Replacement needed for ESP8266

Go To Last Post
15 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Dear freaks

recently i was using ESP8266(ESP-01) with mega8A to log temperature to Thingspeak.Setup was running successfully.But after a long run of 4 days,my esp8266 failed for the first time.I flashed different new version of firmware 2-3 times,but soon after 2/3 hours of run,every time the ESP8266 failed.Now it's returning garbage characters on UART randomly all by itself(without giving any command to it).
I think these modules are pretty cheap and they have no guarantee for long run.Although i ordered another ESP8266(ESP-12) for second testing for long run.

 

My question is:Is it worthy to stick with the ESP8266 modules and use another version for stability or should I consider another alternative?? Xbee-wifi looks promising but will Xbee-wifi fulfill my needs regarding this particular project aspect??Please answer and also mention if any other alternative available....

Tags:

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The ESP8266 is a 3 V device, were you running your Mega8 at 3V or at 5V?

 

Can you post a schematic diagram of your actual project, and a photo of it?

 

A silicon chip is a silicon chip.

Excluding harsh environments, I would expect the 8266 to perform well and last just as long as any other similar device, IF the design is good.

You need a good power supply, noise / voltage spikes cold cause a problem.

You need appropriate by-pass caps on the design, both for the uC and for the 8266.

You need to protect it from static electricity.

It has several power levels, was it running hot?

Did you try running it (the new one) at a lower power setting?

 

Although it is certainly possible to have an "early failure" of the device, it is far more likely to fail based upon the design or something the User did to the chip.

 

JC

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My circuit is powered from a 5v supply.Of course the ESP8266 is fed using a LM1117-3.3v regulator.Also there is a level shifter between the data lines of mega8 and ESP.I am attaching a Diagram in PDF format for your convenience.The 5v is fed from a LM350 based linear PSU.Also i tried to fed from a 5v/1A SMPS.

 

what do you mean by

 It has several power levels, was it running hot?

The ESP was getting a little bit hot while running.

Did you try running it (the new one) at a lower power setting?

No,my new module didn't arrive still now.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Your problem is really interesting to me since I just got an ESP8266 recently and I'm hoping to use it for long-term weather monitoring. I've yet to program it or anything, but your attached schematic looks reasonable to me (at least comparing it against the other schematics I've already looked at for programming one).

 

When you say the ESP was getting a little hot while running, was it so hot that you couldn't handle it with your hands? Or was it more just warm to the touch?

 

How was the design you attached all wired up? Did you have the regulator, etc on a breadboard, a PCB, dead-bug, ...?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The ESP was little bit hot than warm  but not at all very hot.My design is installed on a breadboard.The LM1117-3.3v with two 10uF caps are installed in a small homemade PCB.

Last Edited: Thu. Mar 24, 2016 - 07:49 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Are ALL of the micro's Vcc and Ground pins connected?

 

Do you have two caps connected to the 16 MHz crystal?

(perhaps 18 pF +/- a little bit)

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You seem to be using the minimum value for the output capacitor (10uF) on the  LM1117 3.3V regulator.

 

Here is what the datasheet says:

8.2.2.1.3 Output Capacitor
The output capacitor is critical in maintaining regulator stability, and must meet the required conditions for both
minimum amount of capacitance and equivalent series resistance (ESR). The minimum output capacitance
required by the LM1117 is 10 μF, if a tantalum capacitor is used. Any increase of the output capacitance will
merely improve the loop stability and transient response. The ESR of the output capacitor should range between
0.3 Ω to 22 Ω
. In the case of the adjustable regulator, when the CADJ is used, a larger output capacitance (22-μF
tantalum) is required.

Is the 10uF output cap a tantalum?

What is the ESR of the cap?

(Note: most of the example schematics in the datasheet show a 100uF capacitor.)

 

If the regulator becomes unstable and oscillates, it would cause all kinds of problems for the ESP8266.

 

Also, have you measured the output current of  the LM1117 and calculated the power dissipation?  What is the package of the LM1117?  Is it getting hot?

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes,all the 5v pins are connected.Yes,i used 22pf caps.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No,I didn't use tantalum caps cause i don't had them.Instead i used electrolytic cap.The ESR is very hard to tell as i have no datasheets regarding them.No,the LM1117 was not getting hot.I used the SMD version installed on a small home made pcb along with two caps.

Last Edited: Fri. Mar 25, 2016 - 08:14 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The ESP8266 will have quite large consumption (compared to "average ICs") in sudden bursts because of radio transmission. Thus you should probably aim for the upper end of the LM1117 recommended capacitance, not the lower one. Specially if you cannot meet the tantalum recommendation (in which case put several caps in parallel to lower the ESR). It wouldn't be surprising that this was the cause of your ESP8266 early failure.

ɴᴇᴛɪᴢᴇᴎ

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have first hand experience with getting the supply to the ESP8266 wrong - it is unreliable as hell. Give it a decent supply and it is much more reliable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I see....

Ok,i will order some tantalum caps as per the discussion.What is the recommended value for Tantalums to power up the ESP8266 ????

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yesterday i received my ESP8266(ESP-12F) module with a breakout board.Also i was able to manage the "Tantalum" caps for LM1117-3.3.Now i replaced the 10uF/25v electrolytic caps with 10uF/25v tantalum caps(at both input and output).Wiring descriptions of the ESP8266(ESP-12F) module is as follows.

 

Vcc - 3.3v

GND - GND

CH_PD - pulled up(3.3v) via 10k resistor (Previously soldered in the breakout board)

GPIO0 - pulled up(3.3v) via 10k resistor

GPIO2 - pulled up(3.3v) via 10k resistor

GPIO15 - pulled down(GND) via 10k (Previously soldered in the breakout board)

 

Now the ESP is up and running and transmitting data to Thingspeak correctly.You can see it here in my channel https://thingspeak.com/channels/92431/

I gill get back to you guys if anything wrong happens.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok freaks...i faced it again...the incident is as follows

 

I powered the module from my LM350 based linear power supply and worked well for almost 2 days.The LM350 power supply output was set to 5v.After that, i needed the power supply for another experiment,so i switched it off,then disconnected it and thought to power the ESP-MEGA8 circuit via a 5v HTC original mobile charger.It is rated as 5v/1A output.

 

Also i connected a 1000uF/25v electrolytic cap at the 5v output line of the HTC charger and powered the circuit from the charger.As soon i powered the circuit,the ESP was deliberately throwing Blue lights after ever 5-6 seconds.So in order to know,whats happening,i connected the FTDI module between the Tx pin and GND pin of the ESP8266 and logged the data using Realterm.I am attching the logfile.

 

I then switched it off and disconnected the HTC charger.This time i Pulled up the RST pin via a 10k resistor.Also i added a 470uF/25v cap between VCC & GND of the ESP module and powered it back from the LM350 based linear PSU.This time it also misbehaved and was taking random resets.I then again kill the power and restart it after some time.Then everything came back to normal and it was uploading data to the Thingspeak.com normally.

after 10-12 hrs of operation,i saw that the module was updating data to thingspeak discontinously.I then killed the power and restarted it after 5-6 hours.But this time,it is taking resets like before.My mega8 is giving commands to it as it should be.Whats the problem behind this failure???Did i bricked the module by feeding the power from the mobile charger???I.Please help.......

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Also in addition,i tried to flash a new firmware exactly described as per this page.But also after firmware upgrade,it is taking reset like previously.Also in addition,it is sometimes showing "wifi connected" without issuing any command to it.At that time,my home wi-fi network is being jammed/inaccessible.After powering off the module,the network is restoring.cryingcryingcrying