Transformers Redundancy

[rrui00]

I would like to receive advice. We are going to receive in a new facility 115 KV, and we need to step it down to 13.8 KV. We are thinking of installing 2 transformers in paralell for redundancy. Is that a normal practice? or should we just install one transformer.

 

[Noway2]

I would suggest at least planning for the second transformer, but not necessarily installing it. I suggest this thinking along the lines of expansion needs rather than redundancy. Transformers are one of the most perfect machines made and have a low failure rate to where I would doubt you would get your money out of it. If redundancy is a concern you would be better off looking into either a UPS system or a transfer switch that works with two utility feeds.

It is a lot cheaper to plan for the second transformer and allocate space, add the conduit, mounting pad, etc during planning and initial construction that to try to add it in later.

 

[HCBFlash]

Purely for redundancy? I'd say it'd be a waste. I'll echo Noway2.

 

[DaveScott]

I think it depends on the value of the process, and details of the load.
For example, if you can drop 40% load in an emergency, then you could purchase two transformers, each with 60% capacity.

 

[dpc]

It depends on a how much an outage will cost. With one transformer, if it fails, you will be out of business for a long, long time, unless you can locate a replacement in a hurry.

For facilities that can't tolerate downtime, two transformers are pretty common. For some facilities in the US, such as wastewater plants, it's basically the law.

Typically, with two transformers, you would size each one to be able to carry the entire load using the maximum forced cooled rating.

 

[slavag]

Hi.
I'm totaly agree with DPC.
We also recommend in lot of cases add coupler on BB.
In case of petroleum/FAB plants we also recommend check with utilities optional separated connection of MV infeed for back-up and black-out starts.
Regards
Slava.

 

[dpc]

For a new large transformer with load tap changers, we are getting delivery time estimates of 100 weeks in some cases.

There seems to be long lead times these days for transformers - some shortage of electrical-grade steel. I guess China is using it a lot of it up for their internal growth.

 

[VEBill]

This is not my field, but just to ensure clarity...

Two such transformers would not be simply ~wired~ in parallel for redundancy, right?

OP mentioned ~installing~ 'in parallel'.

 

[itsmoked]

Infernal growth...

Would it not be a poor idea to have two transformers that can each run the entire load, for redundancy, running in parallel? Could not something then take out both at once, like lightning? And could that also require much larger interruption requirements on both sides of the bank? Also perhaps some current hogging issues?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[slavag]

Hi.
"Redandancy" is not mean parallel operation of two xfr's.
In this case are possible two options:
1. Both of transformer in operation , but coupler is open
disandvantage : not full load of xfr and twice losses.
2. Coupler is close and only one xfr in operation.
Of course some ATS system.
Regards.
Slava

 

[davidbeach]

Pretty much all possible means of using two transformers to supply one overall load are in use. How closely they are connected depends on the sensitivity of the load to outages and the cost of those outages. Both connected in parallel on the secondary means twice the available fault current on the secondary compared to no connection. Connection in parallel will also require additional protection to avoid faults on the high side of one transformer being fed through the other transformer.

It all boils down to the specifics of the application, and we probably can't fill in the necessary details here. It would probably be best if you hire a consultant familiar with these kinds of installations who can look at the drawings rather than trying to design the system through the internet.

 

[RRaghunath]

rrui00,

I have seen the practice followed in power plants as well as in oil & gas installations. The cost of unavailability of transformer is much larger than the extra money you spend on having 100% redundancy. This applies to most of process / production industries too.

It is always 2x100% transformers. When it comes to paralleling at the low bus, I suggest it can safely be done and can contribute to better reliability if the design issues are properly addressed. A few things I can recommend,
1. The over current/earth fault protection at the transformer incomers shall be summation type, that will open the respective incomer breaker and bus coupler for any uncleared downstream faults or faults in the respective bus section.
2. Directional overcurrent earth fault relays in the transformer incomers set sensitive and to look towards the transformer and can act as back up to the transformer differential / REF protection.

IEEE 242 (Buff book) and IEEE 141 (Red book) are quite helpful in this regard.

 

[Noway2]

If you are concerned about reliability, I would still recommend a medium voltage transfer switch over redundant transformers.

The transfer switch can detect problems and get you away from the poor source to a reliable power source where transformers can't.

Take for example, if a fault occurs stream from one of your transformers, the fault condition will still be seen by the second (transformer) system where as a transfer switch can isolate you from the fault. Note that a transfer switch will also operate against quality type faults, not just over current faults. My point here is that a transformer with breakers is NOT equivalent.

 

[ScottyUK]

Transfer switches require maintenance and fail too - redundant transformers give something closer to full independance. Consider how far back into the utility the redundancy exists - at some point you will hit a single point failure unless you are near a major load centre with multiple grid infeeds, and can afford the cost of providing the redundant connection.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[slavag]

Hi.
I think our discussion is about different things.
Word "redundancy" make some misunderstanding.
I told about "redundancy" on HW level, that means:
Maintenance of one of xfr.
Power supply of plant in case of trip in one of xfr.( is reduce outage time in case of xfr. repair)
Flexibility in case of problem on one of BB (CB fault, SC on one of BB, etc.)., that mean possible fed MCC,etc on low level from energ. BB.
ATS as "simple" system, long delay ( several seconds), operated by UV situation. Not special MV transfer switch
and not high speed bus transfer switch.
Paralleling operation of two xf. for full redundancy is other story:
Protection coordination ( see post of Raghun).
Tap changer coordination ( master, follower, etc.).
Other level of current fault.
May be other primary swg.
I think is good for critical MCC's and switchboards, but not for grid connection to plant. Is not hospitals, FAB's.
Regards.
Slava