Driver Infineon-6EDL7141 - Signals propagate to MOSFET only if you pet the chip

[eTobey]

Hi,

i am already trying hours to get this chip working. The outputs will only get put through after about 2 - 30 seconds after starting of the chip. I already checked the under voltage and overvoltage lockouts, both, via SPI (error flags) and measuring. But everything looks fine and stable. I have read the values that i have written to the driver. They are correct.

After they propagate, then stays like that, and everything looks good.

These are my settings:
(Settings that are not shown here are left in default)

Code: [Select]

writeDataSPIDriver(ADDR_FAULTS_CLR, 0x0, 0x03); // clear faults


writeDataSPIDriver(ADDR_PWM_CFG, 0x0, BRAKE_HIGHZ << PWM_CFG_BRAKE_CFG_Pos); // no brake


uint16_t csampCfg = (CS_GAIN_8V << CSAMP_CFG_CS_GAIN_Pos) |
(CS_GAIN_PROG_DIG << CSAMP_CFG_CS_GAIN_ANA_Pos) |
(CS_A_DIS_B_DIS_C_DIS << CSAMP_CFG_CS_EN_Pos) | //current amplifier enable
(CS_BLANK_300ns << CSAMP_CFG_CS_BLANK_Pos) |
(CS_DEGLITCH_0us << CSAMP_CFG_CS_OCP_DEGLITCH_Pos) |
(OCP_FLT_TRIG_8 << CSAMP_CFG_CS_OCPFLT_CFG_Pos);

writeDataSPIDriver(ADDR_CSAMP_CFG, (byte) (csampCfg >> 8), (byte) csampCfg);

uint16_t csampCfg2 = (OCP_POS_THR_300mV << CSAMP_CFG2_CS_OCP_PTHR_Pos) |
(OCP_NEG_THR_300mV << CSAMP_CFG2_CS_OCP_NTHR_Pos) |
(OCP_FLT_LATCH_EN << CSAMP_CFG2_CS_OCP_LATCH_Pos) |
(CS_SENSE_RDSON << CSAMP_CFG2_CS_MODE_Pos) |
(OCP_FLT_BRAKE_DIS << CSAMP_CFG2_CS_OCP_BRAKE_Pos) |
(OCP_PWM_TRUNC_EN << CSAMP_CFG2_CS_TRUNC_DIS_Pos) |
(CS_VREF_INT << CSAMP_CFG2_VREF_INSEL_Pos) |
(CS_AUTOZERO_DIS << CSAMP_CFG2_CS_AZ_CFG_Pos);

writeDataSPIDriver(ADDR_CSAMP_CFG2, (byte) (csampCfg2 >> 8), (byte) csampCfg2);

https://www.infineon.com/dgdl/Infineon-6EDL7141-DataSheet-v01_08-EN.pdf?fileId=5546d46279cccfdb0179f4c1d3bf0756

[eTobey]

I just found out, that if you pet it, it works. 

Further more, it seems that it only works, if there is a sudden field change, that comes over the human body. No electrical connections are touched.

[Wolfram]

Did you leave any pins floating? Unconnected CMOS inputs will not have a defined value, and they are often sensitive enough that stray electric fields can influence the state.

Did you connect EN_DRV anywhere, or could it be floating due to a bad solder joints? QFNs are notorious for having bad joints that can be hard to spot

[Psi]

I just found out, that if you pet it, it works. 

I had to type that into DALL-E 3   

[eTobey]

Did you leave any pins floating? Unconnected CMOS inputs will not have a defined value, and they are often sensitive enough that stray electric fields can influence the state.

Ignorant me!  

Just set up Brake "High-z" and thinking i dont need a brake. But indeed, i left pin (Vsens - pin 13) therefore floating. It does not have a pull-up resistor. Strangly it uses a resistor to GND to set up start configuration. It also matches my observation, that touching it on the vsense pin region, it was more often reproducible.

Now that whole case gets interesting:
- Can i somehow send a little boost on the DVDD line to make it work?
- Will it work, and not stop at any time?

It would have been awesome, if i would have been able to use the DVDD that is right next to it. But Vsense has an open drain mosfet. So i might have to get a bit creative.