Use this forum to chat about hardware specific topics for the ESP8266 (peripherals, memory, clocks, JTAG, programming)

User avatar
By torntrousers
#61350 One thing is that with Lipo batteries most of the power is used up by the time the voltage drops below about 3.6v anyway, so although they can be discharged down to 2.something volts there isn't actually that much useful power left doing that.

It might be worth considering just an LDO regulator like the XC6206, which has a dropout of 250mV, connected directly to the LIPO so will work fine down to about 3.5v.

Would be interesting to try both this and the buck/boost approach and compare the cost, weight, and actual running time.
User avatar
By Linton_Samuel_Dawson
#61418
schufti wrote:
Linton_Samuel_Dawson wrote:@piersfinlayson: I found on the interwebz a RT9013 version (the RICHTEK RT9013-33GB) with fixed 3.3V out and 2~5.5V in, which would be perfect for me if only I could find someone selling them :?


RT9013 is a low dropout linear voltage regulator. It is not an up/down boost/bucket converter!
The min. input of 2.2V is valid for the 1.6-1.2V output versions.


Thank you for the heads-up, I might have misread the specs

torntrousers wrote:One thing is that with Lipo batteries most of the power is used up by the time the voltage drops below about 3.6v anyway, so although they can be discharged down to 2.something volts there isn't actually that much useful power left doing that.

It might be worth considering just an LDO regulator like the XC6206, which has a dropout of 250mV, connected directly to the LIPO so will work fine down to about 3.5v.

Would be interesting to try both this and the buck/boost approach and compare the cost, weight, and actual running time.


This is interesting, had no idea about this kind of behaviour. For now I got the boost-buck converters from aliexpress (https://it.aliexpress.com/item/5pcs-min ... 0.0.CSOBwa) and I will see how they work, but if I find these XC6206 i will try them and compare the results ;)
User avatar
By lethe
#61423
Linton_Samuel_Dawson wrote:This is interesting, had no idea about this kind of behaviour. For now I got the boost-buck converters from aliexpress (https://it.aliexpress.com/item/5pcs-min ... 0.0.CSOBwa) and I will see how they work, but if I find these XC6206 i will try them and compare the results ;)

I'm pretty sure those won't work. There's no inductor on these boards, so they can only work as charge pump with tiny capacitors and thus are unlikely to supply enough current for the ESP.

The graph for Vin vs. Vout on page 8 of the XC6206's datasheet, suggests that the regulator can actually work with input voltages below its rated output voltage. In this case you will have an unregulated output that is a couple 100 mV lower than the battery voltage.
Given that the ESP is rated for 3.0-3.6V input and the flash should work in the same range (W25Q32: 2.7-3.6V), the ESP should work down to about 3.2V input. At that point your LiPo is drained almost completely anyway, so going any lower with a buck-boost regulator won't give you much of a benefit.
If your ESP is sleeping most of the time anyway, a buck-boost regulator will probably drain the battery even faster, since they typically have a higher quiescence current.