From my experience, if you were to step into a college electronics lab session and ask a student “When would you set an oscilloscope channel to either AC or DC coupling” chances are you’ll get something along the line, “AC coupling is for AC signals and DC coupling is for DC signals”. While there is some truth to that, that’s not quite the full picture. Up until recently that would have been my response till I decided to take it upon myself to figure out why some past engineers thought that this functionality was worth implementing in oscilloscopes.
A good place to start when trying to understand the use cases of AC and DC coupling is by understanding the word coupling. One of the very first results you’ll get when you google coupling is it’s Wikipedia page which defines coupling as:
A coupling is a device used to connect two shafts together at their ends for the purpose of transmitting power.
While this is a description suited to mechanical engineering, the useful thing to get here is that coupling it basically connecting two things so that something can be transmitted from thing A to thing B.
One of the most common uses of coupling in electronics is the coupling capacitors which feature prominently when learning about transistor circuits. So what do coupling capacitors couple? They couple AC signals from one system to another while blocking DC signals from being transmitted from the source system to the destination system.
So in an oscilloscope, when should you use DC coupling, and when should you use AC coupling?
DC coupling passes all signals through without filtering anything. Thus both DC and AC signals will be displayed by your scope when probing a circuit.
AC coupling, similar to coupling capacitors, will only display AC signals on your scope filtering any DC signals.
To appreciate which coupling to use at any point when testing your designs, let’s look at a couple of simple use cases.
Let case one be the case of a DC signal 5 V that has a small signal AC riding on it.
From the above display you can’t really tell there’s an AC signal, for that you’d have to zoom in as below
We get to see the 100 mV AC signal that’s riding the DC signal but unfortunately this is as close as you can get on DC coupling, 200 mV/Div. Furthermore to get to this point, you’d have to consistently adjust the vertical position of the channel trace to get it to display on screen.
Thus, this is a case that is suited for AC coupling, you’ve acknowledged the fact that you have a DC signal, but you want to specifically investigate the AC signal. The AC coupled signal is shown below:
With AC coupling it is possible to zoom in up to 20 mV and even more to investigate the signal. This technique can be used when dealing with power supplies that convert AC voltage to DC and you want to see how clean of a DC output you are getting.
So we’ve seen when AC coupling would come in handy, the next question would be, why would it be beneficial to use DC coupling with AC signals?
Let’s look at the simple op amp inverting amplifier powered by 5 V and -5 V having a gain of Av = -2 and an input of 1 Vpp at 100 Hz:
The above output shows an input signal of 1 Vpp 100 Hz amplified to about 2 Vpp, which is what is expected. However, let’s say you set the scope to AC coupling since after all, you are investigating AC signals, but then you get the following output:
You output is clipping, but your input is only 1 Vpp which shouldn’t be enough to get the op amp output to saturate. The output is expected to be 2 Vpp and saturation is at 5 V or -5 V. So what’s going on? The answer to this question is clear when you set the channels to DC coupling as shown below:
From the output above, it is clear to see that the problem is that the input has a DC offset that is causing the actual maximum voltage to be 2.2 V and when the maximum is amplifier by the inverting amp the output is -4.4 V at which point the op amp is getting saturated. This is a characteristic that is totally masked when operating with AC coupling. From this you can trace where the DC offset is coming from by either fixing the problem component e.g. a poorly designed power supply, or using coupling capacitors.
Thus from this discussion, I hope one is able to appreciate AC and DC coupling when it comes probing circuits. A general rule of thumb I use it that always operate in DC coupling and only activate AC coupling when you specifically want to investigate particular AC characteristics and you aren’t interested in the DC signals either because you’ve already acknowledged them or they are inconsequential.