UART communication between 2.8V and 1.8V

[AAnkur]

I want to establish a UART communication between two modules. The Rx/Tx Voltage levels of the modules are 2.8V and 1.8V. Also, I have only 3.3V as the supply voltage on my board.
Kindly guide me as to how can I perform this level shifting.

Thanks

 

[Nigel Goodwin]

A simple resistive attenuator to drop the 2.8V to 1.8V is all that's required, the other way (1.8 to 2.8) should probably be fine, as it should fall within the permitted range.

 

[languer]

Something like the following is probably the simplest (yet effective) solution. I have used it some man times now; it's like one of my glue logic go-to items: https://www.sparkfun.com/products/12009 - you can find it in Amazon if you want it in larger quantities.

If you want to go a bit more elegant the you can also try the following: https://www.sparkfun.com/products/11955

 

[Nigel Goodwin]

Something like the following is probably the simplest (yet effective) solution.

Hardly 'simpliest' when I've just given a two resistor solution, which is what that module actually uses one way - the other way uses an FET and resistors, but isn't normally needed, hence my two resistor solution.

 

[languer]

Hardly 'simpliest' when I've just given a two resistor solution, which is what that module actually uses one way - the other way uses an FET and resistors, but isn't normally needed, hence my two resistor solution.

While this may be the simplest it is subject to threshold levels, voltage limits (both low and high), and communication speed. It probably does work in many cases, but there are also other cases where having a high impedance series resistor may not work (due to speed), or having a low impedance shunt resistor may also not work (due to loading constraints). So yes, it is simplest, but it has the potential of needing it to be tailored to the application. And on a high speed communication bus it is very likely to fail. But like everything else, you get what you pay for. But you are correct in terms of simplicity, for low speed simple bi-directional comms the resistor divider solution is truly hard to beat.

 

[Nigel Goodwin]

While this may be the simplest it is subject to threshold levels, voltage limits (both low and high), and communication speed. It probably does work in many cases, but there are also other cases where having a high impedance series resistor may not work (due to speed), or having a low impedance shunt resistor may also not work (due to loading constraints). So yes, it is simplest, but it has the potential of needing it to be tailored to the application. And on a high speed communication bus it is very likely to fail. But like everything else, you get what you pay for. But you are correct in terms of simplicity, for low speed simple bi-directional comms the resistor divider solution is truly hard to beat.

As I already mentioned, the board you suggested does exactly the same thing, using a simple resistive divider to drop the voltage, it merely adds an FET to increase the voltage the other way, which almost certainly isn't required (the lower voltage should fall WELL within the specified input limits of the higher one).

Also, the simple FET solution on those boards has been widely criticised, as unsuitable for high speed conversion - although it is quite clever I've always thought, and I have a number of them in my spares draws for any suitable occasion