LMG1210EVM-012: Bootstrap Capacitor not charging up to 5V at every switching cycle

Hey Ck,

Thanks for reaching out here! Understanding the following will help us determine what's going on:

- I notice the PWM signal changes after the 2nd cursor in the screenshot. Are you manipulating the Enable at this point?

- Is there a reason that you are operating with a 0V as the bus voltage? Would you test if you experience any change by applying a Bus Voltage?

- Have you tried operating the EVM at a slower switching speed to verify if C-Boot is able to fully charge?

Thank you,

Grant 

Hi Grant, thanks for taking up my questions.

1. No PWM manipulation. CH4, green trace is the PWM as seen after the onboard buffer U2A at TP9.

2. Per section 7.1 of the SNVU572–January2018 file, I wanted to confirm the gate signals first before powering Vbus. Wanting to get the below as per Pg.13 of  SNVU572–January2018.

3.  Yellow = HB-HS, Red = HO, Blue = LO, Green = PWM @TP9
at 500kHz, C-boot is barely charged to 2V, about 1.7V. 
I noticed that this 1.7V is appearing at the Vbus rail. Is it normal?

 A zoom in view of above.

4. From day 1, the HO is as below
 Yellow = HO, Green = PWM @TP9

5.  Yellow = HO, Red = Vbus 24V, Blue = LO, Green = PWM @TP9
Tried with 24V Vbus. I'm not comfortable to power Vbus for further testing given the situation the board has not been giving HO correctly since day1.

Hey Ck,

With the 12V and 1.8W power draw, you get 150mA current which exceeds the spec for ILDOM of 100mA in the 5V LDO section of Section 6.5 in the datasheet. There is a chance the driver got damaged. Have you tried using a different driver on the EVM?

Would you provide a scope capture of the HI/EN via test point T8? In many of the above captures it looks like there are signals that are not outputting in accordance to the PWM signal. Also, many inconsistent LO pulses. If EN is not remaining high, then the driver will not output. 

Thank you,

Grant

Grant, If by different driver you mean driving HI/LI independently, No, I don't have the equipment to do so.
Also, my final app is to use this EVM directly and PWM mode is preferred.

The EN scope traces:

Hey Ck,

Based on the waveform of Enable it seems that your LMG1210 device may be damaged. The Enable is tied to the LDO output and should be constant at ~5V. The ripple you are seeing every time HO switches is abnormal. 

By using a different driver, I mean have you tried replacing the LMG1210 device on this EVM Board? By replacing the LMG1210 we can verify if the root cause comes from the gate driver itself, or if the issue comes from the board/other setup.

Best,

Grant 

i don't have the correct tools and equipment to rework to replace the LMG1210 device.
Anyways I can claim warranty on this?
Since this is the result I got from the EVM on day 1 using it.

Hey Ck,

Without verifying the health of the gate driver it will be hard to pinpoint if there are other factors at the root cause of the issue. 

Unfortunately as this was purchased through Digikey I am not sure what the warranty process will look like. Their return policy does state the following: "All returns must be in the original packaging, unused and like-new condition." Considering you have used the EVM, they may be able to claim any incorrect operation stems from usage. I'm sorry I cannot give better insight on that end. 

Best,

Grant