DRV8312: Power dissipation calculation

Hello Nir,

Unfortunately, I don't have a solid answer for you besides, maybe.

In general, yes, the DRV8332 was designed to withstand currents that are high only with the use of a heat sink. Those heat sink recommendations are outlined in the Thermal Considerations (section 11.3) in the datasheet.

Calculations:

If you want to do some calculations, you'll need to calculate the power dissipated equation and then multiply it by the R_JC_top added to the heat sink thermal resistance and thermal grease thermal resistance.

This turns into:

T_J = P_D + (R_JC_top + R_heatSink + R_thermalGrease) + T_A

Where:

• R_JC_top = the thermal resistance from junction to case which can be found in the datasheet
• R_heatSink = thermal resistance of heat sink, given my manufacturer
• R_thermalGrease = thermal resistance of thermal grease, given my manufacturer
• P_D = Power Dissipated (more on that below)
• T_A = Ambient temp, which is 85C in your case
• T_J = Junction temperature which must be below 150C, to be considered safe 

Power Dissipation:

The power dissipated equation is as follows:

P_D = P_RDS + P_SW + P_GVDD + P_VDD + P_PVDD

Where:

• P_RDS = Conduction loss  = IRMS^2*RDS(ON) * N, N = number of MOSFETs ON 
• P_SW = Switching loss = ½*VM*IRMS * N * (trise + tfall) * fPWM, N = number of MOSFETs switching per cycle
• P_GVDD = Supply current for GVDD  = V_GVDD * (3*I_GVDD_X). Typical current in datasheet, one for each half bridge
• P_VDD = Supply current for VDD = (V_GVDD - V_VDD)* I_VDD. Current and voltage in datasheet

Then you plug in all those numbers and see if the T_J exceeds 150C. If it excessively exceeds 150C, it definitely can't work. If it is close to 150C, you may or may not be able to get it lower with good layout techniques (you can pay to get your layout simulated by many 3rd party vendors who will replace the R_x section of the equation. If it is lower, then, with good layout, the device should be able to work with the device.

Note, these equations don't handle all of the sources of power dissipation but these should be 90% of the true value. To be fair, most of these are based on typical values and don't account for temperature or process variation.

The most direct, is to build a board with a good layout and see if it works or at least try an EVM (sadly, this part is so old that an EVM was not developed for it).

Best,

-Cole

Sources:

• https://www.ti.com/lit/an/slva938a/slva938a.pdf Thermal app note
• https://www.ti.com/lit/an/spra953c/spra953c.pdf Thermal app note
• https://www.ti.com/lit/ds/symlink/drv8312.pdf datasheet 
• https://www.ti.com/lit/an/slva959a/slva959a.pdf Layout guidelines