Capacitance coupling
- Thread starter domnic
- Start date
- Status
- Location
- Illinois
- Occupation
- retired electrician
You can't really get rid of it. Anytime you run two conductors in close proximity you create it. It is a very rare case where it causes any type of problems, other than confusing troubleshooters with high impedance meters.
ptonsparky
Tom
- Occupation
- EC - retired
Eliminate it? Put a load on the wires of sufficient size to short the capacitor.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
151011-1652 EDT
Two conductive surfaces separated by an insulator form a capacitor (called a condenser in times past). The magnitude of the capacitance is a function of the surface area of the conductors, the conductor spacing, and the dielectric constant of the insulator. The shape of the conductors and the resulting electric field map with dielectric constant are what really defines the capactance, but it is simpler to say the spacing.
If the conductors have their shape retained, and the spacing is increased, then as spacing is increased the capacitance between them reduces. Thus, an inverse relationship between capacitance and spacing.
If you understand this, then I or someone will answer how to reduce the coupling.
You can experiment with capactive coupling as follows:
1. Cut two 30" lengths of #12 Romex. Either use a cable without an EGC wire, or just let the EGC float.
2. Strip the needed ends.
3. Attach one 30" length to a plug so as to connect the white wire to 120 neutral, and black to 120 hot.
4. Using a 10 megohm input impedance AC digital voltmeter measure the voltages between the hot and neutral wires relative to a neutral found at the socket that the 30" wire is plugged into.
5. Disconnect the neutral wire at the plug which leaves the previously neutral wire floating, and make the same voltage measurements.
6.
Measurement 4 produces about 120 V and zero volts.
Measurement 5 produces 120 V and some small voltage, like in the range around 30 V. Your small voltage may be different than mine because of different cable construction.
7. Add 30" to the first section of Romex. The capacitively coupled voltage will approximately double. As you add more cable length the voltage won't increase as rapidly.
.
Two conductive surfaces separated by an insulator form a capacitor (called a condenser in times past). The magnitude of the capacitance is a function of the surface area of the conductors, the conductor spacing, and the dielectric constant of the insulator. The shape of the conductors and the resulting electric field map with dielectric constant are what really defines the capactance, but it is simpler to say the spacing.
If the conductors have their shape retained, and the spacing is increased, then as spacing is increased the capacitance between them reduces. Thus, an inverse relationship between capacitance and spacing.
If you understand this, then I or someone will answer how to reduce the coupling.
You can experiment with capactive coupling as follows:
1. Cut two 30" lengths of #12 Romex. Either use a cable without an EGC wire, or just let the EGC float.
2. Strip the needed ends.
3. Attach one 30" length to a plug so as to connect the white wire to 120 neutral, and black to 120 hot.
4. Using a 10 megohm input impedance AC digital voltmeter measure the voltages between the hot and neutral wires relative to a neutral found at the socket that the 30" wire is plugged into.
5. Disconnect the neutral wire at the plug which leaves the previously neutral wire floating, and make the same voltage measurements.
6.
Measurement 4 produces about 120 V and zero volts.
Measurement 5 produces 120 V and some small voltage, like in the range around 30 V. Your small voltage may be different than mine because of different cable construction.
7. Add 30" to the first section of Romex. The capacitively coupled voltage will approximately double. As you add more cable length the voltage won't increase as rapidly.
.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
Domnic,
Anytime two conductive surfaces are near each other it forms a capacitor.
When those two surfaces are insulated wires are near each other in open air or in the same raceway the resulting capacitance is one type of stray (unwanted) capacitance.
It can be a design factor in HV transmission lines and in digital and analog signal lines including audio, but can usually be ignored in residential wiring.
Some ways of reducing the unwanted side effects of capacitance include keeping wires separated (which makes stray magnetic fields worse) and, in the case of signalling, use balanced differential signals and twisted wire pairs.
As mentioned earlier, the most commonly seen effect of stray capacitance is when it raises the voltage of an open wire that runs near a hot wire. A low input impedance on the meter used to measure that voltage will act as a voltage divider that reduces the measured voltage.
Anytime two conductive surfaces are near each other it forms a capacitor.
When those two surfaces are insulated wires are near each other in open air or in the same raceway the resulting capacitance is one type of stray (unwanted) capacitance.
It can be a design factor in HV transmission lines and in digital and analog signal lines including audio, but can usually be ignored in residential wiring.
Some ways of reducing the unwanted side effects of capacitance include keeping wires separated (which makes stray magnetic fields worse) and, in the case of signalling, use balanced differential signals and twisted wire pairs.
As mentioned earlier, the most commonly seen effect of stray capacitance is when it raises the voltage of an open wire that runs near a hot wire. A low input impedance on the meter used to measure that voltage will act as a voltage divider that reduces the measured voltage.
rlundsrud
Senior Member
- Location
- chicago, il, USA
As has been previously mentioned, it is always there when electricity flows in a conductor that is isolated by a dielectric. I suppose you could exacerbate the phenomenon by having a hot (with an AC current) alongside an unterminated cable. You would then be able to measure voltage with a high impedance meter at the ends of the unterminated cable. Different dielectrics will yield different results.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
That dielectric could just as well be air or vacuum. A "good" dielectric can produce a higher capacitance than just air. The key is the two conductors need to be isolated rather than in metallic contact.
- Location
- Illinois
- Occupation
- retired electrician
I don't understand the comment about metallic conduits. The capacitive coupling also happens in metallic conduit.
In fact, I first learned this issue when I was trying to ring out spare conductors in a metallic conduit that had energized wires in it along with the spares.
just the cowboy
Inactive, Email Never Verified
- Location
- newburgh,ny
4160 Volts too
4160 Volts too
When we worked on med voltage, the cable would build up acap charge between the conductor and shield in the plastic insulator. Aftershutting off the supply we would ground out that charge before working on it. Isaw an electrical engineer reach in and touch the wire and get zapped from the capacitancecharge on the open wire.
4160 Volts too
When we worked on med voltage, the cable would build up acap charge between the conductor and shield in the plastic insulator. Aftershutting off the supply we would ground out that charge before working on it. Isaw an electrical engineer reach in and touch the wire and get zapped from the capacitancecharge on the open wire.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
Just that there is no effective capacitance between a wire inside a metallic raceway and one outside the raceway.
ramsy
Roger Ruhle dba NoFixNoPay
- Location
- LA basin, CA
- Occupation
- Service Electrician 2020 NEC
Great idea, must ad this to my procedures. Engineers need this kind of interdisciplinary experience when working with field personal.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
151012-1553 EDT
domnic:
"how do I eliminate capacitance coupling ?"
Consider a stiff AC voltage source (meaning a constant voltage relatively independent of the load on the voltage source), a reference electode (a big flat sheet of copper), a hot wire a substantial distance from the copper reference plate, a floating isolated wire close to the hot wire, and the voltage source connected between the hot wire and the copper plate. You can call the copper plate ground if you want, I will.
There is capacitance between the hot wire and ground, This is effectively a shunt across the voltage source, and its current is of little concern in this discussion at this point.
There is capacitance between the hot wire and the floating wire. Also there is capacitance between the floating wire and ground, but this second value is much lower than between hot and the floating wire because of the relative spacings. The two capacitors form a voltage divider across the voltage source. Use an infinite input impedance meter and the voltage of the floating wire will be closer to that of the hot wire than ground if measured relative to ground.
You can reduce the effect of capacitive coupling by connecting a low impedance between the floating electrode, and ground. GoldDigger mentioned this. This lowers the voltage on the floating conductor relative to ground.
If you assume 1000 pfd for the hot to float capacitance, capacitive reactance is about 2.7 megohms at 60 Hz, then a 10 megohm meter would read about 10/10.36 = 0.96 times the source voltage. This calculation is from a right angle vector diagram.
A Simpson 260 is today 5000 ohms per volt or 250,000 ohms on the 50 V scale. Earlier models were 1000 ohms per volt on AC. For 5000 ohms per volt this would have a voltage reading to ground of 0.25/2.72 = 0.092 times the source voltage. At 120 V source this is 11 V, on the 50 V scale. In an actual experiment I read 11 V from 125 V. On the 250 V range the calculation is 1.25 meghoms, and thus 1.25/2.975 = 0.42 times source voltage. At 120 calculates to 50.4 V. Measured value at 125 is 45 V.
Now to elimination of the capacitive coupling. GoldDigger mentioned a conduit with the hot wire running inside the conduit. So to our above example add the conduit enclosing the hot wire with the floating wire outside the conduit, and ground spaced some distance away as before.
First, float the conduit also. Now there is capacitance from hot to conduit, conduit to ground, conduit to float, float to ground, and extremely little capacitance directly from hot to ground. But effectively there is capacitance from hot to float. Conduit will have a voltage fairly close to hot voltage relative to ground. The floating wire voltage will be somewhat less relative to ground than the conduit.
Second, connect conduit to ground. There is capacitance from hot to ground, but virtually none from hot to float. There is capacitance from float to both conduit and ground, but there is no voltage from float to ground because the conduit has now created a shield between the hot and float.
.
domnic:
"how do I eliminate capacitance coupling ?"
Consider a stiff AC voltage source (meaning a constant voltage relatively independent of the load on the voltage source), a reference electode (a big flat sheet of copper), a hot wire a substantial distance from the copper reference plate, a floating isolated wire close to the hot wire, and the voltage source connected between the hot wire and the copper plate. You can call the copper plate ground if you want, I will.
There is capacitance between the hot wire and ground, This is effectively a shunt across the voltage source, and its current is of little concern in this discussion at this point.
There is capacitance between the hot wire and the floating wire. Also there is capacitance between the floating wire and ground, but this second value is much lower than between hot and the floating wire because of the relative spacings. The two capacitors form a voltage divider across the voltage source. Use an infinite input impedance meter and the voltage of the floating wire will be closer to that of the hot wire than ground if measured relative to ground.
You can reduce the effect of capacitive coupling by connecting a low impedance between the floating electrode, and ground. GoldDigger mentioned this. This lowers the voltage on the floating conductor relative to ground.
If you assume 1000 pfd for the hot to float capacitance, capacitive reactance is about 2.7 megohms at 60 Hz, then a 10 megohm meter would read about 10/10.36 = 0.96 times the source voltage. This calculation is from a right angle vector diagram.
A Simpson 260 is today 5000 ohms per volt or 250,000 ohms on the 50 V scale. Earlier models were 1000 ohms per volt on AC. For 5000 ohms per volt this would have a voltage reading to ground of 0.25/2.72 = 0.092 times the source voltage. At 120 V source this is 11 V, on the 50 V scale. In an actual experiment I read 11 V from 125 V. On the 250 V range the calculation is 1.25 meghoms, and thus 1.25/2.975 = 0.42 times source voltage. At 120 calculates to 50.4 V. Measured value at 125 is 45 V.
Now to elimination of the capacitive coupling. GoldDigger mentioned a conduit with the hot wire running inside the conduit. So to our above example add the conduit enclosing the hot wire with the floating wire outside the conduit, and ground spaced some distance away as before.
First, float the conduit also. Now there is capacitance from hot to conduit, conduit to ground, conduit to float, float to ground, and extremely little capacitance directly from hot to ground. But effectively there is capacitance from hot to float. Conduit will have a voltage fairly close to hot voltage relative to ground. The floating wire voltage will be somewhat less relative to ground than the conduit.
Second, connect conduit to ground. There is capacitance from hot to ground, but virtually none from hot to float. There is capacitance from float to both conduit and ground, but there is no voltage from float to ground because the conduit has now created a shield between the hot and float.
.
- Location
- Illinois
- Occupation
- retired electrician
Yes, I agree with that....I just didn't read it that way the first time.
It is also a reason for grounding neutral: any capacitive voltage resulting from contact of a MV line with it is eliminated.
- Status