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.

DRV8308: DRV8308 driving AIRFAN MFA0500 - the driver ic getting heated up and the blower not delivering the maximum 250 lpm at 100 percentage of pwm

Part Number: DRV8308
Other Parts Discussed in Thread: , CSD18540Q5B

HI all

I am driving airfan mfa 0500 BLDC blower using drv8308 driver, i am using stm32 to program it and to generate the pwm and control the drv8308 the issue is at 100 percent pwm the drv8308 is getting heated up and stops the blower after 5 minutes of run time. The current drawn at 100 % pwm is 6 ampere and also the output of the blower is not 250 lpm which the airfan datasheet denotes at 24v. I want to know whether i can run the airfan mfa 0500 blower with the drv8308, do the specs of drv 8308 supports airfan mfa0500 below is the schematic of the drv8308

20040367_A.pdf

The below is the airfan blower data sheeet

Air Fan MFA 0500.pdf

the below is the drv8308 setting 

#define DRV_CTRL_CFG_0 (9<<AG_SETPT) | (1<<DIRPOL) | (1<<SYNRECT) | (3<<PWMF) | (1<<SPDMODE) | (1<<FGSEL) | (0<<BRKMOD) | (1<<RETRY)
#define DRV_CTRL_CFG_00 (9<<AG_SETPT) | (1<<DIRPOL) | (1<<SYNRECT) | (0<<PWMF) | (1<<SPDMODE) | (1<<FGSEL) | (0<<BRKMOD) | (1<<RETRY)
// Address 0x01 Configuration
// --------------------------
#define DRV_CTRL_CFG_1 (132<<ADVANCE)

// Address 0x02 Configuration
// --------------------------
#define DRV_CTRL_CFG_2 (16<<SPDREVS) | (1<<MINSPD)

// Address 0x03 Configuration
// --------------------------
#define DRV_CTRL_CFG_3 (1<<BASIC) | (7<<SPEEDTH) | (2048<< MOD120)

// Address 0x04 Configuration
// --------------------------
#define DRV_CTRL_CFG_4 (0<<HALLRST) | (1<<DELAY)| (0<<AUTOADV) | (0<<ENSINE) | (1<<TDRIVE) | (5<<DTIME) | (1<<IDRIVE)

// Address 0x05 Configuration
// --------------------------
#define DRV_CTRL_CFG_5 (0<<INTCLK) | (10<<SPDGAIN)

// Address 0x06 Configuration
// --------------------------
#define DRV_CTRL_CFG_6 (1<<HALLPOL) | (0<<FILK1)

// Address 0x07 Configuration
// --------------------------
#define DRV_CTRL_CFG_7 (0<<FILK2)

// Address 0x08 Configuration
// --------------------------
#define DRV_CTRL_CFG_8 (0<<BYPCOMP)|(0<<COMK1)

// Address 0x09 Configuration
// --------------------------
#define DRV_CTRL_CFG_9 (9<<AA_SETPT) | (0<<COMK2)

// Address 0x0A Configuration
// --------------------------
#define DRV_CTRL_CFG_A (3<<OCPDEG) | (3<<OCPTH) | (10<<LOOPGAIN)

#define DRV_CTRL_RESET_SPEED	0
#define DRV_CTRL_CLEAR_FAULT	0


#define SFM3019_ADDRESS_READ 0x5D                  //Actual is 2E. We need to shift left by 1 to make space of R/W bit
#define SFM3019_ADDRESS_WRITE 0x5C                 //Actual is 2E. We need to shift left by 1 to make space of R/W bit

The below is the pwm vs flow rate which I got practically from the airfan

INPUT VOLTAGE - 23.8V
SNO  PWM(%) FLOWRATE IN LPM(TSI) AMPERE(A) POWER CONSUMED(W)
1 0 0 0.045 1.071
2 10 5.03 0.09 2.142
3 20 31.58 0.444 10.5672
4 30 53.08 0.841 20.0158
5 40 74.44 1.26 29.988
6 50 94.96 1.712 40.7456
7 60 112.56 2.23 53.074
8 70 128.19 2.813 66.9494
9 80 140.51 3.362 80.0156
10 90 150.04 3.92 93.296
11 100 163.43 5.092 121.1896

Any thoughts please share me.

  • Hi Nagaraj,

    The MOSFET you are using has very small Qgd value. (2.3nC) I think you are using 20mA IDRIVE setting. Can you use the smallest setting (IDRIVE = 10mA) to see if power is decreased? Chances are they may be a shoot-through event occurring at the higher duty cycles if the low side FET is turning on through gate voltage coupling. 

    Have you tuned your motor correctly using section 4 of the DRV8308EVM User's Guide?

    https://www.ti.com/lit/pdf/slvua41

    Thanks,
    Aaron

  • Thankyou Aaron for the reply 

    "The MOSFET you are using has very small Qgd value. (2.3nC) " yes but the drain current is 16 ampere for this ic so i thought it would drive a motor of 90w at 24v. Should i select a mosfet of high Qgd value and what are all the parameters i should consider to select the mosfet for this driver.

    I think you are using 20mA IDRIVE setting. Can you use the smallest setting (IDRIVE = 10mA) to see if power is decreased?

    i changed to 10 ma though the driver is heating and also the mosfet is heating. Today while testing the mosfet is also burnt.

    Have you tuned your motor correctly using section 4 of the DRV8308EVM User's Guide?

    since the mosfet has been burnt i could not implement the steps which the document has provided will implement the points and check with the mosfet replaced.

    Thankyou

  • Hi Nagaraj,

    Understood. 

    Can you check on an oscilloscope the waveforms of, for instance, UHSG and ULSG to see if the 10mA gate drive current is causing an overlap in gate drive voltages with respect to their sources (UHSG-U; ULSG-GND). If so, if they're both on at the same time (Vgs > Vth), then shoot-through current is likely causing the MOSFET heating to occur. 

    In which case, higher IDRIVE currents may be required. 

    Thanks,

    Aaron

  • Hi Aaron since the mosfets are burnt im in replacing of mosfets can you suggest me any mosfets , i need to replace the mosfet and check the waveform in the oscilloscope and will update you in this forum.

  • HI Aaron will this DRV8308 can drive the airfan blower which i have mentioned, since the  the airfan blower is drawing current of more than 5.0 A at 60 %pwm and gets heated up and gets burned. Can you suggest me whether the circuit will support running our airfan blower.

    Thankyou 

  • Hi Nagaraj, 

    We will review further and aim to provide a response by middle of next week 

    Best Regards, 
    Andrew 

  • Hi andrew 

    ok

  • Hi Nagaraj,

    Hi Aaron since the mosfets are burnt im in replacing of mosfets can you suggest me any mosfets , i need to replace the mosfet and check the waveform in the oscilloscope and will update you in this forum.

    You may want to try CSD18540Q5B, which has Qgd of 6.7nC. Here are all TI N-channel MOSFET suggestions: https://www.ti.com/power-management/mosfets/n-channel-transistors/products.html

    HI Aaron will this DRV8308 can drive the airfan blower which i have mentioned, since the  the airfan blower is drawing current of more than 5.0 A at 60 %pwm and gets heated up and gets burned. Can you suggest me whether the circuit will support running our airfan blower.

    I will review your circuit today and provide any feedback if there are any critical issues. 

    In the meantime, please ensure your gate drive signals are not encountering any shoot through conditions by monitoring their respective Vgs voltages. MOSFET heating occurs at higher duty cycles if there is overlap in gate turn on behavior, or if when switching the MOSFETs, if current is coupling into the low-side MOSFET like below. 

    High-power design app note: https://www.ti.com/lit/pdf/slvaf66

    Thanks,
    Aaron

  • Hi Aaron

    You may want to try CSD18540Q5B, which has Qgd of 6.7nC. Here are all TI N-channel MOSFET suggestions: https://www.ti.com/power-management/mosfets/n-channel-transistors/products.html

    Thankyou I will consider this for selection of the MOSFET.

    I will review your circuit today and provide any feedback if there are any critical issues. 

    Thankyou 

    I have measured the oscilloscope signal of U phase UHSG and ULSG with respect to ground at 100 percent PWM. The below is the image of the oscilloscope with 

    Yellow line - (UHSG - GND)

    Blue line - (ULSG - GND)

    The below is the image of the of the oscilloscope of UHSG - U and ULSG - GND

    Yellow line - (UHSG - U)

    Blue line - (ULSG - GND)

    (Since the negative terminal of all the probe in oscilloscope are connected together I could not measure this for higher PWM so I measured it for only low PWM.)

    Thankyou

  • Hi Nagaraj, 

    I believe what is causing your excessive heating is the events shown on UHSG-U and ULSG-GND. The white circles show shoot-through events likely occurring, which causes excessive heating on your MOSFETs at the higher duty cycles. 

    To fix this, you would need a higher IDRIVE current or FET with smaller Qg value because there is likely too much capacitance at the FET gates, which are causing slow turnoffs for the high side FET gate and slow turnons for the low side FET gate. 

    Can you try increasing the IDRIVE current to higher settings to see if the heating is reduced?

    Thanks,
    Aaron

  • Hi Aaron

    To fix this, you would need a higher IDRIVE current or FET with smaller Qg value because there is likely too much capacitance at the FET gates, which are causing slow turnoffs for the high side FET gate and slow turnons for the low side FET gate. 

    Thankyou for the suggestion I will consider this. I will try with smaller Qg  value FET.

    Can you try increasing the IDRIVE current to higher settings to see if the heating is reduced?

    yes I tried the setting with high current IDRIVE of up to 100 mA though the IC is getting heated up and the output was not as expected. Meanwhile I tried with another blower from Micronel make this blower is also a BLDC blower which gives the maximum throughput of flowrate as desired, only the Airfan blower is drawing more than 4 A and the flow rate is also less but the micronel blower is drawing less current of 3.3 A at 100 percentage of duty cycle at 24v. What would be the difference between two blowers with one causing higher current for same PWM duty cycle any thoughts please share me.

    Thankyou 

    Nagaraj JP

  • Hi Nagaraj,

    What would be the difference between two blowers with one causing higher current for same PWM duty cycle any thoughts please share me.

    I would ensure the speed loops are turned correctly for each motor. Is the SYNRECT bit enabled? 

    Do you have layout of your PCB? Can you share if possible?

    Thanks,
    Aaron

  • Hi Aaron

    I would ensure the speed loops are turned correctly for each motor

    The below is the settings for the drv8308 registers.

    #define DRV_CTRL_CFG_0 (9<<AG_SETPT) |(0<<ENPOL)| (1<<DIRPOL) | (1<<SYNRECT) | (0<<PWMF) | (1<<SPDMODE) | (1<<FGSEL) | (0<<BRKMOD) | (1<<RETRY)
    // Address 0x01 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_1 (132<<ADVANCE)
    
    // Address 0x02 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_2 (16<<SPDREVS) | (1<<MINSPD)
    
    // Address 0x03 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_3 (1<<BASIC) | (7<<SPEEDTH) | (2048<< MOD120)
    
    // Address 0x04 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_4 (0<<HALLRST) | (1<<DELAY)| (0<<AUTOADV) | (0<<ENSINE) | (1<<TDRIVE) | (5<<DTIME) | (0<<IDRIVE)
    
    // Address 0x05 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_5 (0<<INTCLK) | (10<<SPDGAIN)
    
    // Address 0x06 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_6 (1<<HALLPOL) | (0<<FILK1)
    
    // Address 0x07 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_7 (0<<FILK2)
    
    // Address 0x08 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_8 (0<<BYPCOMP)|(0<<COMK1)
    
    // Address 0x09 Configuration
    // --------------------------
    #define DRV_CTRL_CFG_9 (9<<AA_SETPT) | (0<<COMK2)
    
    // Address 0x0A Configuration
    // --------------------------
    #define DRV_CTRL_CFG_A (3<<OCPDEG) | (3<<OCPTH) | (10<<LOOPGAIN)

     Is the SYNRECT bit enabled? 

    yes I have enabled the SYNRECT.

    Do you have layout of your PCB? Can you share if possible?

    yes the below is the pcb layout

     vital.PcbDoc

    Thankyou

    Nagaraj JP

  • Hi Nagaraj,

    I do see a big difference between the schematic you shared and recommended parts in the datasheet. We recommend 0.1uF, 35V rated capacitor for CPH-CPL flying capacitor. 

    You are using 0.47uF, 100V rated capacitor. This cap is large and may not be switching the charge pump correctly by design to provide the VCP charge pump current used to provide IDRIVE currents. 

    This may be why there is not enough load current to turn off the FETs (VGS going low), so it has more a RC curve rather than a fast switch on/off to sink the gate current and turn on/off the FETs. Since Vth of the MOSFETs = 3V, it can cause both FETs to turn on at the same time causing the FET heating. 

    Can you replace the cap with a 0.1uF cap? You only need 35-V rating so a smaller cap can be used, i.e. 0805. 

    Also as well, I recommend direct connects on vias/pads, not thermal connects, to improve thermal dissipation and reduce trace inductance. 

    Thanks,
    Aaron

  • Hi Aaron

    Thankyou for the support you have given.

    The problem is with the blower itself, Today we bought a new blower and checked the characteristics it was giving 250 lpm at 100 percentage pwm and the current is also 2.7 ampere at max. So the problem is with the blower and i am very sorry about that.

    Now the issue is that the drv8308 start only at 9 % pwm and for lesser than 9% of pwm the drv8308 does not gives any output so the minimum flowrate we can achieve is that 2.2 lpm for 9% of pwm, is there any ware i am missing that the precise control of the blower can be achieved.

    Thankyou

    Nagararj JP

  • Hi Nagaraj,

    Aaron will get back to you on this next week once he has had a chance to look over this in more detail. 

    Regards,

    Anthony Lodi

  • Hi Nagaraj,

    No worries about the blower causing the issues. 

    But for the variation in blower, you may need to re-tune the DRV8308 for this new blower if there are different resistance/inductance characteristics using the tuning guide here: https://www.ti.com/lit/pdf/slvua41

    Thanks,
    Aaron

  • Hi Aaron

    But for the variation in blower, you may need to re-tune the DRV8308 for this new blower if there are different resistance/inductance characteristics

    Yes sure I will consider the above while using a new blower.

    Thankyou

    Nagaraj JP