• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 30 subscribers
  • Views 1867 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV10970: DRV10970 and DRV8308/DRV8307 Comparison

Jim Gavin
Intellectual 920 points
Part Number: DRV10970
Other Parts Discussed in Thread: DRV8308, DRV8307,

I am in the process of selecting a motor driver/controller and I have narrowed it down to the DRV10970, DRV8308 or DRV8307. I need to drive Faulhaber 2-pole BLDC motors from 1000-8000 RPM. The motors have digital Hall sensor outputs, so having Hall sensor rotor position feedback functionality is a requirement. In addition, the driver/controller must use Trapezoidal 3-Hall Commutation to maximize torque output.

After reviewing the datasheets, I can see clear differences between the DRV8307 and DRV8308 with the configurability. And it appears that the DRV8307 will automatically switch to single-Hall Commutation when the speed is "relatively" constant, so that seems to rule out the DRV8307. It also appears that the DRV8308 can be forced to use 3-Hall Commutation by setting the BASIC bit, is this correct?

As for the DRV8308 vs. the DRV10970, the DRV8308 is about 2 years older than the DRV10970 and they appear to have almost the same functionality, just implemented differently. Is this correct? Are there any differences in the output drive or otherwise that would affect driving a motor under variable torque load (specifically, an increasing torque load for tensioning in my application).

Thank you,

Jim

  • 0
    TI__Mastermind 18900 points

    Hey Jim,

    I'm an applications engineer for the DRV10x products and I can say that the DRV8307/8 are gate drivers (in which they drive external FETs) where the DRV10970 is a driver (has the FETs integrated into the chip). I don't know about the DRV8307, but the DRV8308 and the DRV10970 can drive with 180 degree commutation to maximize torque and has the 3 Hall sensor setup like you require. Also, they both have the same protection.

    Essentially, this comes down to you wanting integrated solution versus non-integrated solution (for FETs). I'd look into your requirements: board size, a voltage range, RDS on, and how much current you need for the application. Looking at the motor datasheet should help you answer these if you haven't thought about them.

    I'll let a DRV8x expert speak about the differences between the DRV8307 and DRV8308.

    Hope this helps,

    -Cole

  • 0 in reply to Cole Macias
    Intellectual 920 points
    Thanks for the quick response Cole.

    Looking at the DRV8308 eval board, the RDSon is about 25 mOhm (HS + LS) with the CSD88537 FETs, but the RDSon isn't the real culprit in my application since I have a relatively low motor current (~125mA). The issue is switching losses, so I am more concerned about the switching time of the DRV10970, which is mostly dictated by the gate drive strength and the gate charge of the internal FETs, but that information is not available on the datasheet. Can you provide any additional information about the switching time, gate drive strength and/or gate charge of the DRV10970?

    Thanks,
    Jim
  • 0 in reply to Jim Gavin
    TI__Guru** 114905 points
    Hi Jim,

    The DRV8308 can be forced to use 3-Hall Commutation by setting the BASIC bit.
    As Cole mentioned, the DRV8308 is a gate driver allowing you to optimize the FETs for your needs. Also the voltage range is higher (32V recommended maximum vs 18V for the DRV10970).
  • 0 in reply to Cole Macias
    Intellectual 920 points
    Cole,

    I forgot to ask about what you said regarding 180° commutation and torque. You said that the DRV8308 and DRV10970 both can drive using 180° commutation to maximize torque, but the DRV10970 datasheet states "Trapezoidal mode provides higher driving torque" on page 16. I'm assuming that when you say 180° commutation, you mean sinusoidal commutation, correct? I know that sinusoidal commutation provides a smoother and quieter operation, but not much more than that. Also, I have a strict torque limit (thus current limit) and the DRV10870 datasheet indicates that the current limit is 1.5x the current limit in trapezoidal mode (pg. 14).

    Thanks,
    Jim
  • 0 in reply to Rick Duncan
    Intellectual 920 points

    Cole and Rick,

    I am really trying to determine which driver to use for my application, so could you help me understand why I might pick the DRV10970 over the DRV8308 and vice versa?

    My application has the following parameters.

    • Relatively low current (worst-case motor configuration is 250mA RMS and up to 900mA peak)
    • Motors are 2-pole Faulhaber coreless BLDCs with low inductance (85uH)
    • Motors have digital Hall sensors for rotor position feedback (spaced 120°)
    • Motor speed is controlled to 1000 to 8000 RPM 
    • Speed is controlled to within ±10%
    • Speed is maintained with a variable torque load at low RPM (specifically, an increasing torque load for tensioning at 1000 RPM)
    • Cost of the driver is not a concern

    One thing that I haven't quite figured out is the drive angle and whether or not it affects torque.

    Thank you,

    Jim

  • 0 in reply to Jim Gavin
    TI__Mastermind 18900 points

    Hey Jim,

    Let me address your questions:

    1. Can you provide any additional information about the switching time, gate drive strength and/or gate charge of the DRV10970?
      1. The DRV10970 datasheet shows that the PWM output frequency on the phase is always 25 kHz.
    2. You said that the DRV8308 and DRV10970 both can drive using 180° commutation to maximize torque, but the DRV10970 datasheet states "Trapezoidal mode provides higher driving torque" on page 16. I'm assuming that when you say 180° commutation, you mean sinusoidal commutation, correct? I know that sinusoidal commutation provides a smoother and quieter operation, but not much more than that. Also, I have a strict torque limit (thus current limit) and the DRV10870 datasheet indicates that the current limit is 1.5x the current limit in trapezoidal mode (pg. 14).
      1. You're correct, I mispoke earlier when I said, "maximize torque". I really meant the "smooth operation" you're mentioning. The reason why 180 degree/sinusoidal is a features is that it torque is constant during every angle of commutation. While you're correct trapezoidal/120 will offer more torque, sometimes the motion can be described as "uneven"
      2. To summarize, if you're concerned about meeting an amount of torque, 120 will offer the most but 180 could work too. If you concerned about quality and smoothness of torque, then 180 is much better but 120 might be good enough for your application. Unfortunately, both devices can do 120 and 180 so it doesn't help you wil differentiation.
    3. why I might pick the DRV10970 over the DRV8308 and vice versa?
    1. Rick already mentioned some of the benefits of using DRV8308 over DRV10970: gate drivers allow you to use FETs for your needs. Also voltage range is higher for the DRV8308 vs the DRV10970
    2. Gate drivers equate to more board space but more flexibility, where you get what you get in an integrated solution.
    3. Otherwise, it looks like your application parameters seem just fine with both of our devices.

    Hope this helps,

    -Cole

  • 0 in reply to Cole Macias
    Intellectual 920 points

    Thanks for the detailed response Cole. I think that helps with my decision making.

    There's one additional item/feature that I'd still like to get clarification on: drive angle and how it affects torque. Can either of you give any explanation?

    Thanks,

    Jim

  • 0 in reply to Jim Gavin
    TI__Mastermind 18900 points

    Hey Jim,

    I'm sorry for the delay, let me answer your question:

    In the DRV10970 datasheet, the section 8.4.3 talks about Adaptive Drive Angle Adjustment (ADAA) and Drive Angle Adjustment (DAA). The purpose of this feature is to align the BEMF and current when driving the motor. When the current and BEMF are not aligned, there is some error introduced into the torque applied to the motor. This means, the torque does depend on the drive angle--whose goal is to put the BEMF in phase with the current.

    The math to prove this is rather complex but I encourage you to watch the training video on commutation techniques for three phase brushless DC motors for more information about the relationship between current and torque.

    To summarize, yes the drive angle can affect torque. However, as long as the drive angle is changed to align the BEMF with the current, you should not have a problem.

    Hope this helps,

    -Cole