1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

DC voltage on an AC line

Discussion in 'Electronic Projects Design/Ideas/Reviews' started by embedded_sys, Sep 28, 2017.

  1. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    I have a main line stretching about 1200 feet. I would like know if the line has been cut when the main line doesn't have voltage. I'm currently thinking of putting a 100K resistor between the hot wire and the ground wire at the end of the line and applying a 3.3VDC or 12VDC voltage with a resistor to the hot line at the beginning of the circuit. This would create a simple voltage divider that I could measure. If it is a 3.3VDC, I'd know the line was cut. If it is lower, the set voltage of my voltage divider, I'd know the line is cut.

    The biggest problem is knowing how to put 3.3VDC on a line that sometimes has 120VAC. I would never try when 120VAC is on the wire, but my worry is someone will accidentally apply 120VAC to this line when I am applying 3.3VDC. Obviously this would destroy my DC circuit. I've thought about using an isolated dc-dc convertor to apply the voltage just for that line to reduce the risk of the rest of my circuit being damaged. Is there a good protection method I can use to make this safe?

    Screen Shot 2017-09-28 at 4.03.09 PM.png
     
  2. AnalogKid

    AnalogKid Well-Known Member

    Joined:
    Sep 22, 2011
    Messages:
    1,139
    Likes:
    161
    Location:
    Ohio - GMT-5
    With a 100K buildout resistance, the reverse current from 120 Vac would be 1.2 mA rms, or +/-1.7 mA peak-to-peak. Other *lethal* lack-of-isolation issues notwithstanding, that should not be a problem for whatever small power supply is supplying the DC.

    Note: *lethal*

    ak
     
  3. Pommie

    Pommie Well-Known Member Most Helpful Member

    Joined:
    Mar 18, 2005
    Messages:
    10,162
    Likes:
    340
    Location:
    Brisbane Australia
    Just use a zener to ensure the DC circuit never sees above 3.3V.

    Mike.
     
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. dr pepper

    dr pepper Well-Known Member Most Helpful Member

    Joined:
    Oct 6, 2008
    Messages:
    4,763
    Likes:
    259
    Location:
    North west UK

    Dc isnt good for things like transformers and through a resistor would probably be decked to ground by some connected load triggering your loss circuit.
    A better way would be to use a 555 and a 567, the 555 generating an audio tone applied to the power line via a suitable cap, and the '567 tone decoder detecting said tone at 'tother end, this would work with power applied or not, and with a load or not, if your working with processors you could do this in code.
    Obviously I need to point out mains and power lines are dangerous dont mess with them.
     
  6. schmitt trigger

    schmitt trigger Active Member

    Joined:
    May 24, 2009
    Messages:
    914
    Likes:
    85
    What dr pepper said..... motors and transformers present a very low impedance at DC and would completely shunt it out.
     
  7. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    This sounds like a good idea. If I put a 555 timer at the beginning, how could I use the 567 decoder to read it also at the beginning? You post suggests a circuit on both sides, but my application really needs to be all at the beginning.

    Also, how would I connect a 555 and a 567 to a line that has 120vac main voltage? Could a frequency generated by the 555 and not drop the voltage enough to be detected by the 567 decoder after a 3000ft round trip on 12 gauge wire?
     
  8. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    If the line doesn't have 120Vac on it, what is its equivalent circuit looking back toward the source end from the load end? Neutral still connected? Line side opened by a switch/relay/breaker?

    While you are checking the integrity of the line, what is connected across the line at the load end?
     
  9. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    Ground and Neutral will still be connected, but the line wire goes through a switch so it is unconnected. The end has a 120VAC relay.
     
  10. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    So you are trying to test the integrity of a line that looks like this:
    line.png

    If the relay is in circuit while measuring, I would need to know R1, R2, L1 and R3.
    What is the measurement system referenced to?
     
  11. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    Your diagram is correct.

    Using a wire resistance calculator, I find that the resistance of line R1 and R2 to be 1.906 ohms (length=1200, gauge=12). I'm not sure of the resistance (R3) and inductance (L1) of the relay, but here is the datasheet.
     
  12. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    Can you measure the DC resistance of the contactor coil using an Ohmmeter?

    There has to be a separate data sheet for the contactor coil?
     
  13. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    It doesn't read on my ohmmeter.
     
  14. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    You are not measuring between the correct terminals.
     
  15. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    Tried a different relay. 282 ohms.
     
  16. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    Here is an approach. Relay can be left connected at the load end. I'm guessing at its inductance. You really need to measure it. Add the network at the source end. The oscillator circuit can be left connected.

    Here is V(out) with switch open: Note the frequency.
    line1.png

    Here is Vout with the switch closed: Note the frequency.
    line2.png
     
  17. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    Wow this looks great. I need to test the line integrity from the other side.
    Screen Shot 2017-09-29 at 1.04.27 PM.png
     
  18. dr pepper

    dr pepper Well-Known Member Most Helpful Member

    Joined:
    Oct 6, 2008
    Messages:
    4,763
    Likes:
    259
    Location:
    North west UK
    Oops I must not have read your post proper.
    You probably could get a tone to make the 3000ft round trip, it has the same chance as the Dc would, but both at the same end is tricky, you could do it but you'd need a big choke in one of the lines so probably isnt practical.
     
  19. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    I misunderstood. Ignore the previous post #15.
    Twisted pair has a capacitance of ~20pF/ft
    1200 ft of line would have a capacitance of ~20*1200 = 24000pF = 24nF
    If the line is intact, that is shunted at the far end by 282Ω in series with whatever the solenoid inductance is.
    If the line is open, it will have a bit of residual capacitance but the solenoid inductance and resistance would not be there.
    Your idea of injecting some DC bias through a 50K to 100K resistor will work (use high voltage resistors for R1 and R2, check breakdown voltage).

    Here is my proposal:

    line.png

    There are three periods shown, each lasts 50ms in the sim.
    1. normal 120Vac applied to distal relay.
    2. normal line, 120Vac is off.
    3. Faulted line, 120Vac is off.

    Note V(out) at each period. V(out) is sampled by the AD in your microprocessor. You will have to discriminate as to what is happening on the line in your processor.
     
    Last edited: Oct 3, 2017
  20. embedded_sys

    embedded_sys New Member

    Joined:
    Sep 8, 2017
    Messages:
    22
    Likes:
    1
    This looks really good. What kind of voltage source would I use for V1, V3, V4? I typically use something like this. How can I safely connect it?
     
  21. MikeMl

    MikeMl Well-Known Member Most Helpful Member

    Joined:
    Mar 17, 2009
    Messages:
    11,139
    Likes:
    566
    Location:
    AZ 86334
    V1 is the Vdd for your microprocessor. It could be 3.3V or 5V...

    V3 and V4 are just for simulation; they control when the simulated switches open/close. Think of "switch" and "fault" as being time-delay relays.

    V3 opens the 120Vac switch S1 at 50ms in the simulation.

    V4 opens switch S2 to create the simulated fault on the line at 100ms in the simulation.

    With the circuit I showed, the microprocessor can tell if the ac is on/off. It can tell if the line is continuous to the relay or not. It cannot tell if the line is shorted; the ac breaker will blow in that case.

    You are obviously confused by a simulation schematic and how to convert that to a circuit schematic. C1, R4, R5 are not "real" components; they represent the line resistance and capacitance. Just like R6 and L2 represent the relay.
     
    Last edited: Oct 3, 2017

Share This Page