Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Grounding in electrical circuits

  1. Aug 31, 2011 #1
    I feel as if the difference between grounding in a circuit and grounding in mains power is not very clear. So my questions are, what does a ground do in circuits and is it necessary?
  2. jcsd
  3. Aug 31, 2011 #2
  4. Aug 31, 2011 #3
  5. Aug 31, 2011 #4
  6. Aug 31, 2011 #5
    Ground is a reference - in some cases (AC Mains), ground is actually the GROUND. Neutral is often connected to ground outside your house to provide a consistent voltage reference.

    In other cases, it is called ground when it is really just circuit common (or 0V reference). For example, on battery power circuits, the (-) terminal of the batteries which makes up the supply negative is often called ground.

    Thank you Studiot.
  7. Aug 31, 2011 #6
    Thank you Studiot and mdjensen22! But this answer also raised a few more questions from me. What purpose does a circuit common/0V reference provide? Also, if I were to hook up the positive terminal of a battery to a ground (such as one used in AC mains), would there be a current?
  8. Aug 31, 2011 #7
    Why would you expect more than the momentary equalisation current you get whenever you touch two pieces of metal?

    Search the forums, this has been debated as naseam

    Suppose you have a microphone running off a 1.5 volt battery feeding a preamplifier, running off a 15 volt supply, driving a power amplifier running off a 60 volt supply.

    You would designate (in this case) the negative terminal of each supply as the 'common' and call the common value zero.
    This way you could measure voltage from common to any point in any of the parts and get a consistent set of readings.
    This is using the common as a reference.

    Sometimes we designate the high side of the supply as reference.
  9. Aug 31, 2011 #8

    So, what you're saying, is that the ground common is actually just the negative terminal of the battery/power supply?
  10. Aug 31, 2011 #9
    No I didn't mention ground and I said use the negative in this case.
  11. Aug 31, 2011 #10
    I see....sorry, I was misreading your response. You were supplying an answer to the question about what the common is. It is the point where you can measure voltage relative to other components of the circuit.

    Anyways, thank you guys so much!
  12. Aug 31, 2011 #11
    The common may or may not be grounded, depending upon circumstances and the application. Many mike/amp applications are not grounded but have 'earth lifts' incorporated to avoid earth loop hum pickup.

    Say you have some sort of transducer providing a signal to a signal processor which outputs to a data logger. You also decide to monitor the signal at the transducer, the processor and the logger with both an oscilloscope and a voltmeter (3 of each). Sorry I can't draw a diagram tonight, but convince yourself that you can choose one terminal for each of these pieces of equipment and connect them together - this is obviously the common terminal. The other terminal on each instrument will be connected differently as appropriate.

    go well
  13. Aug 31, 2011 #12


    User Avatar
    Science Advisor
    Gold Member

    From a theoretical point of view this has to do with the fact that voltage is relative. Explicitly it means nothing to say that one point in a circuit is at a given voltage. One can only say there is a voltage difference between two points. But by arbitrarily setting one point in the circuit to be "the ground" you are implicitly finding the potential difference between it and any other point when you say that point is at V voltage.

    It is like saying a given particle is at position (x,y,z). This is implicitly saying it is displaced by the amounts x, y, and z from an origin point. Without the origin a coordinate is meaningless. The ground acts as an "origin point" so we can use a language of absolute voltage for what is really a relative difference.

    Now from a practical standpoint in a practical electrical device one can use the chassis or even the earth itself as part of our circuit. All the grounds in a circuit are implicitly connected. (You only need to run one wire to your tail light since it uses the metal structure of the car as the second wire in the circuit... although this is not done as frequently in modern vehicles for various reasons.)
  14. Sep 1, 2011 #13
    Good morning, Jambaugh.

    I think earths or grounds are more complicated than your picture. They certainly do not need to be physically connected.

    How do you explain virtual earths or artificial earths?
    They certainly have current flow into/out of them.

    But they satisfy the definition I provided in the thread I linked to, in the first case forced by an op amp in the second forced by the circuit configuration.

    go well
  15. Sep 1, 2011 #14
    Potential with respect to what?
  16. Sep 1, 2011 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Simplify your thinking. I use the humble flashight, an old one with metal case, to teach the concept.

    "Ground" is an unfortunate name that usually means circuit common.

    To grasp the concept go back to Kirchoff's current law which in its simplest form is just:
    "Electricity always gets back to where it came from".
    Hold an old fashioned D cell flashlight battery in your hand with the positive button poinitng up.
    Every electron that comes out the bottom(negative) will make its way back to the top (positive). That's a physical manifestation of Kirchoff's current law.

    If you stack a second D cell atop the first one. You now have a two cell battery.
    If you place that two cell battery in an old fashioned metal flashlight, you now have an electric circuit that will work just fine with no connection to earth ground.

    If you set the flashlight on the ground it is now connected to earth ground through the dirt or grass or whatever it is laying on. It still works fine. There will be no current to earth ground because the electrons coming out of your battery want only to get back to the other end of that battery.

    You can now pick up the flashlihgt and set it atop a Van deGraff generator and charge it up to a million volts. It'll still work fine for the same reason, electrons coming out one end of the battery only want to get back to its other end. Kirchoff says so.

    If you decide to pick a point on your flashlight and call it "Circuit Common" that is you choice. Circuit Common is simply a convenient point to hook one lead of your voltmeter when doing circuit analysis.
    Defining circuit common is COMPLETELY up to the person doing the circuit analysis.
    A prudent analyst usually bases his choice of a point on convenience.
    In an airplane it's usually the aluminum skin, in a not-fiberglass car it's vehicle body, in a computer it's the PC board track where the power supplies all tie together.
    It's often called "zero volt reference" or "Power Supply Return", but all too often it's mis-named "Ground" and that's unfortunate. It may or may not be tied to earth ground.

    In picking a "Circuit Common" for your two cell flashlight it would seem natural to choose the metal body. That's the easiest place to put your voltmeter lead. And it's where your hand touches the circuit.

    If indeed you chose the metal body for your circuit common and measured voltages,
    you'd get:
    at junction of the two batteries(which is where top of lower cell and bottom of upper cell touch), positive 1.5 volts;
    and at top of upper cell positive 3 volts.

    If you decided to be wierd and chose the junction of the two D-cells as your circuit common, and somehow got a voltmeter lead connected there, you would read:
    to bottom of lower cell (which is metal body of flashlight), negative 1.5 volts
    and to top of upper cell, positive 1.5 volts

    House wiring is like a two cell flashlight.
    The transformer on your power pole has a center tapped secondary.
    The center tap is exactly same as the junction of two D-cells we just discussed, it's where two windings inside the transformer join..
    That center tap is tied solidly to earth ground by the big copper wire running down the side of your power pole. (Dont ever cut it....)
    That center tap is tied solidly to the bare cable coming into your house around which the other two are wrapped. So the bare cable serves double duty - it carries current back to the transformer and it holds up the other two (hot) wires.

    The bare wire is also tied to all the white wires in your house wiring, to the longer slot in all your outlets, to the threaded part of your light sockets, and to all the little third prong grounds in your outlets.

    In house wiring that bare wire is regarded as "Circuit Common" because it is power supply return. In house wiring "Circuit Common" JUST HAPPENS TO BE same as earth ground. It is called "Neutral" and is always a white wire.

    Unfortunately the terms have got confused over the years. Car chassis is called "Ground" even though it's insulated from ground by the tires.

    work that in your head a while- you need to separate the concepts.
    old jim
  17. Sep 1, 2011 #16
    Thanks Jim. Nice write up... One question:

    Isn't the white neutral the power supply return?

    In this case the transformer is the power supply and everything (120V anyway) is "returning" on the center-tapped neutral, correct?

    EDIT: I just realized you're talking about the large bare wire between transformer and service entrance. Not to be confused with the bare (sometimes green) wire in the branch circuits of your house. My apologies...

    I was going to be a wise guy and ask if the L1 and L2 leads of the transformer were 180 degrees out of phase with each other, but I decided against it ;-)
    Last edited: Sep 1, 2011
  18. Sep 1, 2011 #17
    Well goodness, jim, that was a long post.

    But you still haven't explained what is special about an earth. In what way is an earth different from other points in the circuit? If it is not special why do we bother?

    Nor have you addressed my comment on virtual or artificial earths.
  19. Sep 1, 2011 #18
    I think this has already been said, but I'll say it again another way...

    "grounding in a circuit" is basically a common reference point to take a voltage measurement at another point. "the ground" is the same as saying "common reference point". When you analyze that circuit and you are reading a voltage, you are reading that voltage with respect to the common "grounding" point. Convenience.

    Grounding in electrical distribution systems literally means attaching one side of the circuit to the ground... The earth we stand on...

    Now your question as to WHY we attach one side of the distribution network to the earth... that is a very complicated and complex question.

    One very important reason is to dissipate static discharges (lightning), and unintentional high voltage spikes, into the ground and divert this current away from the inside of your house and direct it into the earth.
  20. Sep 1, 2011 #19

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    well i just lost a half hour's typing due to that %#@^%&%(@U(*QRT!!!~~*** login gauntlet.

    Why does it time out while you're composing a reply?

    Why does it dump your text if you mispell your password? I have fat fingers and often hit two keys at onse.

    might try again later tonite if nobody else fields these questions.

    old jim

    ps thanks EB
  21. Sep 1, 2011 #20

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Thanks EB

    Indeed the white neutral is power supply return. Neutrals from all over the house are collected in breaker panel and connected to the centertap of the transformer by the big bare cable, thence to earth via that copper wire down side of power pole. There should be a second connection to earth, via a copper plated steel rod driven into the ground near the meter box.

    As you pointed out the white neutrals are the power supply return. Inside the house, if any current gets into a green safety ground wire it's because some insulation somewhere has ceased to insulate and current is leaking into where it shouldn't.
    Ground Fault Circuit breakers, called GFCI's, sense that condition and trip when around 1/100 of an amp is returning by some path other than the white wire. 1/100 amp is about half the accepted threshold for lethal current, wont harm you but is a real healthy "bite" .
    So- the green safety wire is there to provide a path of least resistance for leaked current . Better it flow through the green wire than you, should you be holding onto the leaky appliance.

    Also you are right on about phase on L1 and L2. Which of course leads to your next point - neutral current is the diference not the sum of load currents on the two transformer windings.... have at it my friend!


    ""....But you still haven't explained what is special about an earth. In what way is an earth different from other points in the circuit? If it is not special why do we bother?
    .....Nor have you addressed my comment on virtual or artificial earths. ""

    Good question, Studiot - i hope i understand it !!

    We should be careful in our choice of names for things, they should lead the unaccustomed mind directly to the correct understanding.
    That's why i avoid the term "ground"
    and much prefer to use "circuit common" or "power supply return" for points in the circuit, and "Earth" for a wire that goes into the earth..

    I've worked on a very well thought out system that had a 1 inch square copper bus bar across bottom of cabinet for circuit common/ power supply return. It was labelled "Zero Volt Bus" which painted a good mental picture of what its function was.
    All internal voltages were measured with respect to that zero volt bus bar.
    Zero volt bus bar was mounted on insulators and connected to earth ground, the earthed metal cabinet, by a 47 ohm 200 watt resistor.
    The name "Zero Volt Bus" and the insulators made it real clear that circuit common was not earth or ground.

    Now back to your question "...
    what is special about an earth..."
    Using my definition of an earth, , ,
    If you are standing barefoot on moist ground and touch a point in a circuit that is earthed, you will not get a shock.
    If instead you touch circuit common you will experience whatever voltage exists between circuit common and earth, which may be quite a shock..
    If circuit common is earthed you will experience zero volts and will not get a shock.

    You experience this in cold dry winter weather when exiting your automobile - that shock you get if your hand is stilll touching metal when your foot first hits the ground . The car's power supply return (chassis) forms one plate of a capacitor, the earth below it is the other plate, and you feel it when that capacitance discharges through you. Similarly you'll notice a little wire sticking up at highway toll booths. Its purpose is to discharge that capacitance so the toll booth attendant doesn't get a shock from every motorist as they hand him their quarters.

    That to me is what's special about an earth. It conducts charge into the earth

    Using instead what i think is your definition of earth, which i would instead name "zero volt reference" for the circuit,
    any "specialness" exists ONLY IN THE MIND OF THE CIRCUIT ANALYST.
    It is a convenient place to put a voltmeter lead and nothing more .
    It's usually a big track on printed circuit boards so as to minimize voltage drop along it.
    A prudent analyst takes all his readings with respect to one point just to keep his thinking straight. Power supply return is usually a good place to use, though some high end audio gear has separate signal and power commons. That helps keep down power supply hum.

    ".. virtual earths.."

    I more often see "virtual ground" but i'm stateside. And i again think it should be called "virtual common" for opamps and "virtual neutral" for power systems to distinguish it from a point that is solidly earthed..
    In both cases your "virtual (insert noun) is a point in the circuit that is pushed by active devices toward same voltage as circuit common or neutral.
    Speaking generally, electronics uses opamps and power uses transformers to do the pushing. They push charge around to achieve that virtual zero potential diference.

    Wow i'm way too wordy. Wish i could draw pictures.

    Hope this makes sense. Reason i spent so much time is - over a lifelong career in industry i'd say around five percent of the engineers i knew had a real clear understanding of grounding.
    if this helps one individual, well, pass it on.

    Best boooks i know of on the subject are:
    an IEEE publication "Green Book", it covers power system grounding (EDIT oops i mean earthing;)

    and "Grounding and Shielding Techniques in Instrumentation" ( cant recall author)
    it covers industrial electronics and basic physics of shielding.

    old jim
    Last edited: Sep 2, 2011
  22. Sep 2, 2011 #21

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    "Why do we bother?"

    Industry learned a hard lesson some decades ago.
    A large factory was built with an electrical system that completely floated - no connection at all to earth.
    On dry days electric motors all over the plant would fail.
    Turned out that windage in the rotating machinery made static electricity , it built up high enough voltage to pierce insulation at the motor windings.

    So power systems always have some connection to earth. It can be a solid one like in house wiring, or a high reststance that just bleeds off static electricity, but it's there.

    In my power plant, on the station 125 volt battery, we used ordinary light bulbs. They provided ample drainage for static electricity and if they flickered it told you something peculiar was going on.

    over and out. Y'all must be tired of me by now, g'night...

    old jim
  23. Sep 2, 2011 #22
    Not at all. It is a privilege to hear things from an experienced person like you.
    Great post.
  24. Sep 2, 2011 #23
    Two solutions:

    1) Compose offline in Word/Wordpad/Whatever and paste

    2) Ctrl+A > Ctrl+C immediately before clicking on the submit button. Then if you loose the connection you can paste it into (1) above or just hold it on the clipboard until you log back in.

    go well
  25. Sep 2, 2011 #24
    This is a good question.

    I will try to expand on my definition with the aid of the attached diagrams.


    Why can't we choose any old point in the circuit, use it as a reference and connect it to earth (rather than at F as shown)?

    Well let us examine some candidates in Fig1.

    Say, Point K

    Well this is where the signal comes into the circuit.
    What would the output of our circuit be if we earthed this point?

    Well what about point J?

    J will have some DC bias from the circuit. What effect will connecting J to earth have?

    OK so B has no connection to the DC biasing of the circuit.

    But a significant current flows through B and the load loudspeaker thus the potential at B with respect to the rest of the circuit (ie any other point in the circuit) varies.

    In fact points K,J, L, A and B are all in the signal path and therefore subject to a varying potential with respect to other points in the circuit.

    So hopefully we can all agree none are good choices either as a reference base or as a point of the circuit to connect to earth.

    OK so what about the negative (zero) power rail?

    Well G, C, D, E and F are all directly connected and in simple theory at the same potential so what if we called this zero reference and worked from there?

    The line G, C, D, E and F form a common connection, which was part of the original question. The line could be extended to accept further connections.

    Reality, however, intervenes as soon as our circuit passes current because significant currents flow in the output circuit around TR1 and the loudspeaker, introducing a significant voltage difference between C and D which varies with time. A similar voltage appears between D and E due to TR2. Furthermore, power currents flowing in the rectifier/filter circuit introduce significant voltages between E and F.
    When I say significant, they are not significant in relation to the output, but if any point between C and F are used as a reference for the input to A1 these voltages will be in series with the input and may be a substantial fraction (or even equal to) the signal voltage at this point.

    Moving on to Fig 2 which shows two pieces of equipment, one mains powered, connected by signal leads.
    If the return lead is not earthed the signal generator chassis may attain significant voltage (up to 100 volts) by electrostatic coupling with respect to the other equipment.
    This effect is shown as the dashed capacitor .
    The chassis of the second equipment is pulled up to this voltage by the return connection between the two pieces of equipment, labelled AB.
    Now suppose we wish to make a circuit change to the circuit under test so we switch off its power supply and apply an earthed soldering iron.
    If the chassis is not earthed, the charge flow to earth from the induced voltage on the chassis can easily be enough to destroy semiconductor components.

    Fig 3 shows the classic ground loop where two mains powered apparatus are earthed at their respective mains plugs which is also connected to the signal return between the equipment.
    Shown dotted is the large area of a ground loop that picks up voltages (especially hum) from any magnetic fields that thread the loop.

    This is all a good basis for discussion as to why we don’t want the earth potential to vary and in relation to what this unwanted variation can arise. Part 2 will include safety earths.

    Attached Files:

  26. Sep 2, 2011 #25
    Thank you Jim and Studinot, you have provided a lot of insightful infomation. Anyways, I have one last question?

    Ok, so lets say I have a small circuit with a 555 timer in it. One of the connections has to go to ground. How would I make a ground? Would any piece of metal do?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook