TPS63031: The component causes incorrect results in the system

It looks like you have a voltage drop along the cable. If a load is present then the voltage at the end of the cable is expected to vary according to the cable length. This reference design basically describes this problem and proposes a solution to address this : 

What is your end application? could you share the system block diagram by private message?

Hi Sabrina,

I'll describe again the problem the Indicator provides different results in different sizes of cable length because the use of TPS63031 .

Please note that when they use LP38690 (instead of TPS63031) in their old design with the same setup, the problem was not appears.

4.5 Vbat --> Ferrite Beads (BLA2ABD471SN4D) --> TPS63031 --> 3.3V goes through Wheatstone bridge --> then goes through a cable (which can be length of 0.2-20 meters long) to Indicator.

Please find below the current block diagram:

All capacitors are ceramic with the recommended parameters as shown in the datasheet,

Please can you advise back asap what could case this problem and how to overcome it?

Thanks you

Kind Regards,

Shai Berman

1. Could you capture the waveform:
- at the output of the TPS63031 before the bridge?
- At the end of the cable for different length? for ex. 0.2, 10 and 20meters

2. What is sm?
3. What measurement is the indicator providing? (voltage, current...etc)
4. The usage of PS/SYNC pin is described in section 8.4.3 p. 9 of the TPS63030 DS

The LP38690 is a linear regulator, not a boost converter. Unlike the TPS63030, the LP38690 has SNS that allows remote sensing and eliminates the voltage drop across the cable (description p3 of the LP38690 datasheet).

Hi Sabrina,

After a conversation with the customer, the problem is resolved (partially),

As I mentioned at the beginning the power flow goes as the following:

4.5 Vbat --> Ferrite Beads (BLA2ABD471SN4D) --> TPS63031 --> 3.3V goes through Wheatstone bridge --> then goes through a cable (which can be length of 0.2-20 meters long) to Indicator.

It turns out that ferrite (BLA2ABD471SN4D) created a problem when my client removed the ferrite from the prototype the Indicator did not show any wrong results

But my client needs this kind of filter before the bridge,

Please let me know your professional opinion, why adding ferrite (BLA2ABD471SN4D) to the design can causes such a problem and how to overcome it?

Thanks in advance

BR

Shai Berman

When they used the LP38690 in the previous design what output voltage did they use? Did they use the filter in this design as well?

Hi Juergen,

Yes they used LP38690 and the Ferrite in their previous design, because they don't had the need to have buck/boost but now they use smaller battery pack so the need to have at least Vin(min)=2.5V and have the ability to fix the output voltage at 3.3v .

there is any new regarding this phenomenon?

Thanks in advance,
Shai Berman

I assume the voltage drop across this filter has changed. Can you measure the supply voltage at the Wheatstone bridge using an oscilloscope with and without the filter?

Hi Juergen,

At the moment, the client is unable to get accurate results from the Wheatstone bridge, because the customer has already done production of 1K.
I am trying to find him a solution that will bypass the problem.
Please advise if I'll suggest that adding LDO as the previous used LP38690 (Vout=3.3v) after the TPS63031, we can advise him to change the output voltage to be 3.6V and the LDO to be 3.3V could this will solve the problem?

Thanks in advance,
Kind regards,

Shai Berman

I was asking for more data to better understand the problem. If the problem is related to noise generated by the DCDC converter adding an LDO will make a difference. If the problem is related to limitation of the available power the LDO most likely will make it worse.