Simulating CAN

[SparkFly]

Hi,

I'm trying to simulate a CAN transmitter in LTSpice, based on this diagram:



The circuit works as expected for a CAN '1' (both outputs across 120R resistor are 2.5V, inputs are 1, 0), but when I try to simulate a can '0', the difference across the 120R is far too small. It should be around 3.5V and 1.5V. My circuit is as below.



Can anyone offer any advice as to where I am going wrong? When I increase the resistance of the 120R resistor, I get a bigger differential voltage, but my understanding is that CAN should be terminated by 120R.

Many thanks in advance.

[langwadt]

M5 is upside down, and pick some actual parts for M5/M6 the generic PMOS/NMOS doesn't work

[SparkFly]

Thanks Langwadt, I now have it working as expected.

[T3sl4co1l]

The symbol is upside down to convention (the gate usually comes out from the source side), but as the substrate is connected correctly, and the default model isn't asymmetrical (D and S are interchangeable), it doesn't actually matter.

Also, this is the first time I've seen those outrageously wrong MOSFET symbols in a slightly-less-wrong configuration:



Luckily you put together the circuit correctly, despite that.

If people would realize that the original, and actually correct and useful, MOSFET symbol actually means something, by itself, this wouldn't be confusing.

Tim

[SparkFly]

Thanks T3sl4co1l.

The wide variety of symbols are quite confusing... :/

Thanks for your input.