LP8545: OCP (overcurrent fault) status bit set as soon as part enabled

Hello Robin,

thank you, that helped. The output voltage was equal to the input voltage less diode forward drop, because my circuit has the external FET, but LP8545 default configuration has EN_EXT_FET=0. I temporarily connected the external FET's source to drain and the booster started up, producing the output voltage of 28.7V expected given the default setting in the voltage control register (A5h) and the values of the feedback resistor divider. However, now I have a couple other problems:

(1) when I turn on the EN_EXT_FET bit in register A2h, the output voltage decreases (!) from 28.7V to 20.3V. It changes when I change the target boost voltage in register A5h, so the regulator is obviously doing something:

A5 VoutF VoutN
--------------------
6B 18.07 24.18
6C 18.62 25.28
6D 19.21 26.43
6E 19.78 27.56
6F 20.36 28.72 // default setting
70 20.97 29.89
71 21.56 31.04
72 22.13 32.18
..
78 25.63 34.64
..

7F 29.11 34.64

I checked the voltage on the FB pin and it really becomes smaller when I enable external FET gate drive: 25V when EN_EXT_FET=0 but 17V when EN_EXT_FET=1 at default VBOOST setting. In addition, when the external FET is disabled, the output voltage maxes out at 34.6V.

(2) since the part can't start up at default settings unless the external FET is shorted out, and EN_EXT_FET cannot be set unless the part has started up, I tried to write the settings registers to NVM by following the procedure in the datasheet. However, it seems to have no effect, and when I poll the EE_READY bit after writing EE_INIT=1 and EE_PROG=1, it never goes to 0. Is it possible to program the settings registers to NVM on the bench? I really don't want to redesign the backlight converter. I checked out the parts recommended in this older question on this forum, but none of them can go up to 55V.

My schematics is below:

Hi Anton,

Do you adopt the adaptive mode?

BR,

Robin

Hello Robin,

for the present I disable adaptive mode by writing B7h (EN_EXT_FET=1, EN_ADAPT=0) to register A2h. However, EN_ADAPT does not make a difference: the output voltage is 20.38V whether I write B7h or BFh.

Hi Anton,

Could you present the all register value you configured?  Let me have a check.

For NVM programming, it has been practiced by bench test.

BR,

Robin

Hello Robin,

sure, here are the contents of all I2C registers when LP8545 is outputting 20.3V. I also dumped the 9Xh block which seems to contain diagnostic data, hope it helps.

   | 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
---+------------------------------------------------
00 | 00 05 00 FC 00 65 C0
90 | 32 1E 26 3B 5A 8C 74 94 FD 00 00 00 00 00 00 7C
A0 | 7F B5 B7 7B 28 CF 64 2D

Yuting Shao said:

For NVM programming, it has been practiced by bench test.

I see, but as I wrote above I tried the procedure in the datasheet (write 5 to 01h to turn on, wait for output voltage to settle, write settings to A0..A7h, write 4 to 72h, write 2 to 72h, wait 200ms, write 0 to 72h) multiple times and it had no effect: after a reset the AXh registers are at their default values. What's more, I never observe EE_READY (bit 7 in 72h) be zero during these attempts. What am I doing wrong?

Hello Robin,

The output voltage keeps stuck at 20.3V also when I connect the LED strings. What else I could measure to diagnose this problem?

Hi,

The voltage set by A5h is the initial voltage, and the device will output a proper voltage according to the loading.

As for NVM, could you try again with other devices?

BR,

Robin