Lower Voltage But Louder Motors?

I've been using a set of Shinano Kenshi STP-42D3018 stepper motors for the X, Y, and E motors in my RepStrap that are 8.8V 0.64A 11.6Ω bipolar and 5.2V 0.9A 5.8Ω unipolar. I've been using them as unipolar motors because the bipolar resistance of 11.6 Ohms is apparently too high. They've actually worked quite well, especially the XY, with 47 oz.in. of holding torque. But it seemed like a waste to use these bipolar motors as unipolar. With so many different voltage ratings for stepper motors it also seemed like a waste to use 5.2V if there are motors using 3V and under. Plus the extruder motor was getting pretty warm; so if nothing else I wanted a stronger bipolar motor for that.

So I found a new set of motors that seem more ideally suited, especially for the E motor. These KL17H247-168-4B motors are bipolar with 62 oz in. torque. They use 2.8V at 1.68A 1.65Ω, which is a bit less current than 5.2V 0.9A 5.8Ω (or is it?). I installed one as my E motor and it works great, barely gets warm after hours of printing.

So I just installed one of these as the Y motor and tuned the stepper driver to the point where the motor has a reasonable amount of torque. But a couple of things are apparent right away. First, it seems to need me to dial up more current from the stepper driver than the 5.2V unipolar motor did. So that's interesting. And the motor is significantly louder than the previous motor, to the point where I'm thinking of going back to the old motor even though it's got a slightly higher power rating and gets a little warmer.

So I have a couple of questions...

Is loudness the result of a combination of factors that are hard to gauge? Bipolar motors are supposed to have more torque at lower speeds. Does this extra power come with additional vibration as a consequence? Are bipolar motors just generally louder?

Unipolar motors are supposed to be better for high speed. Am I better off having unipolar motors in my XY anyway, where torque is less important than speed?

When it comes to stepper motor current, sure I can run three 5.2V / 5.8A stepper motors and the printer will be a little more quiet, but am I wasting energy?

What values matter when considering compatibility with RepRap electronics? In stepper motor spec sheets they give helpful values like Driver Vref (2.4V), Inductance/Phase (2.4mH), voltage of course, and Current/Phase (0.9A). How do these values relate to the voltage and signals that stepper drivers are sending out?

As a buyer of stepper motors, how can I know ahead of time whether motors might be loud? Are certain voltage ranges louder than others? Are bipolar motors going to have more vibration than unipolar motors?

How can we reduce the loudness of stepper motors? Hanging the printer from the ceiling helps! But short of total isolation, and within the typical RepRap power range, how is motor noise best kept down?

Edited 4 time(s). Last edit at 04/23/2013 05:58AM by Thinkyhead.

---

|
| Lead Developer of Marlin Firmware
| Help support my work at Patreon and Elsewhere.
|

The loudness of the motor mainly comes from the fact that current through coil is turned on and off in frequencies bellow 30 kiloHertz. The sound mainly goes away when microstepping as this frequency increases over audible range.

The 12v or 5.2v or 2.8v that is written on the motor has nothing to do with what voltage you run it through. You actually need to run the motor at several times more that voltage, e.g. most ppls on cnczone will advice x10 times that value. The ratings on the motor have more to do with ohms law and overall power-watts rating of the motor than it has to do with actual application you want it in.

For example since reprap uses 12V psu, there is no chance you can put 1amp of current through a coil with 40 ohms resistance, because simply ohms law says it would take max 0.3A. In practice that shouldnt be peaked value as the driver needs to switch much more before that happens, so its fast enough, because the 0.3A peak takes a lot of time to hit as the waveform is curving down - almost flat. Also generally speaking high coil resistance means thin wire and more windings, so generally higher inductance, thats at least one of the reasons ppls warn against higher inductance with voltages like 12v. Perhaps another is that high inductance has a harder time increase or collapse its field in a current reversal bipolar motor, so perhaps this ppls dont say just "coil resistance" but more generally "higher inductance" which somehow includes both.

To run a motor 5.2v / 5.8A that means a max 0.9 ohms (ohms law) but again thats practically much less coil resistance, probably again an ideal peak would be much xxlog times lower than that, and thats kinda impossible.

To actually check out ratings of a motor, they usually give 2 of three by ohms law and you can figure out whats missing: e.g. they give either current and coil resistance, or voltage and current, etc. This voltage should be much lower, several times lower than what you plan to actually use. This is about the same thing to say that the ohm's law max current is many times much higher than the peak point where stepper driver will turn off or reverse the coils.

Practically take this as general reference, the images at the right side, (ignore the axis labeling and everything else, just pictures).
[en.wikipedia.org]

Thats somewhat a generic waveform or rising/falling edge. The ohms law give the end (top) value, but in practice notice that it takes a huge time to actually get at that. The driver needs to be fast. So the stepper driver needs to work with its own peak value where it interrupts stuff, and that needs to be in the left side, where the waveform is rising rapidly. In that area it rises rapidly, and it allows the driver to work at high frequencies with full current value. The more operating point is to the left, the fastest operation can be.

In practice that peak can be adjusted withing driver's limits, and depending on the driver and its settings it may or may not reach that peak value either (different driver's business). But at least thats what i think of the frame reference on motor picking, and what the parameters should point at.

Edited 4 time(s). Last edit at 04/24/2013 05:31AM by NoobMan.

It's possible I haven't been getting great micro-stepping and this is adding to the vibration. The G3D stepper drivers that I use include a voltage pot and also a separate 30K trimpot, described this way: "turned fully counterclockwise pulls the ROSC pin to ground, turning it clockwise lets you to choose a reference. By pulling the ROSC pin to ground, mixed decay is set to be active 100% of the time, for both rising and falling currents, and prevents missed steps. If this is not an issue, it is recommended that automatically-selected mixed decay be used, because it will produce reduced ripple currents. Refer to the Fixed Off-Time section for details."

I found that the full counterclockwise position was giving too much vibration. I was able to reduce the vibration a little by setting it about 180° clockwise. The E motor still hums a bit, so I'll probably tweak that also. Neither motor is as quiet as my previous set-as-unipolar motors, but they're reasonable. Still, I might revert to the old motor on the Y axis since there doesn't seem to be any added value with this new motor.

Are there common stepper motors out there that have a good torque range and micro-stepping but just aren't very well suited to RepRaps because of their electrical characteristics? Obviously my new motors are within the range of the Vref pot on the driver, but in a slightly higher voltage range. Should I favor the old motors on the basis that they're (a) quieter, (b) require less current, and (c) have plenty of unipolar torque? On the other hand the new motors are a bit prettier than the old ones.

Edited 2 time(s). Last edit at 04/24/2013 08:02PM by Thinkyhead.

I dunno if comparing the unipolar with bipolar can actually give some conclusions, so i'd rather skip that fishy part.

> Are there common stepper motors out there that have a good torque range and micro-stepping but
> just aren't very well suited to RepRaps because of their electrical characteristics?

Yes, wront type are high inductance ones, and generally means thin wiring + extra turns, so also are a kind of a high coil resistance. For example if the coil resistance is 30 ohms. Or voltage written on the label is 12V, neither are good, in spite of the fact we will actually use it at 12v. Both need to be lower, ideally. From the link of mr Nophead here in the tables at the bottom you can see that second motor with 35ohms, 14v, and 44mH is of the wrong kind. As a side note anyway, motors dont have microstepping, but drivers do.

> Obviously my new motors are within the range of the Vref pot on the driver, but in a slightly higher voltage range.

"The ROSC terminal has three settings:
▪ROSC tied to VDD — off-time internally set to 30 μs, decay
mode is automatic Mixed decay except when in full step where
decay mode is set to Slow decay
▪ROSC tied directly to ground — off-time internally set to
30 μs, current decay is set to Mixed decay for both increasing
and decreasing currents for all step modes.
▪ROSC through a resistor to ground — off-time is determined
by the following formula, the decay mode is automatic Mixed
decay for all step modes. tOFF ≈ROSC⁄825, Where tOFFis in μs."

> Should I favor the old motors on the basis that they're (a) quieter, (b) require less current,
> and (c) have plenty of unipolar torque? On the other hand the new motors are a bit prettier than the old ones.

If i were you i would choose to stay with bipolar ones, yes, and specs like 2.6v ~1.6A ~1.6ohms looks ok.