This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS62808: vs TPS62743 vs TPS62840

Part Number: TPS62808
Other Parts Discussed in Thread: TPS62743, , TPS62840

Hi!

Currently, I'm using TPS62743 to convert 5 V to 1.8 V to power a BME680 which has a peak current of 12 mA during the measurement. Because I can only draw 5 mA at maximum from Vin I'm looking for a more efficient buck converter and found TPS62840 which has lower Iq thant the TPS62743. Later on, I found TPS62808 which has the highest efficiency but a higher Iq. Since Vin is not a battery, I do not care much about Iq except on no load in PWM. Am I right that neither of the bucks will enter PWM at my load conditions?

Would you recommend TPS62808 for my design or should I stay with TPS62743?

BTW: I'm using 47µF for Cout and 220 µF for Cin to reduce ripple and have a large buffer for the measurement which can take up to 120 ms. Sometimes, I encounter startup problems. I believe because of the in-rush current caused by the two capacitors. Would you recommand placing and additional inductor in line before Cin?

--
Many thanks,
Mathias

  • Hi Mathias,

    I understand that you want to run the converter in PWM mode even in no load conditions. TPS62743 is not an option because it will enter Power save mode. Both TPS62840 and TPS62808 have Forced-PWM option which will allow the converter to operate in PWM mode at all load conditions. 

    For highest light load efficiency, my recommendation would be to use TPS62840.

    Did you mean that the start-up time is 120ms? This is not usual. Could you please provide more details regarding your measurement setup, conditions and if possible please share the measurements plots as well. It would also be better if you provide the schematic and layout.

    Regards,

    Febin

  • Hi Febin,

    thanks for your feedback. No, I don't want to run the converter in PWM because I consider 12 mA to be light load. I want to be sure that the converter does not switch to PWM and keeps running in Power Save Mode (always PFM?). The datasheet of TPS62840 contains figure 24 which shows the efficiency in forced PWM mode. It is far less efficient at 12 mA compared to the Power Save Mode efficiency (i. e. figure 23). I assume that TPS62743 behaves similar at this light load. Is this correct? Unfortunately, the DS of TPS62743 does not contain a comparable figure to figure 24, neither it does contain detailed IQ_* values as are shown in section 7.5 at page 7 of TPS62840' DS. Furthermore, I can compare IQ_NO_LOAD only between TPS62840 and TPS62808. For TPS62743 it is not provided.

    Considering Power Save Mode efficiency and my light load of 12 mA max, what converter would you recommend?

    Regarding to my second question: No, I don't mean that the start-up time is 120 ms. This is only the time the BME680 measures the air quality. I did not measure my start-up time exactly. But it is defined by the time Cin (= 220 µF, GRM31CR60J227ME11L) and Cout (= 47 µF) of the TPS62743 need to fully charge during start-up. Because the setup can only draw 5 mA from Vin (= 5V) and a completely uncharged capacitor is like a short-circuit, the setup sometimes fails to start and the upstream power supply device enters the overload state. Your application report SLVA670A explains exactly my problem. The soft-start feature of the converters only helps for Cout but not for my large Cin. Indeed, I can use an additional load switch which increases the footprint. But I want to know if an inductor would be sufficient for my setup with the most efficient converter you recommend. But this can be discussed elsewhere.

    --
    Regards,
    Mathias

  • Hi Mathias,

    TPS62840 has a Forced-PWM option and that is the reason to provide the efficiency plot for this mode. But in TPS62743, we have PWM mode for medium and high load conditions and a Power Save Mode at light loads. The transition between PFM (Power Save) and PWM mode is automatic and this cannot be controlled. Hence, we have only one set of efficiency plots for TPS62743. There is no Forced-PWM mode in this device.

    You may find more information in TPS62840 datasheet, as it is one of our newer devices, and we constantly try to add more useful content based on our experiences with our customers. This will be the case for our newer devices but at the moment we cannot extend datasheet parameters for our old devices.

    Referring to efficiency plots for TPS62840 (Figure 22) and TPS62743 (Figure 11), at 12mA, the efficiency for both the devices are approximately the same. You can choose the device based on other application parameters like size, package, max output current, etc.

    Regarding your second question, I would like to take this offline. I have sent you an email.

    I hope you would be fine to close this thread and we can discuss further through email.

    Regards,

    Febin

  • Hi Febin,

    thanks for the detailed answer. Now, I understand the reason for the less detailed or missing plots in the DS of TPS62743. I still do not know if the TPS62743 considers 12 mA as light load. What's the threshold current to switch to PWM?

    What about TPS62808? It has the highest efficiency in the Webbench Power Designer: Vin = 4.5 - 5 V, Vout = 1.8 V, Ioutmax= 0.012 A, Temp = 25° C

    TPS62743 89.9% 41 mm²
    TPS62840 91.0% 21 mm²
    TPS62808 92.6% 18 mm²

    --
    Regards,
    Mathias

  • Hi Mathias,

    Yes, you are right! For the specific conditions you mentioned, Webench results show that TPS62808 has better efficiency. But choosing from the three, depends on the application and other important factors I mentioned earlier.

    Regarding the threshold for Iout:

    For TPS62743 (figure 16), the switching frequency is very low at 12mA and this indicates that it will fall into power save mode.

    Similarly, for TPS62840 (figure 37) & TPS62808 (figure 30), the device could enter power save mode, if you don' t choose the forced PWM option.

    Regards,

    Febin