DRV8889-Q1: Lose step

Hi Ivy,

Thank you for posting in this forum.

Ivy Jin said:

1. Could you help share what might be the potential causes of step loss?

In 1/2 Step mode there is a possibility of mechanical resonance and result with step loss. When using 1/2 Step is the step rate reduced accordingly compared to 1/16 Step mode? If step rate is too high the current will not reach the target value because of the motor time constant and BEMF voltage and reduce torque output and potential step loss. What is the VM supply voltage and stepper motor L coil and DCR DC resistance of the coil values?

Ivy Jin said:

2. The following is customer's schematic, could you help review and give some comments?

The schematic looks okay. VREF = 2.5 V or 1.14 A IFS, I full-scale assuming TRQ_DAC = 100 %. I think TRQ_DAC was set < 100 % to get IFS = 600 mA, correct? What is the specified coil current for the stepper?

Ivy Jin said:

3. In testing, the customer found that when the DRV8889-Q1 is configured as 1/2 step, the output phase current varies significantly between VMs, as shown in the following figure. However, when the DRV8889-Q1 is configured as 1/16 step, the output phase current value is basically the same for different VMs.  May I ask that why is that? Theoretically, the output phase current should be independent of VM.

Yes theoretically this hold good as long as VM - VBEMF > IFS x DCR. So in customer's waveform VM = 14 V and 16 V current waveform looks constant amplitude suggesting 8 V and 12 V was not sufficient to achieve the target current because VBEMF was not allowing it. Solution could be to reduce PPS step rate and/or reduce target current and/or increase minimum VM to 14 V and above.   

See below example, 1/2 Step 500 mA IFS 200 PPS. I assume Smart Tune Ripple Control decay mode was used in customer tests. 

With VM = 9 V:

With VM = 12 V:

With VM = 16 V:

With VM = 12 V, IFS = 500 mA, 1/16 Step, 1600 PPS:

If the PPS is too high and VM = 9 V the target current may not be reached and could result with step loss. The driver works as intended. Thank you.

Regards, Murugavel 

Thanks for your reply, It helps a lot!

 What is the VM supply voltage and stepper motor L coil and DCR DC resistance of the coil values?

VM supply voltage range from 9V ~16V, DC resistance of the coil value is 22±2.2Ω(20℃)

What is the specified coil current for the stepper?

Specified coil current is 220mA~400mA, the valve vendor this value is tested via 1/16 step control. and the specified rotation speed is 100pps(pulse per second).

I still have a question, when EXV valve (stepper motor) starts, the envelop curve of phase current fluctuates for the 1st drive of target steps. This phenomenon only happens when the vehicle is powered on, and obvious for the 1st drive control of target speed which starts immediately after vehicle is powered on.

Best Regards!

Hi Jianen,

Thanks for the additional details. At 9 V max possible current would be 9/22 = 409 mA. If the motor rotates the BEMF will reduce this further so it cannot achieve 600 mA for example with 9 V while with 16 V it may be possible because max current can be 16/22 = 730 mA.

jianen yuan said:

What is the specified coil current for the stepper?

Specified coil current is 220mA~400mA, the valve vendor this value is tested via 1/16 step control. and the specified rotation speed is 100pps(pulse per second).

Configuring for around 220 mA can help maintain the target current at lower voltages.

jianen yuan said:

I still have a question, when EXV valve (stepper motor) starts, the envelop curve of phase current fluctuates for the 1st drive of target steps. This phenomenon only happens when the vehicle is powered on, and obvious for the 1st drive control of target speed which starts immediately after vehicle is powered on.

I mentioned previously the current through the winding would be (VM - VBEMF) / DCR Ω. So when VM - VBEMF < IFS x DCR the current cannot reach the target IFS. The VBEMF of a stepper motor is a sine wave. So VM - VBEMF will be a sine wave, so the current is superimposed with this sine pattern from the VBEMF of the motor. VM is stable as shown by the battery voltage trace.  

Regards, Murugavel 

Thanks for your reply. I think if the fluctuation of amplitude of phase current can't be avoid, we should make sure the minimum of  phase current should be larger than the specified value.

jianen yuan said:

when EXV valve (stepper motor) starts, the envelop curve of phase current fluctuates for the 1st drive of target steps. This phenomenon only happens when the vehicle is powered on, and obvious for the 1st drive control of target speed which starts immediately after vehicle is powered on.

The graph above shows using 1/16 STEP mode, we configured the TRQ_DAC as 18.75% which regulates the current as 220mA. Basically, we were going to  raise the TRQ_DAC in the hope to get a larger regulated current, but from your reply, this will not have a good effect. It seems that the best way is to reduce the rotation speed. We will test in bench and on the vehicle to have a check of the result.

Best regards!

Hi Jianen,

jianen yuan said:

I still have a question, when EXV valve (stepper motor) starts, the envelop curve of phase current fluctuates for the 1st drive of target steps. This phenomenon only happens when the vehicle is powered on, and obvious for the 1st drive control of target speed which starts immediately after vehicle is powered on.

Kindly explain what you meant by "This phenomenon only happens when the vehicle is powered on, and obvious for the 1st drive control of target speed which starts immediately after vehicle is powered on.". What was the enabling sequence used prior to driving the stepper?

jianen yuan said:

What is the specified coil current for the stepper?

Specified coil current is 220mA~400mA, the valve vendor this value is tested via 1/16 step control. and the specified rotation speed is 100pps(pulse per second).

You mentioned specified speed as 100 pps. I assume this would be with full-step and with 1/16 it would be 1600 pps, is this correct? I have a generic EEV with 20 Ω coil in our lab. I tested this valve with a DRV8889-Q1EVM. IFS set at 400 mA peak (283 mA RMS), 1/16 Steps, 1600 pps.

With voltage stepped from 9 V to 17 V, yellow is VM and blue is current:

jianen yuan said:

I think if the fluctuation of amplitude of phase current can't be avoid, we should make sure the minimum of  phase current should be larger than the specified value.

Yes this will help maintain minimum torque output. Thank you.

Regards, Murugavel 

Hi, Murugavel

Murugavel4637 said:

What was the enabling sequence used prior to driving the stepper?

Our controller will drive 3 water pumps (BLDC, max total current is about 25A), 2 water valves (BDC, max total current 1A), 2 solenoids, 7 EXVs (stepper motor, for 2 small EXVs specified current is 220mA~400mA, for other EXVs specified current is about 400mA). The  drive sequence is not clear. But we have verified the fluctuation phenomenon has no relation with parts other then EXVs by removing parts controlled.

The phase current amplitude fluctuated EXV is named as BEXV, We have checked that if we remove some EXVs, fluctuation of current amplitude of BEXV did not occur. We have no clear idea of the reason. But according to the words mentioned previously "theoretically this hold good as long as VM - VBEMF > IFS x DCR", on occasion of VBEMF reach the max value when motor rotates, regulation of phase current may not hold good and will cause fluctuation. I assume this may be the reason but occurrence condition is not easy to identify.

Murugavel4637 said:

I assume this would be with full-step and with 1/16 it would be 1600 pps, is this correct?

Sorry, I didn't make it clear, rotation speed is100pps with 1/2 step control.

Hi Jianen,

Thanks for further details.

jianen yuan said:

rotation speed is100pps with 1/2 step control.

Understood.

jianen yuan said:

The phase current amplitude fluctuated EXV is named as BEXV, We have checked that if we remove some EXVs, fluctuation of current amplitude of BEXV did not occur. We have no clear idea of the reason. But according to the words mentioned previously "theoretically this hold good as long as VM - VBEMF > IFS x DCR", on occasion of VBEMF reach the max value when motor rotates, regulation of phase current may not hold good and will cause fluctuation. I assume this may be the reason but occurrence condition is not easy to identify.

This would be the highly likely reason. Increasing C71 to > 100 uF placed close to VM of each driver may help mitigate this issue. Thank you.

Regards, Murugavel