USB PD with MPQ4214 IC

DC-DC Power Converters
1

I’m trying to figure out whether the advertised USB PD for this device is at all achievable. For this, an output voltage range of 5V to 36V should be achievable. As of now, we are testing with the evaluation board, which by default would not be able to supply for this range, given that Vref is programmable from 300mV to 2.047V, and the voltage divider allows for a maximum voltage of about 20V on the EVQ4214-U-00A. To address this issue, the voltage divider resistor R2 was exchanged from 10kOhms to 5.1kOhms, which allows for a 38.2 V maximum voltage if my math is not wrong.

But on doing so, the IC looses the precision it used to have on the past voltage divider resistor configuration. And I’m wondering whether it’s my math that’s wrong or if it’s just that I’m pushing the IC well over its operational limits. Steady state error seems to be around ~0.5V now that we’ve changed R2 value (that’s if my math is right), added to this the IC caps at 33.5V rather than the expected 38.5V. I do believe the issue should be one of the following:

  1. My math’s wrong
  2. Should I change the Compensation components now that I’ve changed R2 value? If the answer is yes, how should I determine what the new values of the compensation components should be?
  3. The IC is designed to operate on 36V output, but the evaluation kit is not, could you confirm this?

For a reference, I based my calculation off of this PDF file
PDF with my math

2

Also, it’s needed for the device IC to “boot up” on 5V. With the new voltage divider resistor configuration, it boots to ~8V, my proposed solution is to configure the ADDR pin to be either 64H or 66H
image
Which would boot the IC on “no output mode” (ENPWR default value is zero), and via firmware write the 5V Vref equivalent on the Vref registers before enabling power switching, and just after Vref is set, enable power switching for it to boot on 5V output.

3

0.5V on 36? is 1.4% error. You have a 1% voltage divider ( I would guess) and the Vref internally is also 1%

So that is within reasonable expectation. There is up to 200nA of fb bias current as well which doesn’t look like it will be important with reasonable divider impedance.

1 Like
4

I can’t reach 36 V, it’s capping on 33.5 V for some reason. Also, it’s ~0.5V on any voltage, not just 36V, say, I’m expecting 5.00V and getting 4.5V instead, which is 10% error. It really is not tracking the voltage well as you can see in this picture:
image

5

What is the voltage right at the FB pin? That should equal the reference voltage, might provide a clue also check the SS voltage make sure it isn’t clamping for some weird reason

1 Like
6

I’ll try checking that