Stepper growls like a bear

[Danny Dd]

I am not able to achieve much speed before the steppers lock up or growl on my 

Cartesian clay printer.

Arthur suggested this is caused by resonance. But after reading the thread about Simplify 3d which I was using I thought maybe bad G-code was the problem. 

But it seems to happen almost as frequently with Repetier slicer. I don't think its resonance because I can travel at higher rates of speed and some prints will run fast with no problem but other prints will get halfway through and then start to lock up and growl like a bear. 

Here's is some of my settings. 

default_feed_rate                            750         # Default rate ( mm/minute ) for G1/G2/G3 moves

default_seek_rate                            750           # Default rate ( mm/minute ) for G0 moves

mm_per_arc_segment                           .5              # Arcs are cut into segments ( lines ), this is the length 

                                                              # these segments.  Smaller values mean more resolution,

                                                              # higher values mean faster computation

mm_per_line_segment                          5                # Lines can be cut into segments ( not usefull with cartesian

                                                              # coordinates robots ).

# Arm solution configuration : Cartesian robot. Translates mm positions into stepper positions

alpha_steps_per_mm                           2200           # Steps per mm for alpha stepper

beta_steps_per_mm                            2200             # Steps per mm for beta stepper

gamma_steps_per_mm                           2200           # Steps per mm for gamma stepper

# Planner module configuration : Look-ahead and acceleration configuration

planner_queue_size                           32               # DO NOT CHANGE THIS UNLESS YOU KNOW EXACTLY WHAT YOU ARE DOING

acceleration                                 50         # Acceleration in mm/second/second. 

#z_acceleration                              500              # Acceleration for Z only moves in mm/s^2, 0 uses acceleration which is the default. DO NOT SET ON A DELTA

acceleration_ticks_per_second                1000            # Number of times per second the speed is updated  1000/50

junction_deviation                           0.005             # Similar to the old "max_jerk", in millimeters,.

                                                              # see https://github.com/grbl/grbl/blob/master/planner.c

                                                              # and https://github.com/grbl/grbl/wiki/Configuring-Grbl-v0.8

                                                              # Lower values mean being more careful, higher values means being

                                                              # faster and have more jerk

#z_junction_deviation                        0.0              # for Z only moves, -1 uses junction_deviation, zero disables junction_deviation on z moves DO NOT SET ON A DELTA

#minimum_planner_speed                       0.0              # sets the minimum planner speed in mm/sec

# Stepper module configuration

microseconds_per_step_pulse                  1           # Duration of step pulses to stepper drivers, in microseconds .

base_stepping_frequency                     100000         # Base frequency for stepping, higher gives smoother movement

# Cartesian axis speed limits

x_axis_max_speed                             750            # mm/min

y_axis_max_speed                             750             # mm/min

z_axis_max_speed                             250              # mm/min

   

The only way I can get any speed at all is by lowering the acceleration way down and  lowering the jerk number substantially. I sure could use some help trying to work through these problems.

[Mark Rehorst]

What size motors are you using?

It seems to me that running with such low acceleration and junction deviation is forcing the motors to run very slowly for a long time before they ever get to a moderate speed.  Your maximum speeds are only 12.5 mm/sec in X and Y.  Steppers generate a LOT of vibration when they are run slowly, especially if they are larger size motors.

[Triffid Hunter]

Mid-band resonance is vastly reduced or totally eliminated by microstepping. Smoothie's onboard drivers run at 16x microstepping by default, so resonance should not be a problem.

Far more likely is that you've chosen motors with the wrong specs - ideally, you want motors with the lowest winding resistance possible, and a current rating of 1.5 to 2A.


A common error when choosing stepper motors is to pick motors with a high winding resistance, because they claim a "voltage rating" of 12v.

This "voltage rating" is a red herring - it's simply the rated current multiplied by the winding resistance, and dictates the voltage required to push the rated current when the motors are stationary. When the motor is actually spinning, a higher voltage is required due to the generator effect and winding inductance.

Such motors are suitable for dumb voltage-mode stepper drivers eg repurposed H-bridges in fixed-speed applications like paper+ink printers, but interact very poorly with smart current-mode microstepping drivers like the ones on smoothie.

I have excellent success with RATTM 17HS8401 motors (1.8A, 1.8Ω, 52 N.cm - http://www.aliexpress.com/store/product/10pcs-NEMA17-78-Oz-in-CNC-stepper-motor-stepping-motor-1-8A/704350_423237714.html ) - see http://youtu.be/xQNJgXV6KAQ for a demonstration of my printer using these motors with smoothie.

There are numerous other stepper motors with suitable specs, eg kysan 1124090, Automation Technologies' KL17H248-15-4A, etc.

If you have purchased "12v" motors with high winding resistance and are feeding smoothie only 12v, that is most likely your problem.

To fix it, you must either buy suitable motors, increase the supply voltage to smoothie (24v max), or both.


Another possible cause is setting motor current to a low value, eg less than 1A.

Smoothie's drivers have limited resolution for reading motor current, and struggle to accurately meter such low currents especially when microstepping.

It's common for motors with high winding resistance to have a low current rating; I've seen as little as 0.4A.

Again, replacing the motors with suitable ones will give a vast improvement in performance.


A third possibility is that your motor wires are damaged or poorly connected. If one wire is broken, it's common for the stepper to simply vibrate (with consequent noise) or turn extremely weakly in a random direction.

[adam paul]

There's also the chance that the pair's are wrong. I have found some drivers to be +-+- and some +--+.

[Danny Dd]

Mark 

when I increase the acceleration and junction deviation it only gets worse 
Triffid
 I've included the specifications for the NEMA23, any comments

 Adam 

are you suggesting changing the wire order going into the stepper motor, I've never heard of this before.

https://youtu.be/ogbDH6MfHos

[Arthur Wolf]

That does sound *a lot* like a resonnance problem ( which most definitely can happen even with 1/16 microstepping ).

If you ask google, there is *a lot* of information on the web about how to mitigate those problems.

What are you using for stepper motor drivers ? What voltage ? are you sure you have your current setting set right ?

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-sup...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

[Danny Dd]

Arthur I'm driving directly off of the smoothieboard

alpha_current                                2.0              # X stepper motor current

 power supply 24 volts. Look at the my last post for the motor information chart

[Arthur Wolf]

Well you are running a 2.8A motor at 2A so you are not getting all you can from it.

For a machine like this I'd strongly recommend going to external stepper motor drivers and a 36v or 48v PSU ...

The less power you have, the more likely problems like you have here are.

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-sup...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[David Crocker]

Your stepper motors are a very poor match for the drivers on all the all-in-one electronics boards I know of. It's possible to run Nema 23 motors from these boards, but only if you use 24V or higher supply voltage, you choose motors with a maximum current in the 1.5 to 1.8A range, and you don't need high speeds. My guess is that the A4982 drivers on the Smoothieboard are going into thermal limiting, because they can't handle 2A continuous current. You may get better results if you turn them down to about 1,4A - but you will only be getting half the rated torque from them (which may be enough for a 3D printer). To get good performance from those particular motors, I agree with Arthur - you need external drivers. It may be that 24V supply voltage is still sufficient, it all depends on what speeds you are looking for.

[wolfmanjm]

Sorry but this is BS, Smoothie can handle 2a continuous easily. especially if you run a fan over the bottom of the board.

[Arthur Wolf]

At 2A, unless you are in a place with very good air circulation, or a cold climate, it's *likely* you'll need to aim at least a small fan at the board. But yes the board can definitely do 2A.

--

You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-sup...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Danny Dd]

David 

I'm using a 24 volt power supply with aggressive cooling so I don't believe power is a problem. I have successful ceramic prints that go over an hour with no overheating problems. I've also tried external drivers and experience the same problem see photo attached

[David Crocker]

Personally, I would never run an A4982 at anything like 2A. The datasheet gives the absolute maximum output current as 2A. It gives the source and sink driver resistance as 320mohm typical and 430mohm max @ 1.5A and 25C ambient. So 1.5A is the highest current at which any guarantees are made. The resistance at 2A might be substantially higher. In the absence of more information, it is impossible to tell how much cooling the chip will need if it is run at 2A. Some individual samples of A4982 chips may be capable of continuous operation at 2A given good cooling, but others may not be.

However, as Danny says the problem also occurs when using external drivers, I am inclined to agree that it's probably a resonance problem.

To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[wolfmanjm]

That is because you probably only ever use the pololu style carrier which has lousy heat sinking capability.

The entire smoothieboard is a huge heatsink and they run all day at 2amps with a small fan on the back of the board. No one has ever reported burning one out or even having thermal shutdown at 2amps.

[Triffid Hunter]

Yes, the current limit is dictated entirely by heat.

There's an unavoidable thermal resistance between the silicon and the PCB which sets a hard upper limit depending on ambient temperature, but the more efficiently we can suck heat away with the PCB, the closer we can approach that hard limit - asymptotically of course.

As wolfmanjm says, the entire smoothieboard is a colossal heatsink which greatly exceeds the recommended heatsink requirements in the chip's datasheet , and as a result the limitations and examples listed therein aren't entirely valid.

Real world testing is always more useful than guesstimates from numbers in the datasheet, and indeed we have wide verification that with smoothieboard's epic thermal design (which I'm super proud of) we really can push these chips all the way to their absolute maximum.

The stepper drivers have built-in thermal overload protection, and to my knowledge we have zero reports of it ever kicking in at 2A or less.

The trouble with the daughter board drivers popularised by other electronics is that the heatsink area and access to free moving air is rather less than the requirements in the datasheet - not only does this mean that the drivers overheat at a measly 1A, but the insufficient thermal mass also renders the thermal overload protection less effective than it should be.

Furthermore, I'm firmly convinced that a significant proportion of the daughterboard driver death rate is caused by having them on headers - small spring contacts coupled with moderate currents and lots of vibration is a perfect recipe for intermittent poor contact, noise, inductive kicks, ohmic hot spotting and other electrical effects well known to rapidly destroy chips.

Ironically, the very nature of making the drivers easily replaceable causes them to be replaced often.

On smoothieboard, the drivers simply do not overheat, and do not randomly die at anywhere near the rate reported by the wider reprap community for daughterboard drivers - Our minuscule driver failure rate (excluding user errors) seems to be comparable to a reasonable expectation of unavoidable manufacturing defects.

Of course, silicon degrades rapidly near it's maximum temperature and heat is proportional to square of current, so if the thermal overload protection doesn't kick in at 2A the drivers should last almost forever at 1.5A.

The only downside of smoothie's thermal design is that it's difficult to impossible to rework with a normal soldering iron - hot air is critical.

[David Crocker]

Yes I am fully aware that the Pololu-type driver carriers have totally inadequate cooling. I don't use those drivers. The fact remains that the datasheet makes no guarantees about the Rds(on) of the drivers at currents above 1.5A, so it's pot luck whether you get chips that are OK at 2A. The fact that the manufacturer quotes Rds(on) at 1.5A instead of 2A is a sure indication that not all chips will maintain low Rds(on) at 2A, even if most of them do. It's the same with power mosfet datasheets: the headline figure for maximum current may be a very high figure like 195A, but to find out what current that you can safely switch in all production instances of a design using that mosfet, you need to look at the current(s) for which a maximum Rds(on) is quoted.

So it's not all about cooling: it's about cooling AND having chips that have low enough power dissipation @ 2A that it's actually possible to cool them adequately. Maybe it's because I work in the safety-critical sector and have higher standards than most engineers, but I never rely on semiconductor characteristics that are not guaranteed by the manufacturer.

[Paul Wanamaker]

I have a large delta 3d printer that uses Nema 23's.  

I run the smoothieboard at a higher voltage, and so had to go to greater lengths to cool it.  I designed this SmoothieBoard Enclosure with Fan which I've just published on Youmagine.com, and also added heat sinks.

One thing I did run into was resonance.  I did a great deal of testing, including attaching various sized flywheels.  I ended up getting some digital stepper drives.  Through a very arduous process I was able to tune these to handle the resonance, it quieted down quite a bit and I was able to achieve much higher speeds (needed when the effector is way out at the edge).

Subsequently I used larger pulleys (I have .9 degree motors, didn't need the resolution overkill) and that allowed even faster speeds, and still have 177 steps per mm.

So I do recommend the digital stepper drives.  

Or you could reduce the max rpm by using a larger pitch lead screw.

I can not recommend the company I got the drivers through (who will remain nameless):

- they state clearly they do not support the tuning software

- the cable they provided only works on a serial port!  I borrowed an old XP laptop with a serial port to do it.

- the software also was very out of date and didn't have the model number.

- I think their drivers were a clone of the big Chinese brand - who's software I finally had to use.

You may want to check out the Leadshine DM Series - DSP Based Digital Stepper Drives

[Danny Dd]

Paul that some good advice. I decided to increase my pitch on the lead screw and at least it keeps me away from the problem rpm. So if I understand correctly going to .9 degree motor also helps. I'm trying to stay away from Digital stepper driver because of the added expense and complexity. 

[Paul Wanamaker]

I hope that helps!

I really don't know if .9 degree motors would help -if the resonance is frequency dependent they may hit the first resonance point at 1/2 the speed - just a theory. 

[Mark Rehorst]

The ball screw driven Y axis in my printer is powered by a 425 oz-in NEMA-23 stepper driven by a DSP driver and a 32V power supply.  The X axis is belt driven, with a lower torque NEMA-23 motor, DSP driver, and 32V power supply.  The X axis is silent, the Y axis is noisy.

I tried a vibration damping motor mount (Astrosyn), a (3D printed) harmonic damper mounted on the motor shaft, every possible combo of microstepping, acceleration, speed, and junction deviation, etc., all to no avail.  The Y axis remains noisy.  I have come to the conclusion that all that metal on metal contact in a screw drive is very efficient at transmitting vibrations from the motor to the rest of the machine.

I've learned to live with it, but my wife hasn't, so most of my printing is done at night.