TPS55288: Not working -

Hi Marlon,

1. May I know how did you configure the I2C registers? To ensure the TPS55288 work properly, please set MODE register (06h) OE bit (7 bit) to 1, then TPS55288 starts switching and output default 5V.

2. If you need to change the output voltage, you need to configure the REF register (00h, 01h).

Let me know if the problem still exists.

Hi Zack,

So I have not been able to configure the i2C register. For some reason as soon as I power the TPS55288 chip the SCL goes to LOW. I used a digital analyzer to confirm that.

I have removed any other IC from the bus to confirm that the error is not coming from a different IC, but until now the clock is always 0 as long there is power on the IC.

Do you have any suggestions on what could be the problem?

Could I share my schematics with you? I want to confirm if there is anything wrong with my design.

Regards,

Marlon

Hi Marlon,

Yes, you can share the schematic here.

Only after the I2C register is configured correctly, the output voltage could rise to right value.

Hi Zack,

Thank you very much for reviewing the desing.

Regards

PCBA-PowerControllerV2.7.PDFSRC-PowerControllerV2.7.PDF

Hello Marlon,

Looks like you have to find the root cause of I2C communication failure first. 

In your shared schematic, the I2C communication pull-up voltage 3.3V comes from U3. U3 comes from TPS55288 output. Is it correct? However, you can't let TPS55288 output 4V first without 3.3V pull-up voltage.

Could you try using a external 3.3V voltage?

I'll share you the comments on schematic and layout later. May I know the maximum output current in your application? You didn't mention that in previous post.

Hi Zack,

I have removed all ICs (U3, U4, U2) connected to the I2C bus. I'm providing power through the TP2 and I am connecting an external board to the I2C bus using the 2 ping connector (SDA-PC, SCL-PC) in-between X1 and X2. I still see the SCL going low as soon as power is provided on TP2 and as soon as I disconnect to power SCL goes up. I am using a data analyzer to check the behaviour of  SDA and SCL. So it seems the device is actively pulling down SCL.

Is there any reason why that could be happening?

Hello Marlon,

That's really strange. Do you have any waveform showing once TP2 is powered, the SCL is pulled to low?

Hi Zack,

Yes, it is very weird.

I have attached different pictures

As you can see when U5 is completely OFF I am able to send and I see the NACK

But as soon as I power U5, the clock goes low and I cannot transmit since my board since the clock down and consider that the bus is busy

This is a zoom-in of the previous image

Another Zoom in

In the last one you can see how U5 is power then I disconnect power and then reconnect the power again. You can see how the SCL goes up when the power is off

Regards,

Marlon

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/U5PoweringUp_2D00_2_2D00_ZoomIN.JPG

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/U5PoweringUp_2D00_1.JPG

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/U5Off.JPG

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/U5_2D00_On_2D00_OFF_2D00_On.JPG

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/U5PoweringUp_2D00_3_2D00_ZoomIN.JPG

Hi Marlon,

1. Firstly I would like to confirm with you, you are using TPS55288 (I2C version), not using TPS552882 (NON I2C version). Please check the IC top mark?

2. When you powered U5, I see you use an external VCC source. Have you checked VCC voltage on your board? Is the external 5V correct?

3. How many failure boards do you have now? Could you try replicate it on another board? If you have only one board, could you try replacing the IC.

And the layout could be furthered improved. Let's debug the I2C issue first then we can discuss how to improve the layout.

Hi Zack,

1) I double-checked the IC top mark and yes I did a mistake when ordering the component and I ordered TPS552882 (NON-I2C version). I will order the new component this week and replace it. I think with that hopefully it will work fine.

2) Regarding the external VCC, yes the external VCC is 5V and I do see it on U5

3) Yes all boards have the same problem. I will replace the component in the 3 boards that I have.

I will highly appreciate it if could mention how the layout could be improved. Once I am available to test the current board with the new ICs I would like to modify my design including your comments so I can fabricate a final version.

Regards,

Marlon

Hi Marlon,

1. May I know the maximum load current of your application? I didn't see it in schematic.

2. The I2C communication pull-up voltage 3.3V comes from U3. So only TPS55288 has output voltage (I2C register is configured), U3 has correct input voltage. There is a power up sequence problem. 

3. For TPS55288 layout, please follow the layout guideline. Let me know if you have anything unclear.

Hi Zack,

1. The max current is 2A
2. Yes, there is a power-up sequence. I thought at the beginning that TPS55288 will give a default output voltage of 5V so that will power the 3.3V regulator. Most probably I will use TPS55282 but for now, I have ordered TPS55288 to test the design and I have removed the 3.3V regulator so I do not have the power-up sequence problem
3. Do you have any specific comments? I have followed the guideline as much as possible. My loop 1 is quite small, the input capacitors could be a bit closer but I think they are not that far. My SW1 and SW2 are a bit big than recommend but since I do not have any signal close by and I have the ground plane under I thought it will help thermally. Do you think that should be changed? The VCC capacitor is close to the IC and with vias for current to return. I could not place vias next pin 24. I will appreciate it if you have a specific comment.

Thank you very much for all your help and support.

Regards,

Marlon

Hi Zack,

Thank you so much for your comments and your help with my design and issues. I really appreciate it.

I will include them in the new version. Hopefully, everything will work fine in the new version.

Thanks,

Marlon