ESP8266 Community Forum

- Mon May 18, 2015 4:31 pm #17781 I'm using a very similar part from TI, TPS63011, but what I'm trying to do is use the select function of this part to switch between 2.8V and 3.3V. I'll be sharing my complete design soon but I've run into issues with this power supply that has me 2nd guess if running the part lower than 3.3V is feasible.

I've switched to running my from Agilent power supply and have confirmed that coming from a cold start the ESP-12 module(s) just don't like to run from lower than 3.3V. It's definitely more forgiving with soft resets but I still get a "failure" at least 1 out of ten times.

Not sure if it matters (or maybe it really does) but I'm running NodeMCU (NodeMCU 0.9.5 build 20150318 powered by Lua 5.1.4). What I've also found is that when the module doesn't boot properly it pulls over 300mA and many times I thought I was seeing ASCII junk in the terminal but it turns out it was actually error messages that were coming in at 115200 baud rate (I've been communicating at 9600 normally). It looks like the chip is getting a watch dog timer error. Here's the messages that I see:

Code: Select allets Jan  8 2013,rst cause:4, boot mode:(3,1)
wdt reset
load 0x40100000, len 28740, room 16
tail 4
chksum 0xcd
load 0x3ffe8000, len 2888, room 4
tail 4
chksum 0xbc
load 0x3ffe8b50, len 15252, room 4
tail 0
chksum 0x4f
csum 0x4f
rì


- Mon May 18, 2015 5:12 pm #17785 I seen similar errors when I tried the ESP-01 at 2.1V and I think this was because the memory used on my module would work down to 2.2V but would give errors below this. I don't think it was the ESP chip failing, but I could be wrong.

Q: Are you switching between 3.3 and 2.8V to interface some 2.8V logic? If so, I've used a simple solution in the past by just adding a diode in series with the 3.3V line for all 2.8V logic. 2.8V logic will work very well at 2.7 or 2.6V so using the voltage drop in the diode with low speed data/signals.

- Mon May 18, 2015 5:34 pm #17791 No, I just wanted to run the module at 2.8V and then switch to 3.3V before WiFi transmission (assuming higher Vcc provides better signal strength).

I've worked with many different parts before I haven't run into something that appears to be so picky about the actual Vcc level. I'm not talking about fluctuations, I'm providing about as stable a power rail as you can provide and it just doesn't seem to like much of anything besides 3.3V. Running a 3.3V part from 2.2, heck 2.5V, is way outside most specs so I'm surprised by our contrasting results.

Hopefully others will chime in and provide their findings. Are you running/testing with Lua or the stock AT firmware?

- Mon May 18, 2015 5:46 pm #17792 I'm running it on the stock AT Firmware.

I'll run a few more test on it tomorrow and let you know I find.