Due to the U.S. Thanksgiving holiday, please expect delayed responses during the week of 11/22.

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.

DRV8312: DRV8312, DRV8332 Power up and down sequence

Part Number: DRV8312
Other Parts Discussed in Thread: DRV8332,

Hi 

I have a question about DRV8312 and DRV8332 power up and power down sequence.

According to the datasheet, it mention that there are no requirement about power up and down sequence.

But it mention that RESETx require holding low state while powering up the device.

Also it mention that RESETx require holding low state during power down.

I plan to generate 12V for GVDD and VDD, from 24V(PVDD). In that case,

  - is it necessary to RESETx holding low state when 12V turning on?

  - In case of turning off 12V(equivalent to 24V) suddenly, is it necessary to RESETx holding low state?

We are considering whether to attach a circuit that satisfies both of these externally, so please let us know your opinion.

Best regards

Naoki

  • Hello Naoki,

    Yes, seems like a good plan to hold RESETx for turning on 12V and turning off.

    Note that these are not necessarily specifications for the device, just good practice (which is why it is in the application section). The goal is to set the driver in a known state (which is, stop driving the motor) because voltage transients are happening at this time which are good to protect against. Specifically in the case of turning off, we don't want gates to be asserted when we've made the decision to power down so this just prevents the possibility of that happening. We have seen that the outputs might be asserted with some nonzero voltage and the motor might move a little bit (e.g. <0.3 degree if the interia is low) if the inputs to the device are uncontrolled during the power off state.

    Best,

    -Cole

  • HI 

    I want to ask you further question about power up sequence of PVDD, GVDD and RESETx.

    - If only PVDD(24V) is ON(GVDD is OFF),are there any concern? 

    - As I mentioned before GVDD(12V) will make from PVDD(24V).

       If PVDD is off, GVDD is also off. So I think there are no concerns (such as moving a little bit) whatever RESETx's state is. Is my thinking correct?

    - I  plan to turn on GVDD by putting RESETx in the EN pin of DCDC or LDO.

      Should there be such a circuit?

    I'm thinking of a proper sequence of operations for P VDD, G VDD, RESETx and would appreciate any advice.

    Best regards

    Naoki

  • Hello Naoki,

    Let me give you a proper response tomorrow, 12/9/20.

    Best,

    -Cole

  • Hi 

    I want to ask you additional relevant questions.

      -I found that RESETx absolute max voltage is "VREG + 0.5".

       Does It mean if GVDD and VDD are off, RESETx cannnot receive any signal?

      (So far, I was reading older datasheet revised 2013, I found current datasheet revised 2014  changed this spec)

     - According to the datasheet, it is mentioned that "recommended to design the 12-V power supply with current capability at least 5-10% of your

    load current and no less than 100 mA to assure the device performance across all temperature range."

       Does load current mean motor load current? For example If the motor load current is 5A, Does 12-V power supply need 500mA?

    Best regards

    Naoki

  • Hello Naoki,

    If only PVDD(24V) is ON(GVDD is OFF),are there any concern?   Does It mean if GVDD and VDD are off, RESETx cannnot receive any signal? I'm thinking of a proper sequence of operations for P VDD, G VDD, RESETx and would appreciate any advice. If PVDD is off, GVDD is also off. So I think there are no concerns (such as moving a little bit) whatever RESETx's state is. Is my thinking correct?

    There's a lot going on with these questions so I'm hoping I can just explain the power tree to answer most your questions.

    VDD specifically powers the digital circuitry in the device and VREG. This interprets inputs from the RESETx and PWMx and tries to drive the gates. Without this supply none of the logic will work as expected so its recommended that this powers up first. Essentially, the OUTx never sees any logic input and its considered "floating"

    GVDD is only needed for the boost voltage or gate drive voltage. Most people tie this to VDD so the rails will come up at the same time. Without PVDD, the undervoltage protection will trigger which means the OUTx pins will be made high impedance and not respond to RESETx and PWMx inputs.

    Because GVDD and VDD are usually powered up at the same time, as soon as the power on reset criteria is fulfilled, the device is ready to commutate and the device will try to respond to RESETx and PWMx inputs. As such, this is why we recommend users to change the state of RESETx, which operates on a supply that's independent of the DRVx, so that the gates won't try to commutate.

    PVDD was designed to work as a split supply. That means, inside of the part and on the DRVx die, the GND1,2,3 and PVDDx are not connected to anywhere else (with the exception of some ESD diodes that should never turn on, note the 0.3V abs. max ratings). This means PVDDx was designed to be powered separately from GVDD in most cases (up to 70V abs. max for PVDD where GVDD is only 13.2V abs. max). Without out this voltage, not much will happen but also, the motor can't spin and current won't be delivered through FETs. So, there's not a whole lot of risk to keep it off. I would recommend powering this after VDD and GVDD are up.

    I  plan to turn on GVDD by putting RESETx in the EN pin of DCDC or LDO.

    This might be a good idea. I'm intepretting this as the MCU or whatever controls RESETx and PWMx are finished powering up and the MCU has control over the inputs before allowing GVDD and VDD to turn on.

    Does load current mean motor load current? For example If the motor load current is 5A, Does 12-V power supply need 500mA?

    This text in the datasheet refers mostly to the drift in current consumption by the device. You can generalize them as I_Supply_limit > I_VDDandGVDD. The datasheet is saying don't make I_Supply_limit = 50mA because process variation and temperature might have DRV8312s that will consume more current than expected.

    As such, a good rule of thumb is I_Supply_limit > I_VDDandGVDD > I_Motor*(0.10). As you said,  I_Supply_limit > 5A(RMS)*10% > 0.5A as you calculated

    Let me know if you have any questions.

    Best,

    -Cole

  • Hi 

    Thank you for answering my questions. I want to ask you additional question

    Regarding PWM_X and /RESETX 's absolute MAX voltage is between -0.3V to VREG + 0.5

    You mentioned if I use out of this rule, I understood tDRV8312/8332 cannot respond correctly.

    I thought absolute max meaning is device breaking voltage.

    Your answer does not mentioned about this matter. This absolute max is not mean breaking voltage, isn't it?

    I asked because there is a mode in which PVDD and GVDD are off in our system.

    Regarding 12V power supply load current

    I understood 12V supply current need I_supply_limit *10%.

     -Does I_Supply_limit mean RMS? How should I interpret when driving a motor in a sine wave?

     -I don't understand why 12V power supply is relevant to motor control current.

      I think it depends on the device temperature, right? 

      Does it mean that the temperature of the device is replaced with the current value?

    Best regards

    Naoki

  • Hello Naoki,

    No problem, let's see if I can answer some of the questions.

    Regarding PWM_X and /RESETX 's absolute MAX voltage is between -0.3V to VREG + 0.5

    Let me break this down a bit as there's also a lot to this if you haven't seen semi-conductor design before.

    To start exceeding Absolute maximum will cause damage to the part, notes on this shown in the datasheet statement below:

    In the case of PVDD = ON and GVDD = OFF (which means VDD = OFF, but you can correct me if I'm wrong that VDD is supplied from a different voltage from GVDD), we know that VREG will not be powered up. This means, VREG = 0V or floating (depending on if you think charge is settled for a long time and total current = 0A). As a result, the RESET_X and PWM_X absolute maximum = 0.5V. Anything higher than that falls under the datasheet statement above.

    The reason why I bring this is up is that it gives a clue about how the PWM_X and RESET_X circuit is designed. Every device with a digital core needs a digital rail. These are usually 0.9V, 1.2V, 1.8V, and 3.3V. So anything voltage higher than that means there is a regulator need to step down to that voltage (i.e. VDD to VREG). Traditionally, a GPIO input can be simplified by the below (which I am borrowing from STM32 GPIO lecture):

    While this isn't necessarily the structure inside the DRV8312, its very close. Specifically, those protection diodes at the front of the GPIO need to be powered by some rails. As a result, most of the digital type logic is source from the same logic rail that's powering the digital core. So VDD = VREG for our purposes and VSS = GND.

    So, whenever you see a -0.3V or +0.5V or something like that in an abs. max of a datasheet, you should interpret that as protection diodes are on this pin as the 0.3V and 0.5V represent the forward voltages needed to turn on the diodes and start "protecting" the circuit. If this current or voltage is "large" it will blow and the protection diode will be useless and you would see a short or low impedance between VREG and PWM_X or VREG and RESET_X if you damage the protection diode.

    So with all this knowledge we can say that "Does It mean if GVDD and VDD are off, RESETx cannot receive any signal "because it will blow up the protection diodes.

    Does this make sense?

    Regarding 12V power supply load current

    Does I_Supply_limit mean RMS? How should I interpret when driving a motor in a sine wave?

    Yes I_SupplyLimit means RMS. Anything transient should come from decoupling capacitors or be protected against.

    When you said 5A(RMS) I assumed you had already done the conversion for whatever current you're driving. For sinusoidal calculation, the RMS calculation is easy. I_PeakSine/sqrt(2) = I_RMS. In the case of 5A RMS you gave earlier, I would assume that motor current you see on the oscilloscope is 7.07A_pk which results in 5A_RMS as the value to be used in the equation..

    I don't understand why 12V power supply is relevant to motor control current.  I think it depends on the device temperature, right?

    Exactly, temperature is the key careabout and 12V just refers to the marketing terminology of what voltage level this part was designed for (not to say any other voltage is bad or untested, but 12V was the key voltage we wanted to market towards).

    The theory behind the statement assumes that you will use the device at TJ = 150C.  If you system requirements are lower, then you probably won't need the 500mA.

    Unfortunately, I can't really give a rule of thumb below the one that's given. Its much more common for an engineer to have a power budget in mind, build something and test it, and then either replace the power supply or replace the motor drive. 

    Best,

    -Cole