OK, and I AM an engineer. AND the way to sense a "Short Circuit", or a SORT OF short circuit requires fast thinking, unless you do it off-line, first, as some have suggested. That works very well but it is not what you are seeking.
The scheme to use is like the "short circuit protected " VS drives and soft starters. They measure the current as the voltage rises for each cycle, and if the current is excessive for that voltage they switch off the current. THAT is why it needs to be fast. Fortunately, for protection, it does not need to be so very high resolution. And the good news is that there are schemes for doing the comparison to limits in a real-time analog arrangement with the digital half just setting the limits for each cycle. (I DO listen when I get an explanation of how the system works.) The details belong to the companies and as they will be happy to sell a package that works correctly the first time and every time after, the savings you get from spending zero development budget and zero prototype work make a person look quite good. And when clients come back for more it is a good feeling.

 

The elephant in the room is WHY it's necessary to solve this fiendishly difficult AC and DC at the same time problem.

If you divulged the actual nature of the problem, at the high level, I am sure the people here could point you in the direction of a simple and highly effective solution.

 

YES!

Okay, this is awesome, thank you for that great response; that's exactly what I'm looking for. So is what you are proposing, similar to what I designed on the schematic I uploaded? You need an inductor to filter out the AC from the DC?

Do you only need one, or two, and if one, does it need to be on the high side or low side?

With regards to your statement of connecting an XFMR to a DC power source, that's exactly why this works so well in my situation. If I had to test a device with AC and connect it to DC, bad things could happen. However I am unaware of any condition that would cause a reactive impedance to be less than the linear. So if I did connect a transformer winding upto my tester, it would give me a fault indication, so I could examine the item, see that is a transformer (a coil of wire), and test it with my multi meter or physical examination to see if it is intact or in a failure state. That's the beauty of this.

Now, ideally I'm thinking probably 1 Ohm would be sufficient; although I was planning on using a varistor or potentially a potentiometer so that I can calibrate the signal. My thoughts on that were that the current would flow through the Receptacle and device, through a calibrated resistor and then through the potentiometer into the optocoupler. So the led in the optocoupler would only illuminate (to signal the MCU) of there was sufficiently low enough impedance through the load device. Turn up the resistance and no fault until there is a short. Turn down the resistance and you have an early warning that there might be a higher current flow through the AC circuit.

You say that the inductor would have to be huge in your example. How do you calculate that? I presume you're just calculating the coil impedance based at 120V 60hz. So I'm confused why the DC circuit would affect that variable, would not the ac line voltage and frequency dictate that on its own? Are you merely sizing that to have a sufficiently high impedance compared to the load device?

If that's the case then this would be astronomically difficult. Say for example I unplug the load and the circuit is still energized. Even if my coil is 10MOhm the Receptacle now has an impedance of infinity... wait, although that would break that C2 circuit as well, so that's not a problem. But if you plug in a load with high impedance, that could be an issue right?

Wait, let me think about this... I might not need this functionality. So, I only need to test the device when it's not powered on I presume. Any device not powered on will either be an open circuit, extremely low idle power draw, or a short. The first two are of no concern...

Okay let me think about this for a couple days. I could potentially have this circuit only need to operate with no ac power, then I could use a simple DPDT relay.

Yah,
Thank you!

 

I went back to post #1 and I read " 'm thinking this is possible with some sort of arrangement of inductors and capacitors (to filter each signal out), as well as optocouplers to isolate the controls from mains power. " telling me that there are two different signals to see, without any description.
BUT if those signals are the voltage magnitude and the current magnitude, that is what the smart motor starter looks at.There is no need to apply any DC voltage to test an AC device.

 

I think I should be fine with a double throw relay solution.

Thank you all for your contributions.

 

OK, here is the testing scheme modified to function a bit like one I did long ago. The very large difference is that the test for some form of "short circuit" is done with either DC or AC PRIOR to the decision to connect mains power. So it can work manually unless it is done as an automated scheme. The test is like this: A known low voltage is applied to the load through a known resistance, and the test control logic measures that voltage and decides if it will work to apply the mains power or not. That whole test may take up to five seconds, although two seconds would be quite long enough. If the voltage developed across the load is great enough, close to the sourced voltage, then the decision is that it is OK to apply mains power after disconnecting the check circuit.

 

If you have a DMM, you do not need what you are asking for and also it will damage the meter or your requested circuit to measure resistance on live AC grid. So the best solution is to learn to use the DMM to measure voltage and if 0 then measure resistance.

 

OK, either the TS used one of the ideas or they gave up. The really fast cheap trick I saw in a hardware store MANY years ago used a series light bulb in the mains power feed. If the bulb lit full brightness it saw a short. To apply power for diagnostics a 100 watt bulb was used. No, it would not pass an OSHA inspection, this was probably 30 years before anybody imagined such a thing as OSHA.