I've read that speed of rs-232 depends on length of cable. There are many different tables that describe this dependency. But speed although should depend on voltage of Rx and Tx signals. How maximum theoretical speed of rs-232 depends on voltage? How can i calculate this speed at +-5V and +-12V?
Theoretical baud rate means that we know capacity of the cabling C and inductivity L. Transmitter and reciever are ideal. They can give any waveform. Does the voltage affects baud rate in this terms?
The maximum possible speed depends on the capacitance of the cable and the impedance of the source driving said cable. There are also some tricks that can be done at the receiver end to affect the maximum achievable bandwidth, though you are eventually limited by the communication channel, as described in the Shannon-Hartley theorem.
As such, the voltage range the driver can produce is largely irrelevant, assuming that you have a receiver that will work with whatever voltage range your transmitter supports.
You are incorrect in your belief that the voltage plays a role in determining the maximum speed.
Furthermore, the "speed-vs-cable length" tables are what's called a rule of thumb, e.g. a general rule that works in most cases. They generally describe a safe starting point, though in many situations, the system can actually go faster.
Have you ever heard of a term called slew rate. The higher the voltage, the harder it is for electronics to change state ie from a digital '1' to digital '0' in RS232 will be between -8v to -15v for digital '1' and 8v to 15v for digital '0'. At best the electronics will have to change state over 16v. Internal capacitance's of transistors will affect the slew rate as well as the time it takes to get from one state to the other.
Instead of nice clean square wave's, you will get triangular spikes that would get smaller as the speed goes up.
