L6470 Stepper Motor Driver Vreg No voltage

[Jagadish_93]

I have attached two images of the L6470 Motor diver, In the new design PCB the motor drivers are heating up and don't seem to respond to SPI commands. No voltage is present on the VREG pin. The sense resistors were added basically for measuring the winding current. I have added Led on the Flag pin too. I have highlighted the changes in the Image for reference.

[Peter BENSCH]

Welcome @Jagadish_93, to the community!

you must never apply ADCIN to voltages like VS (see data sheet of the L6470, section 7.5 and abs max ratings in table 2) and should do this via a voltage converter that brings the shunt voltage down from VS level to GND level.

Hope that helps?

Regards
/Peter

[Jagadish_93]

Hi Peter,

I was aware of this that's why I chose a tiny sense resistor so that the drop would be in the order of millivolts, so that even at nominal 12Volts the sense resistor drop would be never reach 3Volts 

[Peter BENSCH]

Unfortunately, you are making a mistake: the differential voltage across your shunt is related to VS. The potential at VSA, VSA2, VSB and VSB2 (let's call it VSA) against GND is therefore smaller than VS by the voltage at the shunt: VSA = VS-Vshunt.

ADCIN, however, measures against GND and within your schematics gets almost the full voltage VS, so that the current barely limited via R22 can flow into ADCIN. It is almost certain that you have destroyed the L6470 and can therefore no longer communicate with it.

My proposals: install a new L6470 and measure the current through the shunt with a current sensing opamp, such as the TSC2010, if you absolutely want to measure on the high side.

[Jagadish_93]

Hi Peter, Would it be okay to leave the ADCIN floating, I have cut the traces connected to the sense resistor.
Now the ADCIN voltage pin is floating

[Peter BENSCH]

On the lab bench, it should not be a big problem to leave ADCIN floating. For a final product, however, I would recommend either pulling ADCIN to GND via a pull down when not in use, or connecting it to VS via a voltage divider as on the EVAL boards.

[Jagadish_93]

Hi Peter,
Thanks for the info, the previous circuit was damaging the IC immediately, after cutting the traces and replacing them with new iCs it's working. I have attached the updated design I'd appreciate your assessment if possible.

[Peter BENSCH]

At least the surroundings of the L6470 look ok except for Q1. Q1 is missing the load capacitors of about 12...18pF at OSCIN and OSCOUT (see also data sheet, fig 11).

[Jagadish_93]

Q1 is DNP, those pins are floating as of now.