DRV8711 - Possible mid band Resonance?

Hi Nick,

I am glad to hear you have been happy with the DRV8711 in the past.

Can you provide the motor specs and the DRV8711 settings?
Can you provide a scope capture of the odd current?
Some motor manufacturers will provide the resonant frequency of the motor. Have you asked your motor manufacturer for this?

In general, avoiding the resonant frequency by accelerating/decelerating through the range quickly is one method. Using microstepping or adding inertia as you state can also help.

Changing decay modes does not have an effect on the start of the next cycle.

Thanks for the quick reply Rick,

I will get this information together and include it in my next post.

With respect to the decay mode question: What I really wanted to know is ....is there a decay mode which restarted the decay cycle to the charging part (beginning) on receiving a new step pulse in?! I thought that I had read this somewhere?

Nick

Rick,

DVR8711 setting are as follows:

• 1/16 microstepping
• ISGAIN = 5
• DTIME = 400ns
• Torque = 39 (give me about 1.5amps with a 0.05ohm shunt resistor and ISGAIN of 5)
• SMPLTH = 100uS
• Toff = 30uS
• Internal indexer
• Tblank = 2.24uS
• ABT = enabled
• TDecay = 8uS
• Mixed decay
• TDRIVEN = 1uS
• TDRIVEP = 1uS
• IDRIVEN = 100mA
• IDRIVEP = 50mA (the FET gate charge is 18nC)

Please see attached PDF for motor specs

MAE Stepper Motor.pdf

See scope trace below for odd current wave form. As you can see the wave form isn't a constant shape. It seems to flicker between two forms before the motor desinks.

 

Again it would be great to know if any of the decay modes restart the chopping cycle on receiving a new step pulse?

Many Thanks

Nick

Hi Nick,

Thank you for the information. Can you clarify which of the three models is being used in the scope captures? If the 0100 or 0300, how are the A/A' leads connected (series or parallel)? What is the VM voltage?

Automixed decay will restart the chopping cycle when a step pulse is received. Please refer to the first paragraph after Figure 10 in the datasheet for more information.

It appears you may be outrunning the ability of the system to inject or remove current into the winding. This has been observed at high microstepping rates with higher inductance motors and at lower voltage.

This is based on the following assumptions:
At 1/16 microstepping, it appears a step pulse is issued every 18us.
So the DRV8711 indexer table may be lagging in the current regulation.

Please try the following:
1) Change to 1/8 or 1/4 usteps and reduce the number of steps per second by half or 1/4, respectively? Lowering the steps/second allows more time for the current level to be reached. Does performance get better?
2) If the voltage is low (12V to 15V), can you raise the voltage to 24 to 36V? Raising the voltage improves the ability to inject or remove current from the winding.

Hi Rick,

Sorry, The motor in this scope trace is the 0300 wired in parallel. The VM is 31V there but I have experimented up to 48V already and it seems to allow me to pass though the trouble point at that speed but the problem becomes present again at a higher speed (I'm guessing the next harmonic of resonance for that motor?? or it just shifts the resonance to a higher frequency).

Thanks for pointing me to the location in the data sheet where it states that auto mixed decay starts a new chopping cycle when it receives a new step input signal. Thought I was going mad!

What you are telling me seems to be matching up exactly with what I have found out through experimentation already. This effect is reduced greatly, but not eliminated, by running in half stepping (or a lower microstep setting as you suggest). Increasing the voltage to a point allows us to pass this band of resonance.

Ideally we would want this motor to run up to 2000 rpm at 1/16 microstepping and 36v like the other motors we use as specified by our customer.

Thanks

Nick

Hi Nick,

Do the other motors have lower inductance than this one?

I am not sure 2000 RPM at 1/16 microstepping is feasible. The torque curve for the 0300 motor is met by raising the voltage to 60V. Lowering the voltage will lower the torque curve and decrease the maximum number of steps the motor can run.

I suggest you check with your motor manufacturer prior to proceeding with the suggestions below. In general, there should be not problem as the RMS current does not exceed the rated current. The motor manufacturer may be aware of a reason not to raise the current.

In my opinion your best option for 2000 RPM is running full step and raising the torque setting to 140% of rated current. This is actually 100% of the rated full step current as the full step setting is 71% of the torque setting. Using this method will allow the current to reach the desired level and remain there for the longest period of time. Each step will be 150us.

Once you have this working, you can try half step and maybe quarter step. If trying half and quarter step, either lower the torque setting back to 100% or stop at a full step location. Stopping at a full step location allows the maximum rated current through each winding.