Hello! I quess I know what SPI is all about, but I may be wrong. Just to have a confirmation: If I use a MCU (like LPC932) which is configured as the SPI master and as slave an ADC (Texas TLC3544) without slave select, the MCU will always be master. Correct? It will generate the SPI clock. The ADC is controlled via certain commands. The commands are sent to the slave by just putting them into the SPDAT SFR. Since commands are 16 bits long, I have to sent two bytes. So far, so good. But if I want to read what the slave is answering to the MCU, I just need to initiate another SPI transfer by sending the (TLC3544 related) command "FIFO read" and in return I get two bytes with the conversion result data put in SPDAT, which I'm reading to some variable for further use. Also correct? Or do I have to put the MCU into slave mode? I don't guess, since the ADC can't be master. Furthermore, will it be sufficient to read out SPDAT, clear the SPI flag and transfer the next byte by putting it into SPDAT and read out again? Looks this way.