Advanced Users can post their questions and comments here for the not so Newbie crowd.

Moderator: eriksl

User avatar
By btidey
#89257 In my circuit an active low turns on the MOSFET which charges up the capacitor connected to EN in a few uSec. If the signal disappears and the MOSFET turns off then the capacitor will keep EN active for about 5 seconds which is plenty of time for the ESP88266 to fully boot up and activate the hold signal on the GPIO to maintain the EN signal. When the ESP8266 prepares to sleep it removes the hold.

I like using the little Lithium cells as they simplify the powering just using an LDO regulator, they are smaller / lighter than AAA and of course can be recharged. I use an external Lithium charging circuit that plugs into the module direct across th ebattery snd I re-use across all similar modules.
User avatar
By eriksl
#89262 My remote control does not need to be very small, I have cases lying around I am going to use and they will fit one or two AA or AAA cells just fine. They are going to be rechargable, so the max voltage that can be reached is 2.4 V, too little for the ESP to run on. Also the voltage will drop quite rapidly. So that's why I am going to use a boost converter anyway. My guess and hope is that the boost converter does the necessary load regulation, so I don't need to add a lineair voltage regulator (which would almost make the boost converter useless in terms of being to be able to run a long time on the batteries). They come in two flavours: 3.3 V and 5 V. With the latter I'd need a regulator, I am going to try both.

I don't like very much relying on an analog circuit using a capacitor for timing, and I have these 7555s lying around, so why not use them ;). It will ensure a very clean and "digital" signal for the CE pin.

For the inputs I've ordered 8-to-1 AND logic IC's, I am still thinking of the timer's output. It needs to be ORed with a GPIO. I guess I could wire-OR them, but I don't like it very much. I may add an OR IC there.
User avatar
By eriksl
#89265 I have an interesting experimental result (long before I have all the stuff to do the "real thing"). When carefully stepped down from 3.3 V downwards, the ESP2866 stops working at about 1.9 V :shock:. For starting it needs a little more power, for that to happen I need 2.1 V. This is al not quite relevant for my project, as I am using a boost converter (probably), but it's interesting. I may mean only one penlite will suffice...
User avatar
By davydnorris
#89276 The unit is officially able to run down to 2.5V, and I have tried running permanently on 2.8V and find it to be stable. I currently use a buck/boost converter and it works really well - they do the regulation so you don't need another LDO.

I've done a lot of mucking around with power circuits and found some really interesting things:
- an LDO is great for regulating a unit that's powered from a supply as they are very cheap, but is a disaster for battery. The excess voltage is dissipated as heat and it essentially acts as a big resistor, so the higher the voltage difference the higher the current it draws while working. The end result is that your batteries drain super fast until you get close to the LDO voltage. My first battery powered box used 3 D cells in series and a 3.3V LDO, and I chewed through the power in days. Really disappointing.

- the buck/boost converter was so much better. In buck mode it gave me 3.3V and used the excess voltage to make up current so the current draw in steady state on mode was only about 20mA for the whole module.

- in boost mode, however, the boost circuit made up the voltage by using more current, which meant that as the batteries drained the circuit pulled more current, which pulled the battery voltage even lower, which pulled even more current, which... so the effect was to drain the batteries very quickly at the end of their life. So for a more complex and expensive circuit, I really only got a few more hours of power

- the LDO circuit did have one good result. It showed (like you have found) that the ESP kept running way lower than the 3.3V that I thought was needed. My sensors stopped working before the ESP did. This led me to change everything to lower voltage and I found versions of my sensors that ran at 2.8V and 1.8V

I am now in the process of changing to a 2.8V buck only converter and an LDO for the 1.8V sensor circuit because the current draw is tiny. This will let me run nicely on batteries right down to the cut off voltage for rechargables, directly off solar, or off dry dells as well.