TPS63710: Unstable Operation of TPS63710

Hi Shyam,

Thanks for your enough waveforms. It looks the device restarts.

There're three possible reason causing restart. Input voltage UVLO, output current limit and over temperature.

1. Input voltage UVLO is not possible since the SW voltage is ~12V during on-time.

2. Output current limit is possible since we can see the input voltage is decreasing before restart.

To confirm whether it's due to output current limit, you can try to probe the inductor current.

You can catch it by disconnect one side of the inductor and use a wire in series to catch the current using a current probe.

3. Over temperature(OT) is possible since we can see the keeping time is shorter and shorter and that's may be related to the temperature.

If you can see the keeping -5V time is long just after first powering up and becomes shorter and shorter then, the OT is a possible reason.

It would be better if you have a camera to catch the case temperature.

Waiting for your response.

Athos

I have seen a blunder happened in my assembly, i have the output 47uF cap changed to tantalum at last minute due to unavailablity of ceramic

the tantulum cap is mounted reverse, postive terminal of tantalum is connected to negative ouput of LDO,due to this it heated so much and spread the heat all over board, i observed this quickly with a thermal camera and reversed it, with which the issue resolved

below are the waveform of output

-5V LDO output :

-5.35V DCDC Output

-2.5V LDO Ouput

-2.85V DCDC Output

5V LDO Output

5.35 DCDC Output

+2.5V LDO ouput

+2.85 DCDC outuput

1. I see the DCDC which is generating +2.85V is being excessively getting heated up, is there a way to reduce this heating ?

2. Also can you see the output switching noise levels if are they as per expectations in datasheets ? i feel i can come down much less of switching noise, placing few extra caps at dcdc output will it reduce the noise ? suggest few methods.

Unfortunately this noise is appearing in my opamp circuits as i am operating at very low noise levels

can you explain or throw some literature about adding snubber and 100nF caps at inputs ?

This would help me to tackle the noise better

For example the yellow one is the first TIA  of gain 10K and pink is Inverting amplifier of gain 10, which has this spurious pulses, i am pretty sure the noises are from power supply and no photo diode is connected, i wish to bring down these noises at opamp to much low.

Hi Shyam,

The two ways I shew you are to reduce the switching noise but here you mentioned is the noise of output voltage. I'm not sure if they are related. Sometimes yes, but I want to first confirm that.

Usually output noise our customer tested is from the probe but not actual noise. You can try to measure the ripple using a pig tail to reduce the space injection. Also you can follow this AN. https://www.ti.com/lit/ta/ssztb25/ssztb25.pdf 

The switching noise means the ringing during FET on/off. You can refer to below AN about RC snubber.

https://www.ti.com/lit/an/slyt465/slyt465.pdf 

Also the noise is highly related to the layout, you can send me the layout for review, too.

https://www.ti.com/lit/ab/snva803/snva803.pdf 

Regarding the noise issue, you can firstly use a pig tail to measure the OPA voltage. If the noise still exists, you can disable our DC/DC and change it  to a DC source to see if the noise is gone.

please can you drop a personal msg with mail id, i shall mail  you the layout file to your mail.

Infact i was measuring noise at output of dcdc and opamp with nearest ground using a pigtail probe only.

athos-zhao@ti.com