Stepper Motor Stalling When Driving Reciprocating Load

Welcome to CR4.

When you say "at certain points in the rev range" - do you mean at certain step rates? If so, I'd suspect a resonance. This is quite common, and can usually be avoided by one of a few techniques.

What kind of drive are you using?

Hi,

Let me be more specific, due to problems in using the motor in the device, we've been testing it on the dyno.

Firstly we took a torque/speed plot of the motor on it's own, it performed as expected.

Then we attched the pump assy attached to the back of the motor as per it's application in the device and the front output shaft is attached to the dyno. We accelerated the motor to a range of operating speeds between 1500 and 4500 Step /sec. We then started to increase the load on the motor using the dyno and measured the stall torque.

The problem is that there is two bands in the rev range that the motor is exibiting a low stall torque when the pump is attached roughly between 1500 -2500 Step/sec and 3500 - 4000 step /sec.

Looking at the motor torque curve the motor seems to have ample torque at these speeds but with the pump attached the motor is stalling with a low additional load.

The load the pump presents is reciprocating between 0.4NM and -0.3 NM.

Any clearer?

Sounds even more like a resonance. Are you using full step/half step/microstep or what?

We're running in full step mode at the moment.

If you can, go straight to half-step. Microstep would be even better.

GA. This does not sound li8ke an appropriate application for a stepper.

I also agree that is is not the right fit for a stepper, but I don't have enough experience in this area, so I didn't say anything.

GA

I agree that half stepping will improve resonance problems but there's still a lot of other parameters that can be causing this resonance problem. Do you have the driver's running current set to the appropriate current for the motor? Does the supply provide a high enough voltage (into the driver) so that the peak current can be achieved in the time interval of each step despite the back EMF? Are you certain that the mechanical torque load at the motor's shaft is always less than 40 Ncm?

From #2:

"Firstly we took a torque/speed plot of the motor on it's own, it performed as expected."

Doesn't sound like there's much wrong with the drive (voltage or current).

I agree that higher torque than anticipated could be a problem, but even a single stack 34 frame single-stack "old tech" motor^[1] should be good for about 80 Ncm.

If half-stepping (or microstepping, if it's an easy option) doesn't fix it, it's time to start playing with the mechanics. First try a compliant coupling (i.e. a lump of rubber) somewhere in the linkage - may need to experiment a bit. If that doesn't help, the next step would be either to muck about with damping, or to use a gearbox. First a simple friction damper (assuming there's plenty of spare torque) - then getting more extreme/expensive whatever and fitting a rotary damper (either electromagnetic or hydraulic), or going for a gearbox.

However, I don't think it will get that far before a fix is found.

[1] A more modern "enhanced performance" 34 size 3-stack motor will deliver up to 300 Ncm at the step rates in question.

Well if the motor performed as specified by the manufacturer with a dynamo-meter load and it still stalls on their rig, then the only conclusion then is that they must have a bigger mechanical load than anticipated.

Are you sure that the drive you are using with the motor is adequate to support the required power at the high step rate. Not all drives are created equal.

When using a stepper motor it becomes increasingly difficult to get full power into the windings as the step rate increases. If the drive does not supply a high enough voltage then the motor will stall.

Resonace try a flexible coupling and/or a fly wheel

Good point and a GA.

The reasons for a stepper motor are manifold, driving the pump is just one of it's jobs, it also drives a geartrain at a specific speed, regardless of supply voltage (110/240, 50hz 60hz), reducing our variant count to 1.

It replaces an AC motor and represents a significant cost saving.

The pump load has been tested using a inline torque transducer an was driven by a smaller frame stepper motor for the test, this showed a ocsilating load of 0.4 to -0.4 Nm at full pressure.

Have you tried half- or micro-stepping it yet?

Just my 2 cents having worked in both the printer industry and industrial automation industry, stepper motors are at the bottom of the list with all specifications related to torque (T/$, T/watt, T/size, T/velocity) so I would compare torque curves on the motors to insure you are comparing apples to apples.

I've been involved in a number of programs where we have redesigned the product to convert from stepper motors to DC servo or synchronous AC to reduce cost while improving performance. The development effort is significant but the BOM cost was always lower.

Different motor designs came about because no one motor design does everything very well. I would be very surprised that a stepping motor would be less expensive than an AC motor. Having only one motor doing multiple things at different times will require the addition of multiple clutch mechanism so that each mechanical load can be added at a time. This will certainly be much more complicated than having multiple motors independently turned ON/OFF electrically. You can now also select the ideal type of motor for each load so that your open loop position control will be operated by the stepping motor, your variable speed motor will be operated by a DC motor and your heavy lifter can be a three phase AC induction motor. If you don't have all that heavy a lift then a single phase AC induction motor can be located that by just providing a user select switch can cover both 110/240 VAC operation.

I'm not actually conducting the test, but it's in the plan to test using half steps soon. I'll let you know the result.