TPS65023: TPS650231 can be a direct replacement?

Hi Archie,

The difference between these two devices is the default output voltage of the two 200mA-LDOs. The following EVM User's Guide provides the default output voltages for the two devices. You can reference Section 7.5.1.1 of the datasheet for additional information on using the DEFLDO1 and DEFLDO2 pins to change the LDO output voltages. 

https://www.ti.com/lit/ug/slvu394/slvu394.pdf?ts=1633450097814 

Hi Emily,

Thanks for your quick response.

We do have some further clarifications here as per the client's requirements:

• The Primary issue is VDCDC3 being 3.3V when it is programmed to be 1.8V. Does the LDO programming affect the output voltage of VDCDC3?
• According to the datasheet and EVM user manual for both the TPS65023B/TPS650231 DEFDCD1,2,3 are tied to DC gnd. to program the outputs to be 1.2V, 1.8V, 1.8V respectively. VDCDC3 is 3.3V instead of 1.8V. The PCB assembly we are using it on was designed for TPS65023 has not changed and VDCDC3 was always 1.8V. Installing the TPS650231 causes VDCDC3 to become 3.3V. To verify no damage to PCB occurred when the part was changed (I replaced twice to be sure), I put the original TPS65023 back and VDCDC3 returned to 1.8V.
• I would also add that when probing the L3 node, the driver appears to begin oscillating and stops after a few hundred mS and then becomes 3.3VDC which of course goes through the inductor to VDCDC3 resulting in 3.3V instead of 1.8V.
• Here is a closeup of the node that we are having trouble with.

Let me know if you need some further information to support these queries.

Thank you.

Regards,
Archie A.

Hi Archie,

The LDO programming will not affect the output voltage of DCDC3. As you correctly stated, tying the DEFDCDC3 pin to GND should default the output to 1.8V. How does VDCDC3 behave when you tie this pin to VCC instead? Or try to use a external divider? What is the value of the DEFCORE register (06h)?

Both of these devices (TPS65023 and TPS650231) have the same default DEFDCDC1-3 operation and voltage setting, so it's not immediately clear to me why VDCDC3 would be outputting 3.3V.

Best,
Emily

Hi Archie,

Emily is out of office today but will be back tomorrow to follow up on this.

Thanks,

Daniel W

Hi Archie,

Thank you for providing your findings; this is very helpful for others down the line as well. I am not aware of any newer datasheets we have available for this device, but will pass along the information to our team to make sure our datasheets reflect accurate information across device packages. 

Best,
Emily

Hello Emily,

Thanks for your response.

Will be waiting on this.

Thank you.

Regards,
Archie A.

Hi Emily,

Any update on this?

Thank you.

Archie,

To my understanding, the PMIC now works as expected. Is the customer asking for a new datasheet immediately? Unfortunately, that's not something that we would be able to support on a short timeline.

Best,
Emily

Hello Emily,

Thanks for your response.

Yes, the client has now a working PMIC after finding out the discrepancy of TPS650231 D/S regarding the programming of DCDC2 voltage. Can you confirm on this?

I will inform the client to submit his own feedback by clicking the footnotes on a specific page of the datasheet.

Thank you. 

Regards,
Archie A.

Hi Archie,

Due to bandwidth constraints, please expect a response on this by EOB tomorrow. 

Best,
Emily

Hi Archie, 

Just to provide a status update, I am in the midst of trying to obtain an EVM to test in the lab (this takes about 3-5 business to arrive once ordered). Until then, I have reached out to various functional teams to try and get some history on this device. Because this device has been in production for the past 11 years, it's taken quite a bit of time to do some digging. I will try and update as soon as we have some more information, but I wouldn't expect a firm answer until (earliest) next week, 10/27. 

Best,
Emily

Hello Emily,

Thanks for this.

Regards,
Archie A.

Hello Emily,

Good day.

Would like to update the confirmation of this case.

Thank you for your support.

Regards,
Archie A.

Hi Archie,

Feedback I am getting from our design and post RTM teams is that there should be no reason why tying DEFDCDC2 to either VIN or GND should not work to set VDCDC2 to 3.3V or 1.8V. (what is in the datasheet for the QFN package should be accurate)

However, I just got back from testing the TPS650231 EVM in the lab and found that tying DEFDCDC2 to GND yielded VDCDC2 = 3.3V and tying DEFDCDC2 to VIN did not provide stable output voltage at all. These conclusions are conflicting and are requiring me to question this issue given that this device has been in production for quite some time. 

I understand your customer's investigation found the culprit to be VDCDC2 and DEFDCDC2. 

Did your customer ever test the VDCDC2 when DEFDCDC2 was tied to VIN? If they did, and their findings match my own, this may be a recent issue with fabrication that is affecting more than just a couple samples. 

Best,
Emily

Hello Emly,

Thanks for this info.

Will confirm it with the client.

Regards,
Archie A.

Hello Emily,

The customer just responded as follows:

"When using DEFDCDC2 a divider must be used to attain voltages other than 3.3V The description of DEFDCDC2 function is correct for the BGA package and the same formula must be used for the QFN package. Please let me know if you have updated tech data sheet."

For your information and confirmation.

Thank you.

Regards,
Archie A.

Hello Emily,

Sorry for waiting as I've just received feedback from client as follows:

For your information.

Thank you for your support.

Regards,
Archie A.