-->
Page 2 of 3

Re: ESP-12S Inductive Reset

PostPosted: Thu Oct 31, 2019 3:13 am
by Bonzo
I will often have the antenna hanging off the board.


Usefull tip

Re: ESP-12S Inductive Reset

PostPosted: Thu Oct 31, 2019 3:26 am
by schufti
there has been an issue (don't remeber if here or github) where (unfortunate shielding and) too high Tx power (by default) led to unpredictable behaviour when wifi communication took place.

Re: ESP-12S Inductive Reset

PostPosted: Thu Oct 31, 2019 11:56 am
by cjohnson355
rudy wrote:This is not related to your resetting problem. In the future, don't put any metal near the antenna. No traces or external wires, and no copper planes. The copper you have under the antenna will degrade the the sensitivity of the signals. Transmit levels will be worse as well as received signals. It is recommended that the antenna is placed on an edge of the board, I will often have the antenna hanging off the board.

see page 13 of https://www.espressif.com/sites/default ... nes_en.pdf


I noticed this in the documentation after I'd already finished/ordered the design. Thanks for the input though, the WiFi strength is pretty weak and I'm only about 30 feet away from the router. Putting the antenna off the board should help with this.

Here's my game plan:
Re-layout the board with the WiFi module off the edge of the board.
Change transistor driven relay to using an optocoupler

I'm hoping this fixes my reset issue, has anyone else experience the reset issue I am having?

I tried the snubber circuit with 120 ohm / 0.1 uF but it still reset pretty much every time the heater shut off.

I can't think of another way to fix this issue other than remotely located the WiFi module off the main board. I've seen other designs that use an ESP to control high current/high inductance loads, and they all use optocouplers. That's got to be the answer, but only time will tell after the design is finished.

Re: ESP-12S Inductive Reset

PostPosted: Fri Nov 01, 2019 5:52 pm
by cjohnson355
schufti wrote:there has been an issue (don't remeber if here or github) where (unfortunate shielding and) too high Tx power (by default) led to unpredictable behaviour when wifi communication took place.


Was there a fix to the issue?

I am using the serial port, but normally nothing is connected. Just serial prints when the server.on callback happens.

Think disabling the serial port would fix this issue on top of converting to optoisolator driven relay?