Surge protection on 27kV ungrounded delta system

[wirenut60]

I'm involved in a project where the existing plant distribution is 27kV with a 5kV standby generator. The plant receives power from the utility’s 27kV grounded wye distribution system. The step up transformer between the 27kV bus and the 5kV bus has a delta ungrounded primary, (27kV), and grounded wye secondary (5kV). The original distribution system had MOV surge arrestors on the primary of every 27kV transformer on site. After completion of the generator addition, the plant experienced the failure of most of the MOV surge arrestors that were connected to the plant distribution system when the plant was operated on generator. All MOV surge arrestors were then removed from the 27kV distribution system. I did a little research and found that the MCOV ratings required for 27kV class MOV surge arrestors were not high enough for a ungrounded delta system.

I believe that the step up transformer between the 27kV system and the 5kV generator should have had a wye primary. This would have avoided damage to the MOV arrestors. I have discussed this system with an engineer familiar with, but not directly involved with the generator installation. He maintains there was a reason for this transformer to have the delta primary… but cannot recall the reason. Does anyone know of a reason for installing this transformer with a delta ungrounded primary? I’m starting to think it was installed in error and should have been a wye, or grounded delta, since I do not know of any reason to validate the ungrounded delta primary. Does anyone have some experience on a similar issue?

 

[davidbeach]

The transformer is definitely backward; it should be delta on the generator side and grounded wye on the system side. The failure of the MOVs was easily predicted and should have been seen coming from the very beginning.

I would call the primary of a GSU to be the generator side even though it is the low side as power will always flow from low to high side. Probably best just to refer to high and low side. Because of this uncertainty in terminology the whole problem may have come about from someone calling for a delta primary meaning low side and the transformer being ordered by someone who took primary to be high side.

 

[jghrist]

See Interconnection Transformer Winding Arrangement Implications on IPP Protection at

http://www.beckwithelectric.com/powerlines/powerlines-37.html#4

 

[wirenut60]

Thanks for the reply. I agree... the xfmr is backwards from what it should have been. I was able dig out the original contract drawings this morning... looks like the design shows the transformer having a delta winding on the 27kV side... so... either engineering error or cad error... either way, not a confused contractor or supplier... they just followed the dwgs. Other than phase to phase surge protection, is there any type of surge protection available for this system?

 

[davidbeach]

On any system that can operate ungrounded the surge protection needs to be rated for the phase-phase voltage but then may be connected phase-to-ground.

 

[apowerengr]

wirenut60 - you state the plant's distribution is 27kV with 5kV busses and multiple transformers. It sounds like the 5kV system is the utilization voltage, as such 5kV wye-gnd is most likely the correct connection. Many of these utility supply systems were orignally three-wire three-phase so the delta winding at 27kV is correct. When they were converted for use as local distribution lines the neutral was extended from the source substation and line-to-ground connected loads were added. As davidbeach stated arrester failures when running on the generator only was a very predictable consequence as isolating from the utility removed the 27kV system's ground source. Without changing multiple transformers I'd agree the arresters need to have an MCOV rating at least 27kV and then are connected line-to-ground.

 

[wirenut60]

apowerengr... i agree, it should have been a predictable issue. However, i've walked into a multitude of existing problems while designing additions and alterations to the existing system. The ungrounded delta and surge arrestor problem is but the tip of the iceberg. I found (4) 2500kva transformers, (23.8kV - 480/277v), that had no primary protection other than the plant's main fusing, which was 200a @ 23.8kV. And any type of coordination was nearly non-existant. The plant does have dual 200a feeders with everything split between the 2 feeders with most fuse swgr incorporating a "main-tie-main" arrangement. I've been able to correct the primary fusing issues by incorporating the addition of proper fused swgr in various projects. And we've been able to provide proper coordination to the system. But this latest issue with the ungrounded delta system has got me wondering what is going to be next!
BTW... most 23.8kV terminations in the plant are 25kV class 200a load break elbows, in order to get a load break elbow with the proper MCOV rating i would have to go to a 35kV class load break elbow... which is GREAT!, except for the fact that 35kV load break MOV will not fit a 25kV bushing!... woe is me... what's next?!... LOL! I won't even go into the use of 15kV terminations on 25kV class cable!