DRV8302: Buck Converter Problem

Hi Cole 

Thank you so much for your information.

I will do it by following your recommendation.

But now I got another problem.

I used external 5v power, not use buck converter. And supply 3.3V to OC_ADJ pin 12V for PVDD1. I checked voltage of GVDD.

It is 12.68 V. 

Do you have any idea for that problem.

Thank you so much.

Hello Cole~

Thank you for your reply.

I kept 2 caps (2.2uF) between GVDD and GND. Then I changed power supply 24V to PVDD1. 

I checked voltage of GVDD pin, it is 12.68V (the same before). 

After that I provide 3.3V on INH_A, and kept 3.3V and EN_GATE. But voltage on PIN GH_A is 33V after I checked it.

I do not know why is that problem.

Do you have any information for that problem

Thank you so much

Hello Vanquang,

Reviewing with the team, you have some conflicting data between the GVDD and the GHx results.

As a result, please complete the following actions and see if the issue goes away:

1. Repeat your measurements with one of the caps on GVDD removed
   1. Too much capacitance can result in stability problems (as mentioned above)
2. Review the voltage ratings on the caps for GVDD and make sure they are above the recommended 16V (>20V is even better)
   1. Too low of voltage rating will result in a lower effective capacitance. GVDD is acting like an LDO and too low of capacitance can result in stability problems
3. Repeat your measurements measuring at the GVDD pin and GND pin, PVDD1 pin and GND pin, and GHx pin and GND pin. Use the math function on the scope to also display GHx - GVDD and PVDD - GVDD. Do not probe at the GND next to capacitors near those pins, probe the ground pin directly
   1. Because your voltage do not apply and we do not expect a 3V drop through the diode from GVDD so we do not think the measurements were done correctly on either the GVDD or GHx.
   2. If you have gotten to step 3, we are looking if these waveforms can tell anything. We might find that the probing location makes a difference, but if we don't the above waveforms will help us understand the problem

Best,

-Cole

Hello  Cole.

Thank you for your information.

Here is my information 

1. I removed 1 cap 2.2uF, and kept only one. EN_GATE = INL_A = INH_B = 3.3v. Then voltage of GH_B = 14V, GL_A = 12.5V.  Is there any problem if I use GH_B = 14V?

2. Voltage of cap ratings is 16V. If I changed 2.2uF, 50V. Then the result is the same. 

3. In all my measurements until now, I used the ground pin directly.

Do you think DRV8302 has the problem?

Many thanks.

Hello Cole~

Yesterday, I did not work for it.

Now I am working in it.

I resoldered again GND thermal pad. And then I measure resistance of internal pull down resistor in pin EN_GATE, INH, INL. they are abot 2 to 3 M ohm.

Then I set up EN_GATE = INH_B = INL_A = 3.3V. voltage of GVDD = 11.90V, voltage of GH_B  = ~ 1V. GL_A = 11.90V.

So my problem not resolve yet.

Now what I have to do to resolve my problem?

Thank you so much

Best.

Hey Vanquang,

Yeah, this is pretty frustrating, apologizes for the trail and error and thanks for your patience.

Now it looks like your GH_B isn't working correctly. The GVDD has much more margin from the abs. max now, so I wouldn't worry about it but I think it still important to find the root cause.

I want to meet with the team again and see if we can give you some better guidance. In preparation for that, can you also provide the snippet of the schematic relating to the FETs?

Thanks,

-Cole

Hello Cole~

Thank you for your reply

Here is my schematic 

Hey Vanquang,

The team thinks we're still dealing with assembly issues. Here are key points from the discussion.

• No waveforms have been provided. Though you have given constant values in text in place of the various waveforms, we haven't evaluated noise, coupling, interference, or any parameter that could help be a clue to the solution to the problem
• Voltage rating and capacitance on GVDD have verified. We realized the charge pump cap has not been verified, please go verify the ratings (PVDD1*2 = 24*2 = 48V < voltage rating of charge pump cap)
• The board you have shown is clearly hand assembled. The replacement of the DRV8302 drastically changing the value of GVDD is enough to cause suspicion that this is the problem

• • For context, none of our EVMs have been close to this voltage. If you have an EVM from us, you can check the voltage on the board, then swap the devices from your board to EVM and vice versa. This will be able to see if the problem follows the part or the board. 
  • As such, you can try and assemble another board and monitor the voltage, we suggest getting an expert to manufacture it
  • Some more evidence to this is that the vias under the power pad seem a bit larger than expected (we could be wrong about this because only a picture was provided instead of gerbers or suitable layout files). Either way, more solder is needed than expected which means most of the solder will go through the vias and not swell up on the power pad. This has caused issues like this in the past

• Don't see anything wrong with FET schematic 
• Earlier you mentioned "I used external 5v power, not use buck converter". We don't understand the context of this statement. Can you confirm this statement is unrelated?
• Also, please feel free to correct the procedure, as we assume this is your set up:

• • Do not attach motor, attach voltage probes to GVDD, GH_B, GL_A, INH_x (at least one of them verification) and appropriate ground lead to GND pin
  • PVDD1 = 24V, wait for rails to power on
  • Use 3.3V external supply attached to INL_A, and ENABLE. Turn on supply
  • Monitor GL_A voltage
  • Use another 3.3V external supply attached INL_B, turn on supply and charge bootstrap cap
  • Move 3.3V supply to INH_B to switch high side on where pull down switches INL_B off
  • Monitor voltage on GH_B

Best,

-Cole

Hello~ cole

Thank you for your email.

I am sorry to reply so late.

From your email, I think my problem is from two issues.

1. My solder was not good, especially for DRV8302.

2. In thermal pad of DRV8302, Via diameter is 1.2mm, and via drill is 0.8. These parameter are so big I think.

For "I used external 5v power, not use buck converter". I mean I did not use  the buck converter of DRV8302 to make 5V. I used external 5V power supply. Then I connected this 5V supply to my PCB, and my PCB converted down to 3.3V. 

Thank you for your support.

Thanks for the follow up Vanquang.

Feel free to a new question, and click the ask related question if button if its applicable to this project specifically.

Thanks for being receptive to feedback.

Best,

-Cole