Software Serial on ESP8266

Projects Libraries
1

Hi,
I've an issue with the SoftwareSerial library, running on ESP8266 12-E NodeMCU.

I'd like to use pin 4 (D2/GPIO4/SDA) as Rx (input for the micro-controller), and pin 0 (D3/GPIO0/FLASH) as Tx (output).

But it seems there's an issue with this setting.
I initialize the SoftwareSerial library with :
SoftwareSerial m_Software_Serial(4, 0);

After this the Rx pin (4) is forced to +3.3V.
Hopefully this doesn't create a short circuit, because I always insert a resistor (1 K) when I connect an external signal to a µC input pin.

Same result if I try to set this pin as input with :
pinMode(4, INPUT);
I guess this is not necessary, because also done by the SoftwareSerial library, it was just a trial.

If I don't initialize the SoftwareSerial library and just set this pin 4 as an input, I can connect a signal to it (via the 1K res.) and set its voltage to 0/3.3 V without problem.

Thanks for any explanation/suggestion.
Regards.

2

Why not use the second hardware serial port?

3

There is an alternate pin assignment for UART0 (GPIO 13 & 15) and there is UART1 which can only be used fir tx.

For the esp you should use espsoftwareserial

4

I can only comment on this. SoftwareSerial more than likely enables the internal pull-up resistor to prevent floating inputs.

1 Like
5

Hi all,

First, thanks for your suggestions and support.

I've been busy this week, so I just worked on this topic last evening.
I found the issue and solution.

The issue was in fact:

  • a inaccurate measurement tool (USB scope),
  • and a bad functional parameter (a too short receive time-out).

This prevented my sub-system to work.
The sub-system is a RS-485 adapter, which output is connected to a pin of the ESP8266.
This pin is configured as 'Rx' pin for library Software Serial.

When I noticed the issue of no message reception on the ESP8266, I put a probe scope on the Rx pin; but this probe has an internal impedance, and is probably connected to a + something inside the scope; this forced the signal to +5V. Using the probe with setting 'x10' (maybe this increases the probe impedance) allowed to restore a nominal signal on the Rx pin (to be sure I've added a 10 K pulldown).
With a correct signal, it was then easy to find and correct the second issue (reception time-out too low).

Every thing ok now. Sorry to have searched (and made some other searched) in the wrong direction.

Best regards
Claude B.

6

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.