TI中文支持网你好，我现在要控制电机转速，要求稳定性很高，DRV系列有没有合适的，电流-速度双闭环的芯片，电压5-15V，电流3A，谢谢。
user467873：
回复 Annie Liu:
电机是带开关霍尔的无刷电机。
user467873：
回复 Annie Liu:
之前用DRV10970控制这个电机，先速度闭环，然后进入锁相环闭环，电机转速波动还是有点大。之前用罗姆的BA6840控制很好，想用DRV10970替换，但是DRV10970一直达不到BA6840的状态。
user467873：
回复 user467873:
用DRV10970控制这个电机，先速度闭环，然后进入锁相环闭环，电机转速波动还是有点大。之前用罗姆的BA6840控制很好，想用DRV10970替换，但是DRV10970一直达不到BA6840的状态。是不是因为BA6840是电流控制的，DRV10970是电压控制的区别啊？
Annie Liu：
回复 user467873:
很抱歉，回复晚了。请查看产品工程师的回复。
About the problem:
Before talking about the two paths. I noticed that 3A current is a specification, which neither the DRV10970 nor the DRV10974 can support. I also can't get a complete datasheet of the BA6840 online so I couldn't determine if it also had that capability.
Another thing I did notice about the BA6840, assuming the datasheet I found is correct, is that it has the capability of a torque loop. Given that motor speed fluctuation is a primary concern, the torque loop is a key feature that is not present in either of the DRV10x devices.
The DRV10x might say "closed loop" but I don't want to confuse you that this means there is a speed loop, torque loop, or speed regulation. The DRV10x has the potential to have very low speed fluctuation (~1% speed fluctuation) if the load is very constant, and the motor is of high quality.
To combat this, some customers will introduce a cheap MCU that will be able to provide a reference speed and monitor the speed through FG to reduce the error between them (to make a speed loop). Usually this results in <1% speed fluctuation but has an increased board size, cost, and coding support. We have some plans to integrate these features in the next gen, but that is a ways off and I'd be more interested solving your problem now.
With these specs in mind, the BA6840 might be the better part for the application, I encourage you to take a look at the specifications and quantify them more rigorously.
DRV10974 Replacement:
It looks like you have sensored motor and you might notice the DRV10974 is sensorless (note that the DRV10970 is our only sensored device right now). Sensorless solutions usually have slower start ups compared to sensored and are more prone to problems (e.g. some tuning would be needed). I encourage you to take a look at the recommended application range and see if your motor falls neatly into the bounds of the algorithm. If it does not, we will not have high confidence that the device will be able to spin the motor; but it doesn't guarantee it is impossible.
user467873：
回复 Annie Liu:
我初步怀疑达不到要求的原因是DRV10970没有电流力矩环引起的，但是不能确定。
user467873：
回复 user467873:
非常感谢。
user467873：
回复 user467873:
DRV10970实现了1%的速度波动了，但是产品要求要远远小于1%的波动。
