RS-232 Communications (software)

Now that we understand the hardware part of the picture, let's dive right into the software part. We'll take a look at each part of the puzzle by defining a few of the common terms. Ever wondered what phrases like 9600-8-N-1 meant? 

Do you use software-handshaking or hardware-handshaking at formal parties for a greeting? If you're not sure, read on!

• ASCII is a human-readable to computer-readable translation code. (i.e. each letter/number is translated to 1's and 0's) It's a 7-bit (a bit is a 1 or a 0) code, so we can translate 128 characters. (2^7 is 128) Character sets that use the 8th bit do exist but they are not true ASCII. Below is an ASCII chart showing its "human-readable" representation. We typically refer to the characters by using hexadecimal terminology. "0" is 30h, "5" is 35h, "E" is 45h, etc. (the "h" simple means hexadecimal)

  	most significant bits
  least sig. bits		0	1	2	3	4	5	6	7
  	0			space	0	@	P	`	p
  	1		XON	!	1	A	Q	a	q
  	2	STX		"	2	B	R	b	r
  	3	ETX	XOFF	#	3	C	S	c	s
  	4			$	4	D	T	d	t
  	5		NAK	%	5	E	U	e	u
  	6	ACK		&	6	F	V	f	v
  	7			'	7	G	W	g	w
  	8			(	8	H	X	h	x
  	9			)	9	I	Y	i	y
  	A	LF		*	:	J	Z	j	z
  	B			+	;	K	[	k	{
  	C			,	<	L	\	l	|
  	D	CR		-	=	M	]	m	}
  	E			.	>	N	^	n	~
  	F			/	?	O	_	o

• start bit- In RS-232 the first thing we send is called a start bit. This start bit ("invented" during WW1 by Kleinschmidt) is a synchronizing bit added just before each character we are sending. This is considered a SPACE or negative voltage or a 0.
• stop bit- The last thing we send is called a stop bit. This stop bit tells us that the last character was just sent. Think of it as an end-of -character bit. This is considered a MARK or positive voltage or a 1. The start and stop bits are commonly called framing bits because they surround the character we are sending.
• parity bit- Since most plcs/external equipment are byte-oriented (8 bits=1byte) it seems natural to handle data as a byte. Although ASCII is a 7-bit code it is rarely transmitted that way. Typically, the 8th bit is used as a parity bit for error checking. This method of error checking gets its name from the math idea of parity. (remember the odd-even property of integers? I didn't think so.) In simple terms, parity means that all characters will either have an odd number of 1's or an even number of 1's. 

  Common forms of parity are None, Even, and Odd. (Mark and Space aren't very common so I won't discuss them). Consider these examples:

  send "E" (45h or 1000101b(inary)) 

  In parity of None, the parity bit is always 0 so we send 10001010.

  In parity of Even we must have an Even number of 1's in our total character so the original character currently has 3 1's (1000101) therefore our parity bit we will add must be a 1. (10001011) Now we have an even number of 1's. 

  In Odd parity we need an odd number of 1's. Since our original character already has an odd number of 1's (3 is an odd number, right?) our parity bit will be a 0. (10001010)

  During transmission, the sender calculates the parity bit and sends it. The receiver calculates parity for the 7-bit character and compares the result to the parity bit received. If the calculated and real parity bits don't match, an error occurred an we act appropriately. 

  It's strange that this parity method is so popular. The reason is because it's only effective half the time. That is, parity checking can only find errors that effect an odd number of bits. If the error affected 2 or 4 or 6 bits the method is useless. Typically, errors are caused by noise which comes in bursts and rarely effects 1 bit. Block redundancy checks are used in other communication methods to prevent this.
• Baud rate- I'll perpetuate the incorrect meaning since it's most commonly used incorrectly. Think of baud rate as referring to the number of bits per second that are being transmitted. So 1200 means 1200 bits per second are being sent and 9600 means 9600 bits are being transmitted every second. Common values (speeds) are 1200, 2400, 4800, 9600, 19200, and 38400.
• RS232 data format- (baud rate-data bits-parity-stop bits) This is the way the data format is typically specified. For example, 9600-8-N-1 means a baud rate of 9600, 8 data bits, parity of None, and 1 stop bit.

The picture below shows how data leaves the serial port for the character "E" (45h 100 0101b) and Even parity.

Another important thing that is sometimes used is called software handshaking (flow control). Like the hardware handshaking we saw in the previous chapter, software handshaking is used to make sure both devices are ready to send/receive data. The most popular "character flow control" is called XON/XOFF. It's very simple to understand. Simply put, the receiver sends the XOFF character when it wants the transmitter to pause sending data. When it's ready to receive data again, it sends the transmitter the XON character. XOFF is sometimes referred to as the holdoff character and XON as the release character.

The last thing we should know about is delimiters. A delimiter is simply added to the end of a message to tell the receiver to process the data it has received. The most common is the CR or the CR and LF pair. The CR (carriage return) is like the old typewriters. (remember them??) When you reached the end of a line while typing, a bell would sound. You would then grab the handle and move the carriage back to the start. In other words, you returned the carriage to the beginning. (This is the same as what a CR delimiter will do if you view it on a computer screen.) The plc/external device receives this and knows to take the data from its buffer. (where the data is stored temporarily before being processed) An LF (line feed) is also sometimes sent with the CR character. If viewed on a computer screen this would look like what happens on the typewriter when the carriage is returned and the page moves down a line so you don't type over what you just typed.

Sometimes an STX and ETX pair is used for transmission/reception as well. STX is "start of text" and ETX is "end of text". The STX is sent before the data and tells the external device that data is coming. After all the data has been sent, an ETX character is sent.

Finally, we might also come across an ACK/NAK pair. This is rarely used but it should be noted as well. Essentially, the transmitter sends its data. If the receiver gets it without error, it sends back an ACK character. If there was an error, the receiver sends back a NAK character and the transmitter resends the data.