# esp8266 power supply problem

It's the fourth esp8266-12e that destroyed with my AMS1117-3.3 voltage regulator.

I'm powering esp8266 with AMS1117-3.3 module and this simple circuit:

Input voltage: 5.4 V from adapter
Input cap: 1000 uF
Output cap: 470 uF


But the esp8266 will burn after some on-off switches.

Am I doing anything wrong?

I'm testing it on a breadboard and the ams1117-3.3 is a copy model. I've measured the output voltage of it with my scope and I saw some ripples with a maximum of 4.6 V and about 20 ns duration at the moment I switch input voltage ON and OFF. Can these short ripples burn the esp8266? How can I protect it?

I saw these ripples also on Lf33 and 7805 (original models)

Ripples on output voltage of AMS1117-3.3 triggering when switching input voltage ON/Off:

• Your large capacitors are effectively not present at high frequencies, add physically small ceramic ones placed close to the regulator in parallel. Consider if your upstream power source is doing something nasty - some cheap bench supplies have horrible transients. Ultimately the cause of your failures may well be something else. – Chris Stratton Jun 16 '20 at 14:30
• Is your scope on 1V/div? I have my ESP on AMS1117-3.3 and is working about a year without any problems, albeit it is constantly on... I would check mine with a scope when powering on. – smajli Jun 16 '20 at 15:01
• @mohamadali Please show how the esp8266 is wired, how is your ground? – Voltage Spike Jun 16 '20 at 15:27
• @smajli yes it's on 1V/div , maybe my problem is that my ams is a copy, if you have an original please check if maximum voltage cross over 4 V at switching on/off , I get this voltage everytime switching input power repeatedly on/off – MohamadAli Zeraatkar Jun 16 '20 at 17:01

This answer is to show my ESP8266 power-on plot. It is provided to compare against originator's trace, so it is possible it is not an ultimate solution.
My setup consist of 5V 2A power supply, connected to AMS1117-3 with some additional capacitors.
Capacitors: on input of AMS is an tantalum 10uF and output of AMS - 220uF electrolytic and close to the ESP8266 Vcc and GND board terminals is located a 100nF capacitor.

Scope trace:

Scope is on 500mV/div, so overshoot is around 250mV.

I would suggest to use capacitors to suppress initial voltage overshoot. Looking at the originator's scope trace it would be beneficial to use small capacitor (100nF) as close as possible to the ESP module.
Also, as is mentioned, using non-original AMS1117 may bring additional problems.

EDIT:

Some wording has been changed to make my post clear.

• How to place esp in power on trace only? Is there any standard way for it ? – MohamadAli Zeraatkar Jun 24 '20 at 17:17
• Try to add 100nF capacitor close to the terminals of ESP and check the power-on voltage with scope. – smajli Jun 30 '20 at 11:43