I'm designing a print for RS485 with the LTC2854 transceiver, but I don't know in what exact situation it will be placed in. So I don't know what voltage levels the other devices will communicate on. I will supply the transceiver with 3.3V and communicate on 0V and +3.3V.
Does this mean I can still receive the full range of voltages -7V – +12V of other devices on the bus? Or does this mean that it can only be implemented on 0V – +3.3V level systems?
In my understanding I should be able to detect difference between a zero and one, as long as the difference a-b of the transceiver is at least +/- 200mV. I won't be using biasing and the termination resistor can be turned on or off with LTC2854. Could you explain if my understanding is correct?
Best Answer
-7V to 12V is the tolerable common mode voltage of the differential line, not the signalling voltage on that line (and definitely not the signalling voltage on the microcontroller side). It's not directly related to the system voltage.
You should also be asking if a 3.3V transceiver can interface with a 5V transceiver. (Yes, yes it can).
It's a bit of a mystery to me how they actually do this in hardware since it is operating on 3.3V/5V but is able to tolerate and transmit over a line that has a common mode voltage of -7V to 12V which exceeds 0V to 3.3V/5 (and some transceivers are capable of much higher...-/+25V).
I think the common mode range specification doesn't actually apply to the transmitter, at least not during transmission. Not in the sense that the transmitter takes 0V or 3.3V/5V and then places it on top of the common mode voltage already present on the balanced line. The transmitter might have protective devices to allow it to be in high impedance mode and tolerate those common mode voltages, but I don't think it can actually transmit it's signal on top of a common mode voltage. I think that it just sets the balanced line by connecting it to 0V or 3.3V/5V and that's what the line ends up being. I don't see how it could be done any other way, but then again, those VLSI guys have a lot of control and are able to do some really strange things when directly working with silicon.
I think the common mode range specification only applies to the receiver. I know for a fact that the the receiver steps down the voltage to within the system voltage's range and then uses a differential amplifier to detect the signal. That means a higher common mode voltage you can operate under, the smaller your signal will actually be after processing just before it is sampled since it's stepped down more so you need more sensitive voltage comparators.