New here. Just have to say, I discovered the online textbook here a couple days ago and I'm enjoying it so far.


But right now I'm having a problem with a bipolar stepper motor drive circuit. At speeds of ~3,500pps the motor stalls.

I'm using a PIC microcontroller for controlling step rate and direction, The big Easy Driver is my motor controller (seen here https://www.sparkfun.com/products/10735), and my motor is a LIN Engineering 211-18-01 (seen here http://www.linengineering.com/LinE/contents/stepmotors/211.aspx). I am running it with a 24V 0.6A unregulated power supply.

The PIC program ramps up the speed from 0-6,000pps, full step mode. At low speeds it runs great, however as soon as I get around 3500pps the motor stalls. From the motors datasheet, I can see that with the right driver it can reach speeds of ~15,000pps.

I have seen the term 'electrical resonance' on the forums, and how it occurs in the 2k-5k pps range. I believe this is my problem, however, I cannot find any more info on what exactly it is, how to find it, or how to fix it.

Does anyone know more about electrical resonance it stepper motors? Or, is this even the problem?

Thanks in Advance

 

Have you tried increasing the number of steps per revolution, i.e. changing the microstep size?

 

Page 14 of Lin's Catalog has some info about torsional resonance. Its common and as mlog indicates, one way is to change microstepping params. Going further, any resonance point is a function of the complete mechanical system including the motor and its driver (microstepping, volts, current etc. i.e. anything that influences the torque on the shaft as it steps).

The way I use to fix this in an all-up system is to slew through the resonance point i.e. always be accelerating or decelerating through the resonance point(s). It will take a little trial and error. You want to accelerate enough to sail through resonance but not so fast that the motor can't keep up and start cogging.

Of course, you'll have to accelerate the motor to speed anyway. You cant go from stop to 15000pps as a step function.

Many specs show the motor's inherent resonances w/various drives in the data but I don't see it here. See what happens with slewing. You might want to use the PIC to read the ADC connected to a speed pot and use it to slew around the steps/sec to see where the issues lie. I've done this to find system resonances then plug in the results into the drive profiles.

BTW, while playing around, mind the motor temperature. Some get HOT (enough to burn skin) under continuous use.

Have fun!

 

Have you tried increasing the number of steps per revolution, i.e. changing the microstep size?

I've tried half-stepping. I got it to around 6-7k pps, which is the same speed as 3.5k pps in full step mode.
I found this strange, and I don't know what to make of it.

Page 14 of Lin's Catalog has some info about torsional resonance. Its common and as mlog indicates, one way is to change microstepping params. Going further, any resonance point is a function of the complete mechanical system including the motor and its driver (microstepping, volts, current etc. i.e. anything that influences the torque on the shaft as it steps).

Reading through it, they mention mechanical resonance at 1 rps and 2 rps. I do notice some extra vibrating at around those speeds, but my motor is failing at around 17 rps. I thought I'd be well out of the mechanical resonance at that speed?

Just to make things more difficult, my project is aiming for precise speed and acceleration control. So I'd like to avoid slewing, and changing the acceleration/deceleration rate

 

The electrical resonance has to do with the electrical properties of the motor windings and the drive circuit. Those electrical properties include the rotor inductance and the inter-winding capacitance. You have enough voltage in your supply to overcome the L/R time constant of switching current into the coil and accelerating the rotor. It seems as though your supply lacks the current to provide the oomph to take the next step of accelerating through the resonance. If your supply has poor transient response or starts to fold over because the coils are drawing current near the load limit you might start there.

How about 48 volts at 4 Amps. The motor doesn't actually care about the exact voltage and it knows how much current it wants if it can dissipate the excess power.

 

By slewing, I just mean that you accelerate through the resonances. Your normal accel will likely be sufficient. You just can't level off and run the motor at the resonance point. Noisy and the torque will be lower.

Since both stepping methods fail at roughly the same PPS, I would look next at the power supply and PIC. Is your unregulated PS handling the higher speeds? As the PPS increases, is/are the pulse output(s) from the PIC regularly spaced or is overhead causing some jitter?

Jus' thinkin'.

 

How about 48 volts at 4 Amps. The motor doesn't actually care about the exact voltage and it knows how much current it wants if it can dissipate the excess power.

The motor driver is a current chopper circuit, I don't think giving it more current will do much. I'm also wary of running it above the rated current.

I did try 24V 1.2A though. No noticeable difference

Since both stepping methods fail at roughly the same PPS, I would look next at the power supply and PIC. Is your unregulated PS handling the higher speeds? As the PPS increases, is/are the pulse output(s) from the PIC regularly spaced or is overhead causing some jitter?

Jus' thinkin'.

I'm going to look over the code again. The Step clock is driven by a PWM wave at half duty cycle. So I think the pulses should be fine, but it is easy to mess things up.

 

Depending on the L and the R the chopper drive might actually be the problem. What is the chopper frequency? Have you looked at the current waveform?

 

I've used Chopper frequencies of 15kHz and 33kHz. I was testing two different drive circuits, but both have the same problems.

I don't have a scope so looking at the current means general readings from a multimeter. I noticed that the current draw dropped to around 200-300mA at higher stepper speeds.

 

It may be the inductance of the motor windings which limits the current rise-time and doesn't allow the current to reach the rated value at high pulse rates (which is indicated by the drop in motor current you measure). If you don't have an oscilloscope, it's rather difficult to determine for sure if that is the problem. The way to increase the current rise-time is to use a higher drive voltage and use a current limit circuit to limit the maximum motor current.

 

It totally sounds like torsional resonance to me- experienced that same problem many times.

I assume you are running the motor without a load connected, just watching the rotor?

Your final system load is an important part of the equation, try your tests with the motor installed in the mechanical system you intend to drive, otherwise you may be trying to solve a non-existent problem in the final system.

The inertia and mechanical losses of the load tend to reduce the mechanical "Q" of the motor, making resonance less of a problem. The bare, unloaded motor is your worst case scenario for resonance- don't fixate on it.

 

I assume you are running the motor without a load connected, just watching the rotor?

Yup

Your final system load is an important part of the equation, try your tests with the motor installed in the mechanical system you intend to drive, otherwise you may be trying to solve a non-existent problem in the final system.

The inertia and mechanical losses of the load tend to reduce the mechanical "Q" of the motor, making resonance less of a problem. The bare, unloaded motor is your worst case scenario for resonance- don't fixate on it.

The final mechanical system hasn't been made yet, but I'll try and replicate the requirements and see what happens

 

The reason for raising the supply voltage is to get current into the coils faster. this is important with a chopper drive. The coil will draw whatever current it wants to draw; don't fixate on the motor rating they will take considerable abuse.

 

I noticed some improvement with a small load attached. I'm going to have to do more thorough testing before I can report anything back.

 

Made a simple flywheel and attached it to the shaft to roughly simulate the load requirements.

I hit my speed cap of around 6,000 pps the majority of the time. However there were still a couple of stalls (my flywheel is roughly made/attached). So I'll consider the problem solved, and continue my prototyping.

Thanks for the help everyone.