In the Standard ICEA S-95-658-1999 "Standard for Nonshielded Power Cables Rated 2000 V or less for the Distribution of Electrical Energy" Section 3.3 it defines the 173% insulation level for cables applied on a system where the time required to deenergize a grounded section is more than one hour. It states that the phase to phase voltage should be multiplied by 1.73 to determine the voltage rating which for a 480 volt system would be 831 volts. This requirement is also described in IEEE 141-1993 in the Chapter on "Cable Systems" Section 12.4.1. Does this not apply to 480 volt High Resistance Grounded Systems?
As far as I know there is no NEC requirement for insulation more than 600V for 480V systems, even ungrounded. I can't recall ever seeing anything other than 100% insulation (and always only implied, never stated) for 600V insulation. above 600V the next step is the 2kV class unless you are looking at very specialized conductor insulation.
The NEC doesn't adrress a lot of issues in industrial applications. I am concerned about the inadequate voltage rating for the power cables used for this type of system when a section is grounded and the two remaining phases have voltages exceeding 800 volts. I think this is a safety issue for people working on these systems. I recently refused to put my PE stamp on electrical drawings because of this issue and removed the high resistance ground from the incoming service in order to approve the drawings. Maybe I went overboard. But I think the NEC should address the issue.
Maybe I am missing something here. During a single-phase ground fault on a balanced HRG system, a neutral shift exists due to the voltage drop across the neutral resistor. The voltage-to-ground of the 2 remaining unfaulted phases to ground is in theory up to the line-to-line voltage (i.e. 480 V line-to-ground vs. 277 V line-to-ground during normal operation).
According to IEC Standards, for 400 V, for instance the cable will be 0.6/1 kV [something like above ICEA standard]. Usually the insulation thickness of European cable is less than for US cable [of the same voltage level].
Tim, 600 V insulation is adequate for 480V systems. Why would a cable rated for up to 600 volts not be suitable for 480 L-L and 277 L-G? Even during a fault, this insulation is sufficient as the L-L and L-G are still only 480 volts.
Read NEC 110.4 - "The rating of electrical equipment shall not be less than the nominal voltage to which it is connected". Thus, your cables rated at 600 volts are sufficient for a 480V system.
------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.
Unless the 480 V HRG network allows a ground fault to stay more than one hour we donot need 173% insulation.(We need 133% if it is less than one hour) During a ground fault it is true that the line to ground voltages of the healthy phases will be equal to line to line voltages. We should remember that these are steady state voltages.
But this 173% insulation is required to look after cable from transient overvoltages till the time the the ground fault is located and isolated.
As 7anoter4 pointed out this question does not arise in LV IEC cables because we use 0.6/ 1 kV cables for LV systems.
The similar thing applies for HV IEC cables. Depending on the time duration for ground fault clearence, we should specify whether the cable belongs to Category A, B or C.
0.6/1kV is (U0/U) U0 = The rated voltage between conductors and earth metallic screen;
U = The rated voltage between conductors, for which the cable designed.
The 1.73 formula is legit on ungrounded systems. During a fault voltage spikes are amplified by short duration capacitance to ground. In that case on 480 volt systems I use 7anoter4's rating of 0.6/1kV cable.
I would venture to guess that HRG would alleviate the need for that however.
I have no references however. Being a marine world we just kind of wing it.
In IEEE 141-1993 Chapter 7 "Grounding" Section 7.2 "System Grounding" it describes the following:
"High-resistance grounding will limit to a moderate value the transient overvoltages created by an inductive reactance connection from one phase to ground or from an intermittent-contact phase-to-ground short circuit. It will not avoid the sustained 73% overvoltage on two phases during the presence of a ground fault on the third phase."
I beleive this is the basis of the requirement for 173% insulation rating if the ground fault on a high-resistance grounded system will remain for more than 1 hour.
Speaking strictly from experience, I've NEVER seen anything other than 600 V cable for any 480 V system, including old 480 V ungrounded delta systems. And I've seen quite a few.
I don't know what to make of the ICEA or IEEE 141 statements, but most 480 V ground faults on HRG systems will exist for MUCH longer than one hour. More like days in many cases.
Transient overvoltages due to repetitive arcing fault re-strikes should be much lower on a high-resistance grounded system as opposed to an ungrounded system. That is the whole point of the HRG system.
Normal line-to-ground voltage on cable: 277V
Line-to-ground voltage during grd fault: 1.73*277 = 480
Required cable rating = max line-to-ground voltage times 1.73 = 1.1*277*480 = 527V assuming a 10% overvoltage <600V.
You neglected the phase relationship between the voltages. In addition this is an unbalanced ground fault on a three phase system which need to be handled by symmetrical components.
If cable insulation is sufficient for the voltage of the enclosed conductor to ground, the insulation of two cables will under all conditions be sufficient for the voltage between two phases. Your 480V system, not matter how grounded, will not see significantly more that 480V between conductors regardless of fault type. You can move the whole voltage triangle around relative to ground, and you can collapse the triangle, but you can't materially expand the triangle. Other than an arcing ground fault on an ungrounded (capacitively grounded) system, you can't get to 600V to ground on a 480V system. The arcing ground fault can cause severe overvoltages that are extinguished when something flashes over and no cable system is intended to protect against those overvoltages.
Why do you continue to want to fight years of accepted practice? 600V insulation is suitable for all 480V systems, no further testings/ratings needed.
I like dpc have never seen other than 600 volt insulated conductors used on 480 volts systems here in the US. I also agree that for an ungrounded system it is likely that the ground fault will exist for days if not weeks before it is corrected.
I don't agree with your statement that this is the accepted practice. You seem to ignore what is in the IEEE Recommended Practices and ICEA industry recognized standards which are, in my opinion, the accepted practice.
There is also a publication by GE on electrical system grounding written in December 1972 with gives the same recommendations. I have also discussed this with a power cable manufacturer which also agreed with the need to specify 173% insulation if a high-resistance ground system is utilized on a low-voltage power distribution system.
resqcapt19,
Just because you never seen other than 600 volt insulated conductors used on a 480 volt system doesn't justify not specifying it when required for a high-restance grounded system 480 volt system.
My friend, you are swimming upstream against the collective knowledge of engineers with about 100 years of combined experience in the field right here. I can sympathize with trying to justify your earlier decision, but I don't think you are going to make much headway. When you've dug yourself into a hole, the first rule is to quit digging.
I've looked at the Red Book comments, and I think it was really written with respect to medium-voltage cable, although I admit it doesn't say so explicitly.
But just because something is written in a "Recommended Practice" doesn't make it accepted practice anyway.
600 V insulation fully complies with the NEC as far as I can see and that is all I have ever seen used in the past 37 years. Trying to justify something in excess of that based on consensus standards and recommended practices is a tough road, unless you are buying the cable.
