Phases in separate pipe?
- Thread starter nightmare
- Start date
- Status
nightmare
Member
- Location
- MD and currently overseas
Where are the ground conductors in this photograph? Is there no need for neutrals because it is a 480 V delta feeding only 3 phase motor loads? Perhaps my monitor is causing me some problems but, the smaller wires in the background look like they are marked with gray tape. Please explain cause I'm very grateful that you provided the pic and I'm, now, working a big problem that exists with my ground wires and I'm not sure, yet, where to look.
ohmhead
Senior Member
- Location
- ORLANDO FLA
Well thats a service from a transformer there is no need for a grounding conductor .
Thats iwire showing his isolated paralleled feeders to a switchboard .
Question do you have lots of nonlinear loads ?
Last edited:
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
Whenever current flows in a wire, a magnetic field is produced which surrounds that wire. If the current flow is AC, then the magnetic field will be a continuously changing magnetic field.
Whenever you have a wire in a _changing_ magnetic flux, voltage will be induced in that wire. This includes both the wire that creates the magnetic field in the first place, and other wires nearby. This is the basis of inductance and transformer action. The magnetic field creates the magnetic flux, depending upon the reluctance (magnetic resistance) of the flux path.
When you have the three phases next to each other, the current flowing in each wire is balanced by the current flowing in the other two. The net result is that you get zero net current in the entire bundle of phases. Even though each phase on its own is generating a magnetic field and has inductance, the set of phases together produces only a slight external field and has slight inductance. The larger the space between the conductors, the more magnetic field on the 'outside' of the bunch.
Any ferromagnetic materials in the vicinity will act to 'amplify' the amount of magnetic flux produced by a given current; thus increasing the inductance and the induced voltage caused by any magnetic field.
The 'isolated phase' installation is fine as long as you don't have ferromagnetic rings that enclose the individual phases, and as long as you don't have any conductive loops that are coupled to the magnetic flux produced by the isolated phases.
I believe that the EGC conductors which you added have _caused_ a serious problem. I presume that these EGC conductors are bonded together at both ends of the run. Now consider the following path: start at the EGC junction/termination at one end of the run; now follow the EGC that goes with phase A to the other end of the run. That EGC lands on a junction/termination with all of the other EGCs. Follow the EGC that goes with phase B back to the start of the run. Again this EGC lands at a common termination. What you have is a nice low impedance loop of wire, sitting right in the magnetic field created by the phase conductors.
If the EGC had _zero_ resistance, then the current in each EGC wire would exactly match the current in the corresponding phase. This is because the current induced in the shorted EGC loop will tend to balance the current creating the magnetic field in the first place, thus reducing the magnetic field that induces the voltage in the EGC loop. The actual current flow is set by the balance of phase conductor current, EGC loop current, with the resulting net current producing a magnetic field which produces a voltage that overcomes the resistance of the EGC loop.
-Jon
Whenever you have a wire in a _changing_ magnetic flux, voltage will be induced in that wire. This includes both the wire that creates the magnetic field in the first place, and other wires nearby. This is the basis of inductance and transformer action. The magnetic field creates the magnetic flux, depending upon the reluctance (magnetic resistance) of the flux path.
When you have the three phases next to each other, the current flowing in each wire is balanced by the current flowing in the other two. The net result is that you get zero net current in the entire bundle of phases. Even though each phase on its own is generating a magnetic field and has inductance, the set of phases together produces only a slight external field and has slight inductance. The larger the space between the conductors, the more magnetic field on the 'outside' of the bunch.
Any ferromagnetic materials in the vicinity will act to 'amplify' the amount of magnetic flux produced by a given current; thus increasing the inductance and the induced voltage caused by any magnetic field.
The 'isolated phase' installation is fine as long as you don't have ferromagnetic rings that enclose the individual phases, and as long as you don't have any conductive loops that are coupled to the magnetic flux produced by the isolated phases.
I believe that the EGC conductors which you added have _caused_ a serious problem. I presume that these EGC conductors are bonded together at both ends of the run. Now consider the following path: start at the EGC junction/termination at one end of the run; now follow the EGC that goes with phase A to the other end of the run. That EGC lands on a junction/termination with all of the other EGCs. Follow the EGC that goes with phase B back to the start of the run. Again this EGC lands at a common termination. What you have is a nice low impedance loop of wire, sitting right in the magnetic field created by the phase conductors.
If the EGC had _zero_ resistance, then the current in each EGC wire would exactly match the current in the corresponding phase. This is because the current induced in the shorted EGC loop will tend to balance the current creating the magnetic field in the first place, thus reducing the magnetic field that induces the voltage in the EGC loop. The actual current flow is set by the balance of phase conductor current, EGC loop current, with the resulting net current producing a magnetic field which produces a voltage that overcomes the resistance of the EGC loop.
-Jon
art82
Member
- Location
- collegepark md
phases in one pipe
phases in one pipe
to ansswer your question its not a good idea and why would apply this applicatiobn its called histeria i think the polarities in each pipe per phase its not a good idea in the long run it might affect if you have linear loads
phases in one pipe
to ansswer your question its not a good idea and why would apply this applicatiobn its called histeria i think the polarities in each pipe per phase its not a good idea in the long run it might affect if you have linear loads
nightmare
Member
- Location
- MD and currently overseas
a relatively small number of flourescents and some HPS lighting and some small split pack heat pumps
nightmare
Member
- Location
- MD and currently overseas
Thank you for the explanation. I'll disconnect immediately.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
You are welcome to the explanation. Please note, however that it is not necessarily a suggestion that the EGC be disconnected, but instead a suggestion that the EGC as configured in this installation is a hazard.
The first question I think that you need to ask is: is the EGC required?
If the EGC is required, then you need to come up with an EGC design that doesn't introduce the current flow issues.
The 'standard' approach is, of course, having A-B-C-N-G in each pipe. This arrangement of conductors is intended to reduce the magnetic field that gets coupled out of the phase sets, avoiding inducing current in the EGC. However you are attempting to avoid this because at the present time you have an isolated phase installation.
I don't know if there is an acceptable method of adding an EGC to an isolated phase installation.
-Jon
- Location
- Massachusetts
If you had an isolated phase installation that required an EGC a full size EGC would have to be included in each raceway.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
That is what I thought.
But if you have an EGC in each raceway, and have all of the EGCs connected in parallel as was done here, then you will have undesired current induced in the EGCs.
Dumb question: Do all of these full size EGCs need to be joined together at _both_ ends, or could an EGC be placed in each conduit but only one actually connected at the downstream end?
-Jon
I'll second that.
Would these currents still show up if the EGC were looped back and forth through the pipes, like a zig zag, with only the one point grounded on either side? And be code compliant, with a full size EGC?
Last edited:
nightmare
Member
- Location
- MD and currently overseas
I guess we are both dumb 'cause I was wondering the same thing. Very seriously, though, it was my plan to leave only one of the egc's remaining and the neutral pipe is the only choice I seem to have. According to table 250.122 the size should be 1/0 for my 800 A main. Perhaps we can talk about the nature of the EGC being colocated with the iso phases. What purpose does the EGC serve for each of those phases if any? I'm having trouble seeing in the code and electrically how locating the EGC outside of the isophase pipes is a violation or a life safety issue. Anybody?
ohmhead
Senior Member
- Location
- ORLANDO FLA
Well just a thought if you run a grounding conductor in each raceway of a paralleled array of raceways and in a isolated arrangement of A / B /C /N each with one grounding conductor with one common same phase per conduit .
You have a ground size per code for your breaker size one in each raceway you have a series & paralleled path in a fault condition .
You have one grounding conductor that must handle the fault of multiple phase conductors of that one phase thats in a fault.
You have a connection of paralleled grounding paths meaning current flows in one ground to that fault but divides up the fault current of one ground to the other grounding paths .
You have a larger current on one ground that now is more damaging to the minor fault only one phase is in a trip condition the other two phases are not in the picture .
So its my thinking that a ground is only used for a fault or a short circuit back to the return circuit to limit the safe return and trip a breaker .
If your ground is small its resistance is long time to trip breaker .
If your ground is a larger its resistance is short time to trip breaker .
The longer it takes to trip the more damage is done .
Heres what i see a longer time to trip but at a higher current level in that same time frame .
You have a ground size per code for your breaker size one in each raceway you have a series & paralleled path in a fault condition .
You have one grounding conductor that must handle the fault of multiple phase conductors of that one phase thats in a fault.
You have a connection of paralleled grounding paths meaning current flows in one ground to that fault but divides up the fault current of one ground to the other grounding paths .
You have a larger current on one ground that now is more damaging to the minor fault only one phase is in a trip condition the other two phases are not in the picture .
So its my thinking that a ground is only used for a fault or a short circuit back to the return circuit to limit the safe return and trip a breaker .
If your ground is small its resistance is long time to trip breaker .
If your ground is a larger its resistance is short time to trip breaker .
The longer it takes to trip the more damage is done .
Heres what i see a longer time to trip but at a higher current level in that same time frame .
Last edited:
The idea of having 3 EGCs is to have 3 paths for ground fault current to flow, giving 3 times the ampacity and 1/3 the resistance. Leaving one end ungrounded prevents the EGC from carrying any fault current. Looping one EGC through 3 conduits just makes the fault current flow 3 times as far and the resistance is 3 times instead of 1/3.
As explained by winnie, having one EGC in close proximity to one phase causes induced currents. This would be the case also if you had only one EGC in one conduit. I'm not sure that this is a serious problem, assuming that the EGC has the ampacity to carry the induced current and that the heating of the EGC from the current doesn't reduce the ampacity of the phase conductor too much. Determining the reduced ampacity would be complicated.
It would be much better, for increased ampacity, for reduced ground return path impedance, and for reduced circuit impedance, to have each conduit have all 3 phases plus an EGC.
As explained by winnie, having one EGC in close proximity to one phase causes induced currents. This would be the case also if you had only one EGC in one conduit. I'm not sure that this is a serious problem, assuming that the EGC has the ampacity to carry the induced current and that the heating of the EGC from the current doesn't reduce the ampacity of the phase conductor too much. Determining the reduced ampacity would be complicated.
It would be much better, for increased ampacity, for reduced ground return path impedance, and for reduced circuit impedance, to have each conduit have all 3 phases plus an EGC.
nightmare
Member
- Location
- MD and currently overseas
I organized a shut down and rephased the conductors so that we, now, have ABCNG in each pipe. I am regretfilled, however, because I think we abandoned the concept/design/hope of an isophase installation without fully understanding or exploring it's total value. An instance where life safety trumps genuine understanding for the greater good. I will be sure to let y'all know if I figure out how to mitigate the induced current loop problem while still providing code compliance and defference to life safety.
- Status