Originally created by: phenixrb

Hello to all ,
after configuring the SPI MASTER and having defined the physical addresses of the different MCP23S17 that I use,
I use the following lines to control the outputs:
for exemple:
MCP23S17.byOutputsA.0:=GVL.BP9;
MCP23S17_1.byOutputsA.6:=GVL.BP15;
MCP23S17_2.byOutputsA.3:=GVL.BP12;

it works perfectly.

however, I do not find the variables that will allow me to acquire the state of the inputs.
and the parameters to be defined in the SPI block MCP23S17.
even by setting port B as input as shown in the description, I can not find any variables that change state when I put a + vcc on the MCP23S17 PORT B pin.

Thanks for your help

Hi,
i would give this a try:

BR
Edwin

Originally created by: phenixrb

I have try this but it doesn't work

if I put "2#11111111" in the spi device configuration ,MCP23S17_3.byInputsB.0 is always true

Originally created by: phenixrb

in this configuration:
the input changes state when I apply the gnd to the pin of the MCP23S17
is there a way for the state of the input change when the + vdc is applied to the pin?

Hi,

should work that way,
with current RASBIAN we have seen problems on some boards (for example PIFACE) with SPI maxspeed. (_diMaxSpeed)
Could you please change the speed to 100000 and give this a try.

BR
Edwin

Originally created by: phenixrb

with this configuration:
- PORT B as input,
- no internal pullup resistor enabled.

some inputs are active at the start of the process,

I think that in this configuration, it is necessary to put external resistors of pull down,
to put the inputs pins to the GND; in the absence of control voltage (3.3vdc) on the pin.
currently when I apply a +3.3 vdc on an input pin of port B,
the state of the input changes to 1, but remains at 1, even if there is no more control voltage.
the input returns to 0 only by applying a GND on the input pin.

but except that, it seems to work,
thanks a lot to EDWIN for the help he gives me on the forum and before that in private.

for information: the speed is at 50 000 Hz, I will see if it is possible to increase it once the resistors connected to the GND.