TPS610981: TPS610981DS

Hi,

i'm afard the voltage ripple of the LDO is large without a cap. please try to change it to a 1uF capacitor, instead totally no cap

Hi Jasper Li,

Thank you very much for the reply.

If we are adding capacitor of any value, starting from smallest 0.1uF as well than also the spike is observed. Thatswhy we have removed the capacitor to observe the effect. We are having decoupling capacitors on sensors. Will it be OK if we don't mount the capacitor on VSUB output ?

Attached are the images for the spikes that we are observing on CRO when 0.1uF capacitor is mounted. Also, what may be the reason for such high spike. The total current requirement of the board including the sensors(LIS2DH12TR & SI7051) and radio transceiver/controller (STM32WLE5JCI6) is not much. Awaiting for your quick response on this.

Regards,
Kamesh Lakhwani

Hi 

the spike is measured in the input voltage of boost, correct? it is quite strange behavior. could you measure VIN, SW and BOOST output, LDO output at the same image? i like to understand how does this happen.

also please put the waveforms and the schematic in the same report (ppt, excel or work), and indicate clear the meaning of the waveform. this will help us to find the root cause quickly. 

Hi Jasper Li,

Please find the attached document for the schematics, images of waveforms observed on the said pins at the same time. Also attached is the pdf of the schematics file separately. Let me know on it asap. TI_Waveforms_17032021.pdfSCH-STLORASEN-2029-1A-01.pdf

Hi  Kamesh Lakhwani6

Thanks for the report. it is very clear for me to understand the behavior of the system. 

but in the waveform, the input voltage is very clean during the LDO is enabled. there is no voltage spike as you mentioned. do I miss any information?

Hi Jasper Li,

Thank you for the quick reply.

Actually the spike that we are looking into is observed during the current measurement when the voltage source is connected at input of the regulator. Current is measured across the input supply. We have connected 1uF capacitor at VSUB terminal and 10uF at VMAIN terminal of the regulator. Attached is the waveform showing the spike during wake-up of the device. Expecting you please on this asap..

Hi 

sorry, but I still don't understand when does the voltage spike happen. how did you measure the current? current probe? 

what is wake-up of the device? boost IC or other IC in the system.

Hi Jasper Li,

We are using a current shunt measurement technique for measuring the current. The spike that is observed is during this, so we can say that it is the spike in the current consumption when the device/the custom board wake-up.

Wake-up of the device means when the Radio transceiver will pull the modesel signal of the boost converter high to provide supply +3.0V to the sensors. So VSUB(LDO output) will have +3.0V generated on it for few milliseconds which I had explained in the document report that I had sent you previously.

Let me know if you have further query.

Hi Jasper Li,

Is there any update ?

Hi,

could you put the current waveform together with the voltage spike? the input voltage should drop down if there is voltage spike, 

and could you explain why can't we see voltage spike in the waveform of the report?

Hi Jasper, 

Thanks for your continuous support. Actually we have no issue with Voltage Regulator performance. We haven't faced Line Regulation and Load Regulation issue so far. LDO regulator output is delivering stable output in all conditions. We have put 1uF Output Capacitor on LDO output and 10uF Output Capacitor at "Vmain" Power Rail.

Our major concern behind this 25mA Current consumption spike (420uSec duration) during every Turn ON cycle (every 5 second LDO Turn ON) of LDO output, is might affecting battery life. Every 5 second, we are consuming 25mA current with 420uSec. We have checked this with "No Load Condition"

That's why we want to confirm that is this normal expected behavior of LDO Regulator or is there any scope to minimize further current consumption without compromising any electrical performance?

Hi Kamesh,

I think this current is normal to charge the capacitor of the LDO and also power system loading. that is why I think that the current should be smaller with smaller output capacitor.

I'm not sure if your system loading requires higher current at the beginning. please also check.

Hi Jasper Li,

The spike sometimes goes high upto about 60-70mA when the LDO is turned on with 1uF capacitor on it as well. This is affecting our battery life to higher extend. The devices attached to LDO output as can be seen from the schematics are only sensors and pull-up resistors which consume current in microamperes only.

Hi Kemesh,

please try to increase the CIN to 22uF or higher, and change the R20 to 10ohm. then check if the spike can still happen.

Hi Jasper,

I have changed the capacitor value to 22uF and observed that there is some effect of it on the spike. Attached are the images of the waveform for the same. The amplitude of the spike is reduced but the timing of the spike has increased. Is this due to change in RC time constant by changes done in R and C values ?

Also, we are planning to use a reed switch (SW1) as shown in the schematics instead of R20. R20 will be removed. Will this make any impact ? Upto what value can we increase the value of CIN ? Is there any disadvantage of increasing the value of CIN in the circuit ?

Hi Kamesh,

yes. that is because the RC value becomes larger. 

large CIN doesn't not impact the boost device. but i'm not sure it impacts on the battery when the battery and CIN are firstly connected.

Hi Jasper Li,

Not sure but CIN may not put much effect on the battery but R value may put an impact as it will be in series to the battery supply. 10R resistance will provide voltage drop to the battery supply input and also provide power dissipation. 

I had tried by removing the 10 ohm resistance and only keeping 22uF capacitor but alone the capacitor does not suppress or limit the spike/current. We had also tried by replacing the resistor by the inductor which is almost short for battery DC voltage but haven't seen improvement or the impact the resistor puts on the spike. Is there any other way by which we can limit the current/spike ? Please let us know asap as this is becoming a greater challenge for us.

Hi, Kamesh,

what is the maximum output current of this application. it is small, the 10ohm maybe OK. 

also we can try to reduce the capacitor in the VMAIN rail to 4.7uF.

Hi Jasper Li,

The maximum output current can go upto 100mA.

Hi Kemesh,

if the output current is 100mA, the input current is 200mA level. such current should impact the battery lifetime a lot? 

did you try 4.7uF capacitor in VMAIN?

Hi Jasper Li,

We will be using Forte ER14505 battery which is non-rechargeable and do need to provide 5 years of battery life to the client. 

What value of capacitor do we need to put on VSUB if we use 4.7uF capacitor on VMAIN ?

BAT0101_1.pdf

Hi Jasper Li,

We have measured the current using 4.7uF on VMain and 1uF on VSUB. We are observing reduction in current spike at wake-up but there is a higher current spike observed at radio transmit time. The same phenomenon is also occuring when 10ohm resistor is used at input side with CIN=22uF.

Kamesh 

do you have current probe to measure the inductor current and output current? 

please also share the inductor part number. please check the saturation current is good enough.

Hi Jasper Li,

The inductor part number that we are using is LQM2HPN4R7MG0L from Murata.

Hi Kamesh,

the inductance is only 2uF at 400mA  condition. could you try another inductor with higher saturation current. 

please also share me the layout of the circuit.

Hi Jasper Li,

The peak current requirement in our application is 100mA. So considering this current, the inductance will be 4uH at 20degC temperature. Please correct me if my understanding is wrong. Can we use LQM2HPN2R2MJ0L inductor instead ?

Attached are the pdf files for the layout. Please let us know on this. 

individual_layer.pdfsingle.pdf

Regards,
Kamesh Lakhwani

Hi Kemesh,

the input current will be 200mA level if the output current is 100mA. it is boost topology at which the VOUT is higher than VIN. The peak current of the inductor will be higher.  anyway, it is not a solution, but try to  find root cause of the spike.

the layout looks OK.

Hi Jasper Li,

Yes I understand that it is boost converter and the output will higher as compared to input and current will be lower at output than at the input. Can we change the part and use LQM2HPN2R2MJ0L inductor and observe ? 

Thank you very much for the layout review.

Hi Kamesh,

you can try LQM2HPN2R2MJ0L, but I also want to see if larger inductance would help to reduce the current spike.

could you also measure the SW pin voltage during the spike period? do you have current probe to measure the inductor current?

Hi Kamesh,

was the issue solved?

Hi Jasper,

Thank you for following up. The issue is not solved. But we are operating the board on battery and are rarely observing the spike on battery supply.

Hi Kamesh,

OK. that is good. do we still keep to find the root cause? or we can close the post.

Hi Jasper Li,

We are planning to use TPS63900DSKR instead. I hope we do not face the same spike issue by using this regulator. Also we are changing the temperature sensor part whose current consumption is less. Let me know your view on this.