Unsigned long stepEnd = score.time + score.step The pin start at the last value set by master (HIGH or LOW)ĭigitalWrite(score.pin, !digitalRead(score.pin)) Int currentStep // for playing the array. Unsigned long time // start time of current step. The relevant methods involved are: int PortExtender::reset() ,Ĭonst short nPins = sizeof(pins) / sizeof(pins) Slave, however, receive the command and return the right response, as checked thru Serial Monitor connected to the slave.Įverything seems to work fine, but answer from slave looks like it changed 'in the wire'. Response is some random value that change from run to run. This is the result from Master 2 (nodeMCU): In fact, I can run a complete test suit with no problems, testing the button, controlling the LED and reading the LDR. This is the correct result from Master 1 (Arduino Pro Mini):

The slave answers with its own address (42), just to check that the command works. The test just sent a soft-reset to the slave. PortExtender port = PortExtender(SLAVE_ADDRESS) This is the Master test program: #include

I compiled and ran the same Master program in the two master, while the slave remains the same. The inactive one is powered off and disconected from the two others. I have two master, but only one is used at a time.

I'm writing PortExtender, a library to convert an Arduino Pro Mini in a port expander and I ran into what seems to be a compatibility problem between Arduino and nodeMCU: I got different answers depending on who is the master (Arduino or nodeMCU).Īt the center, the Slave Arduino, with buttons, leds, and ldr for testing pin access.

Arduino zero wire library