I know this is way off the OP post, but since we seem to be going there.....my understanding is that in California, the tanks of overhead transformers are never grounded. GO95 or some such regulation. The thinking is that there is less potential for incidental contact with ground when working on live lines in the vicinity of the transformer primaries. That's why you see two bushing transformers instead of single bushing. Is there a more appropriate forum to discuss this kind of stuff? Don't want to get blacklisted for hijacking threads.
Transformer tank grounding varies in California. There are indeed floating tanks in California employed by some utlities. Some one told me on here its done to help prevent flashover and make servicing lines safer. An equal potential method is employed.
The double bushing transformers are actually required for different reasons. In California PUC 95 applies to line construction which also has rules on EMFs and limiting stray currents on grounding systems much like other countries. As a result all transformers must be connected phase to phase or if connected in wye phase to an isolated neutral kept up on insulators and minimally grounded. Hence 3 wire is the dominant design and where a neutral is present its propt up on insulators. A 4th or 5th ground wire may be run but is only for grounding and bonding, no neutral load current.
Most (not all [I will explain latter]) 3 wire lines in California are actually uni grounded wye (supplied via a wye secondary substation transformer with the neutral solidly grounded to the substation grounding mat but often not run along with the poles). Just because all loads are connected in delta out on a line doesn't mean the substation supply transformers is also delta on the secondary. Common misconception. When a uni grounded line does ground down current does flow. The amount is dependent on the resistance of the fault, the soil resistance and the distance from the supply substation. Generally faults close to the substation are much higher in current magnitude then those furthest since less soil impedance is present. Running an EGC along the line drastically helps to increase the current since its no longer dependent on soil resistance. If faults don't generate a high current (they don't reach the breakers or fuses normal trip curves) they are cleared via GFI logic in the recloser or substation breaker. Because no phase to neutral loads are connected to the grounding system like in a multi grounded neutral system, its possible to sum the readings across the 3 phases. Exact same concept that applies to GFCIs, RCDs, and GFI logic. One could have 459, 429 and 487 amps for example on each phase from all the delta connected loads. If no ground faults exist the zero sequence sum would equal to zero. However, if a 15 amper fault took place on phase A, when the new values 474, 429 and 487 are sumned in the recloser's or breaker's ground fault logic relay it will come out as +15, indicating a ground fault exists out on the line. If the ground fault persists long enough over the established time current curve it will open the breaker or recloser.
"Also, with an ungrounded Delta primary, how would any ground faults trip adjacent reclosers or substation breakers. No ground fault current, since there's no reference to ground."
Now, Ungrounded lines. Current, believe it or not does flow during a ground fault :happyyes:There are genuine ungrounded delta lines in California where the supply substation transformer is indeed delta secondary. Actually a lot more common in California then in most other places. When a phase does ground down current does flow believe it or not. Not much but it does due to the fact a capacitive potential does exist between the phases and earth. The longer the line is the more capacitive potential and thus the more current. One average, assuming an average sized system that current is about an amp or two. During the fault the current also rises to the phase to phase voltage on the none faulted phases so lines are built with fully rated insulator, bushings and surge suppressors. (More expensive) These systems have the advantage that they can run with a faulted phase as well as fewer momentary interruptions (reclosing and voltage sags from single phase to ground faults) since a squirrel bridging a bushing or tree touching a single phase wont cause a violent phase to ground flash over or massive currents to flow to ground. Better service continuity and far less thermal/magnetic stress on the system over time as long as the first fault is fixed before a second one occurs latter on. (2 faults on different phases cause massive current flows). There is one major disadvantage in that an arcing fault can cause voltage stress above the phase to phase voltage due to a type of "resonant" effect. This of course can be deterred by making an artificial neutral point or using the supply transfomer's XO (if it was previously a floating wye secondary) and grounding that neutral point or XO either through a properly sized reactor (Peterson coil earthing) or via a resistor that that will pull current slightly higher over what would normally flow during a solid phase to ground fault. IE, capacitive phase to ground reactance causes 1 amp to flow during a ground fault; the resistor's ohm value is sized to allow 1.5 or 2 amps to flow. It causes more current to flow during a fault yes (still a few amps and can be continues if the resistor is rated for it), but this "shunts" the resonant effect of an arcing ground fault on an ungrounded system. At lower voltages such as 240 its usually not a concern but at 16kv an arcing ground fault can cause significant voltage spikes.
FWIW this also holds true for ungrounded industrial systems. Hence why most newer systems are high resistance grounded. The factory can still run on a faulted phase, but the resistor prevents an arcing fault from destroying every MOV and puncturing cable insulation. A 1.73 rise still happens during a fault just not 6, 8 or 12 times voltage surges
:thumbsup:
Now, even though current does flow its not enough to trip a standard recloser or breaker. Standard RCD logic is hard to pull of to due to the nature of it all, both the small amount (often under 5 amps) and the need to do complex differential current flows. You could add sensitive enough CTs on all the breakers and use complex logic (it does exist) to monitor currents going in and out of the feeders and phases, but its complicated an expensive. Setting up the relays to differentiate between a none faulted feeder and faulted one is difficult as well as accuracy under so little current. Generally the best thing to do is just add a 15 amp draw (continues rated if you want to have the benefit of running the feeder until the fault is found) resistor and use standard RCD logic. Makes things a whole lot easier.
However, there is an easy way to indicate a ground fault exists. California utilities that operate genuine ungrounded systems will take 3 25kva pole pigs and wire them up grounded wye primary broken delta secondary. The secondary is wired across a relay. The primary neutral is connected to the substation's ground mat. This is the only wye grounded load in the system. Under normal conditions not much voltage is present across the relay because the primary voltages are fairly even between phase to ground, but when a phase to ground fault occurs the delta secondary will attempt to circulate currents to balance the phases out. Because its a broken delta and it can't (relay is across it) a voltage rise takes place enough to close the relay. The relay limits the current flow but the voltage remains high enough to keep it closed as long as the ground fault exists. If the relay was not present (closed delta) a massive amount of current would flow eventually blowing the protective fuses. If those transformers were large enough, say 6MVA each, low impedance and had a closed delta secondary any ground fault out on the line would result in thousands of ampers of current flow! The system would behave as a solidly grounded wye, since it now has a neutral point!
An artificial neutral yes, but none the less one that will behave like a normal one. The current would blow regular fuses and trip recloser on regular time current curves since the bank now gives a place for neutral current to flow. This is exactly what happens in a wye grounded delta banks of any size. They are basically an artificial neutral that tries to balance voltage discrepancies out. Your not hijacking this thread. Just giving a detailed account of why a Y delta bank wont work for the op in his predicament
:lol:
Anyway, as for the poco once the relay closes it alarms them a ground fault exists. Its then up to them to do the hard part and track it down. Adding resistors in parallel with the relay can increase the current to the point regular RCD logic can pick it up, so it helps to pinpoint what circuit is in trouble.
Hope I didn't throw you off. Grounding and bonding complex topic with a lot of dynamic parts to it. Let me know if confused you on anything
