Differential Nmea0183 outputs and Oscilloscopes

This forum is for discussing hardware (electronics) and software design aspects of multiplexers in general.
Post Reply
Luis Sa
Site Admin
Posts: 751
Joined: Thu May 04, 2017 4:12 am

Differential Nmea0183 outputs and Oscilloscopes

Post by Luis Sa » Wed Dec 04, 2019 7:47 pm


We need sometimes to measure the differential voltage from a Nmea 0183 talker. If the talker has a differential output the labels of the wires are, normally, "TX+" and "TX-". In the Nmea2Wifi multiplexer, for example, I labelled these wires as A3 and B3. It is important to know if the talker is really a differential talker or if it is simply a single ended talker. In a single ended talker we only have a "TX" wire and Ground. I will refer here to the first drawing that I made in another post showing waveforms of serial signals. For the sake of the explanation, I will take the case of the P3 output of the Nmea2Wifi multiplexer as I could had taken P5 port of the Nmea4Wifi multiplexer or any other instrument that has 3 terminals to output a full RS-422 compatible signal (sometimes terminal, GND, is not directly available). Port P3 can be:
  • a differential output using A3 and B3 that should only connect to listeners that have optocoupler inputs or any other kind of galvanic isolation. The idle state is -5 volts.
  • a single ended output using A3 and GND that is mandatory to use when the listener is single ended input (if you use as above you will short circuit B3 to GND!!!). The idle sate is zero volts.
  • a single ended output with inverse polarity using B3 and GND. In this case we have not a "RS-232 or RS-422" type of waveform but, instead we have a "serial TTL" waveform which is useful, for example, to connect to an Arduino board. The idle state is +5 volts.
In the end of this post I will show waveforms of these 3 types of outputs. Now the question - how can we measure these outputs in order to check if they are operating properly? The multimeter used as a voltmeter can only give us limited information. It can, and this is most important, to give us the nature of the idle state. That is the reason why I had emphasized above the value of the idle sate. If the pulses are frequent, you can only get an average and varying voltage depending on how frequent are the pulses. You will notice a higher voltage (in relation to the idle state voltage) in the 1st 2 cases above and a lower voltage in the 3rd case. The ideal instrument is an oscilloscope. I carry one in my boat that costs less than 20 euros. It has the advantage that its power is given by a 9V battery in such a way that I can measure directly the voltage between A3 and B3. In my lab I cannot measure the differential voltage directly as I have a situation like the one shown in this picture:


In reality, my oscilloscope ground terminal (that I call "aux input" in the drawing) is connected to the mains ground of my house installation. This is correct and safe. If I connect a multimeter between the crocodile tip of the oscilloscope probe and the ground contact of the oscilloscope mains plug, it beeps! On the contrary, the negative terminal of my 12V power supply is not "directly" connected to the ground of my house mains wiring. I tried to measure the waveform between A3 and B3 as shown in the bottom right part of the picture and after some minutes I noticed that the MAX485 IC that governs A3 and B3 was very warm. I removed the connection, noticed that the MAX485 was still working, but as precaution it went directly to the waste bin.

So what we can do? The answer is to measure A3 to GND and then B3 to GND. If you have a 2-channel oscilloscope you can use, for example, CH1 (channel 1) to measure A3 to GND and CH2 to measure B3 to GND. Then you set the oscilloscope to show the difference CH1 - CH2. In the following picture I show A3 to GND in yellow (CH1) and B3 to GND in blue (CH2). The zero reference for CH2 is on the bottom of the display so that the 2 waveforms can be observed apart from each other:


In the next and final picture I show the CH1-CH2 waveform in pink. I moved the zero references of CH1 and CH2 to be out of the display. Note the idle sate voltage and compare with the values I wrote in the beginning of this post.


Regards, Luis

Post Reply