Inverted TTL

[alter-jx38]

Hi guys. I was faced with the task of inverting TTL signals to a different logic and can't get how to did it correctly.
I have a dive-computer with communication interface that interprets LOW= +3V and HIGH= 0V. It has an internal pull-up resistor of about 12k, which is connected to the internal supply of 3.0V
Since the device already has a pull-up resistor, a relatively large pull-up resistor in the pc-interface can be used, since this resistor is only needed to keep the data line in a defined state (high) when the dive-computer is not yet connected.

To connect him I suppose to use simple UART-USB controller based on CP2102 with ordinary TTL logic where LOW is 0V and HIGH represented by +5V. The communication is pretty slow with 1200 baud, 8 bits, 1 stop bit and no parity.

I'm guessing there should be a few NPN transistors but can't realise exactly how?

 

[dick_freebird]

I'd recommend something like a 74HCT14, a Schmitt trigger
inverter (hex) with a thrshold closer to the 1.4V TTL norm,
and some hysteresis which will clean up noise / chatter
somewhat. It will work below 5V, better than TTL will
(what with its Vcc-2*Vbe output high, loaded) and just
get a bit slower as supply reduces, but have rail-rail output.

The rest of them you can use for additional inversions.
Or to add more hysteresis by positive feedback and
resistor network.

This is a common part (TI, et al). Nexperia also makes
them for lower voltages (like 2.7-3.6V). Your shown
supply sort of sits in the gap, I'd use the 5V one, keep
the loading under 1-2mA and plan on a bit more prop
delay than the datasheet says.