2
\$\begingroup\$

I'd like to understand the purpose of an inverted logic channel on device-to-device communication that is being done via UART.

For some background - I'm trying to reverse engineer this device-to-device communication with a logic analyzer that I have determined uses UART with the following configuration:

Parity: None
Data Bits: 9
Baud Rate: 38400bps
Stop Bits: 1

On the device, i determined three pins that seemed to be transmitting within an expected logic level. After recording for 20 seconds, I noticed the first and second pin seemed to be linked - a perfect inverse logic level of each other at all times, an example is shown below:

UART Decoding

My first instinct, is that one channel is TX and the other is RX, and when data is being transmitted on TX the RX line is being used to confirm the data is being received as a method of validation. Is this something that is commonly done for UART communication?

The transmission medium is through a flat-untwisted RJ-12 cable, with lengths around 1M-3M. Context is for an audio device connecting to its remote.

\$\endgroup\$
  • 2
    \$\begingroup\$ Might it be a balanced-differential data transmission? Have you considered that? \$\endgroup\$ – Andy aka Apr 21 at 15:46
  • 1
    \$\begingroup\$ It's likely to be differential signalling for the data Tx. If it was a loopback system, where everything transmitted on Tx was echoed back on Rx by the receiver, you'd see a time delay on what you think might be Rx. \$\endgroup\$ – TonyM Apr 21 at 15:50
6
\$\begingroup\$

More than likely it's a balanced-differential data transmission. This offers much superior noise immunity on long lines compared to a single-ended data transmission system. It also doesn't generate the interference that a single-ended data transmission generates because E and H fields cancel. But it requires (for optimum performance), a differential signal that is transmitted simultaneously on two balanced wires such as twisted pair for example: -

enter image description here

Picture from this wiki page.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ I've added some information about the physical medium of transmision. Since RJ-12 is not twisted, is there any advantage to this? \$\endgroup\$ – Naxin Apr 21 at 15:58
  • 3
    \$\begingroup\$ @Naxin It will still reduce the noise emmission even if not twisted. As you say its for audio related equipment, that would be a compelling reason to use differential. \$\endgroup\$ – Tom Carpenter Apr 21 at 16:00
  • 3
    \$\begingroup\$ RJ12 can be twisted - see this \$\endgroup\$ – Andy aka Apr 21 at 16:06
  • \$\begingroup\$ Thanks for the correction. I've updated to specify this is via an untwisted RJ-12 cable. \$\endgroup\$ – Naxin Apr 21 at 16:08
  • \$\begingroup\$ It still doesn't detract from the basic principle of a balanced and differential drive signal. \$\endgroup\$ – Andy aka Apr 21 at 16:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.