate, except that the mechanical part has some inertia and rotor just cant keep up with that kind of movement, but otherwise they are just sort of giant vibrators So i'd say the main cuplrit is the mode of functioning itself. Arguably my impression is that most audible noise comes from decay part, which can be tweaked, except this particular driver and chip are made for the cheap market and do not offer much flexibility for that adjustment. On the other hand shorting out the Rosc to gnd its one functioning mode, its in the datasheet, i think it makes sense in the case it can theoretically help, so i think you should get some improvement with that, at least its worth trying coz there isnt much else to do.

Yes, I totally agree, thanks again for your help, I'll see if I'm able to cross that resistance without damaging the drivers, hehehe.

After lots of trying I have changed my A4988 stepper drivers with DRV 8825 ones and now it goes much better. It still vibrates but not much...
What I have now is that they get quite hot and they sometimes emit strange sounds like they were grating when rotating, strange thing.