What Serial Link to use

Try asking the question the other way around. Give us your communications requirements then people can suggest the most appropriate protocol.

which

but

communications

systems.

Thanks for the reply Richard, Truly speaking i do not have any requirements in hand for which i a trying to differentiate these protocol. My intention is to understand thes protocols and compare them so that i get a reasoning for having so man protocols at first place. Why not just one or two serial protocols. My Question might sound little dumb but as i said i am Novice :-)

Please help.. Thanks

but

It is not always simple. You should consider plain SPI/I2C for board leve connections, very short distance, very susceptible to electrical noise.

RS232 is good for point to point connections at several meters distanc (slow speed). RS422 is good for point to point connections at a much longer distanc and/or high speed (the distance is a function of speed). RS485 is just like RS422 but it allows multi points connections.

RS422/RS485 are very reliable because they use differential lines ove twisted pairs that have high immunity to the electrical noise.

Of course on top of that there are a lot of other things to consider, i is just a starting point. Wikipedia has articles about the above systems you could start from there then read some data sheets.

regards, Giovanni

SPI and I2C is for use on the PCB between the MCU and peripheral devices. SPI is faster and easier to get going than I2C, but uses more pins than I2C. SPI is also a master-slave protcol. i.e. the master initiates all transactions. I2C uses only 2 pins even for multiple peripherals. A peripheral can initiate a transfer.

RS-485 and RS432 are electrical specifications. It says nothing about the protocol. One tend to use RS-485 compatible buffers when one wants to communicate over fairly long distances (Up to 3000m). Some of the newer RS485 buffers can handle data rates of 50 MBaud, although at a shorter distance. RS-485 buffers also support a multi-drop connection.

The above is only a very simplistic discription.

Regards Anton Erasmus

Something I would add is the the asynchronous protocols work well between devices that are functionally computers having their own clock sources. On the other hand, the synchronous protocols work very well when a computer/microcontroller master wants to talk to simple state- machine-like devices that may not have their own clock - regardless if these are dedicated ICs or in-house FPGA functions.

This isn't an absolute - you can do synchronous to a slave that has it's own clock, it's just more complex. You'll commonly see this to peripheral chips that may have a data clock independent of the control clock. And you also see slave-mode synchronous interfaces on computing devices that need fairly high data rates - CODEC interfaces on DSP-capable microcontrollers for example.

but

Thanks all for your inputs, i am trying to understand on these lines an reading some other sources also.

Thanks Again, Jalon