TPS7A8300: Why did TPS7A8300 output 2V when it work at 3.3V adjustable mode and EN is pull down?

Hey Jason,

Can you please provide a scope shot with VIN, VEN and VOUT during this behavior?

Thanks,

Vahnroy

Hi, Vahnroy:

Thanks!

Please see below scope shot with VIN 5V5, VEN and VOUT 3.3V. You can see that VOUT source 2V before EN is driven high. 

I also get a strange slot marked orange which VIN is behind VOUT. It violate causality

Is there other source inserting into 3V3 rail before VIN 5V5 start up?

Best regards!

Jason 

Hey Shen,

It appears another source is pulling up the output. Is there anyway you can isolate the LDO from the rest of the system?

Thanks,

Vahnroy

Hi, Vahnroy:

There are only two source rail ahead of VOUT 3.3V, those are MCU_3.3V and VIN 5.5V( this 5.5V is VIN shown above scope shot). The sequence between MCU_3.3V and VOUT 3.3V is  shown in below figure. Then I disable MCU_3.3V through pull down EN pin of buck which source MCU_3.3V. So, the MCU will not be power on. EN PIN of TPS7A8300  always be pull down. But, I get fixed 2V by using multimeter. 

why?

Best regards!

Jason

Hey Jason,

Is this behavior being seen across multiple devices?

If not can you try an ABA swap. Swapping the device for a new device and seeing if the problem persists.

Thanks,

Vahnroy

Hi, Vahnroy:

I try another board, also has the same issue. This issue only happen on VOUT 3.3V, but VOUT 5V which another TPS7A8300 on the same board could ramp up from 0V to 5V normally after EN is driven high. So I ask for you that hope you can try the same setup on demo board. Please note that must keep EN low, and run at 3.3V adjustable mode, then to see if the 2V  appear on the VOUT of TPS7A8300 possibly. 

Thanks in advance!

Best regards!

Jason

Hey Shen,

Since this seems like a system level problem testing on our EVM won't be representative of your system. Due to the nature of LDO's the output cannot be high before the input.

In order to confirm that this is not the LDO can you desolder the device and see if the voltage still appears?

Thanks,

Vahnroy

Hi, Vahnroy:

I found root cause: a reference source(2.5V) insert or crosstalk into the VOUT 3.3V power plane in the ADC chip. When I cut off the 2.5V reference, 2V on the VOUT 3.3V disappear. So, there is sequence requirement between the VOUT 3.3V and  2.5V reference actually, but the datasheet don't mention this requirement. Ok, I will power VOUT 3.3V firstly then 2.5V reference.

But I have another question as below figure shown：

when I remove bead FB27 to cut off the input power to TPS7A8300, 2V on the VOUT 3.3V from reference source(2.5V) will flow to input of LDO reversely?

I can measure this fixed 1.6V on the input of LDO. This kind of reverse flow come through the body diode of NMOS in the LDO?

Thanks in advance！

Best regards!

Jason