The total transformer losses will be less at 50/50; the total core losses of two energized transformers is the same regardless of load split, but the load losses are related to I^2, the 50/50 split results in half the total load losses. If you have to do post contingency switching to recover the load following the loss of one transformer, in the 50/50 case you only lose half the load. In the 100/0 you're odds on going to lose 100% of the load, the stressed transformer is much more likely to have issues than the unloaded one.
Unless there are absolutely no prospects for load growth, two transformers loaded to < 50% - - preferably a good deal less than 50% - - will provide adequate capacity for the future; nobody will thank you if the new substation goes on load on a Friday and the following Monday Asset Management has to start planning for another yard or another substation...
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
For how long? An auto switching scheme could be harnessed to promptly [within no more than a few seconds] pick up all shed load on loss of one transformer; who knows, it might not even make the news...
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
Selecting the transformer size at 100% capacity it is recommended to perform a reliability-based approach to determine the number and timing of the transformer spare share strategy. The following should be considered:
- Cost reduction of 100% transformer size compared with the 50% transformer capacity.
- Check load growth and consider the ability to withstand overloading In both normal and emergency conditions
- Option to increase efficiency using lower-loss core materials at adder price.
- Explore if there is an opportunity to use a spare transformer or available mobile substation shared with other stations to reduce the
the capital cost of a third 100% transformer onsite.
- Analyze the N-1 principle in the event of simultaneous transformer failure and risk impact in load curtailment.
Below is a shape of loss-efficiency curve that hopes to help to visualize the losses for 50% and 100% loading transformers and associated losses.
.....
To serve 100 MVA, would the choice be either:
A)2 transformer each rated 200 MVA
B)3 transformer each rated 100 MVA
I suspect option A would have much lower capital cost than option B.
Having a hot spare and two transformers at 100% loading seems like you will be doing thermal damage to both fully loaded transformers. It sounds like fault current limits you to no more than 200 MVA connected.
One other option would be having two transformers loaded at 50% and the third loaded at 100%.
I will vote for 2-200 MVA transformers to meet N-1 criteria.
Although the cost differential to purchase 3-100 MVA (300 MVA total)is lower than 2-200 MVA (400 MVA total) a holistic evaluation is recommended to consider the extra cost impact of additional unid such as transportation, rigging, installation, testing, commissioning extra oil containment/foundation, fire wall, extra aux. power, protection and control, raceways plus losses and O&M cost made the 3-100 MVA option less attractive. However, this need to be validated with actual cost based in this specific case.
I'd say so. There aren't many installations with hot spares.
One of the reasons to avoid 100/0/100 loading (besides perfect balancing which can be impossible) is switching from neighboring circuits must from the same trafo. Further, because of this, its a good idea to replicate both MV busbars- so that when one is taken from service circuits can be placed on the other still fed by the same trafo. Further, any secondary automatic throw outside the substation must be taken into consideration- meaning it will have to be disabled at times or loading below 100% is to be considered or live with trafo life reduction.
Transformer cost and losses vary as MVA raised to 0.75 ratio ie 100 MVA transformer cost and losses will not be two times those of 50 MVA, but 2 raised to 0.75 ie 1.7 times of 50 MVA. It is always better to go for 2nos P rating ( provided your MV breaker rating is not exceeded) with both loaded at 50 % of total load. Remember any transformer kept in deenergized state is prone to absorb more moisture than a loaded transformer. Paper aging also will be less in a 50 % loaded unit, giving more life.
It depends on the rating of MV breaker. A 300 MVA 220/33 kV is impossible as you will not have a suitable breaker for LV side. So for 300 MVA load, 4x100 MVA or 3X150 MVA will be ideal.
4x100 would result in an overload of one unit operated as two M-T-Ms - unless you mean switching operations or those 4 running in parallel. The prior has given me something to think about.
PRC - I saw a utility using pairs and trios of 75MVA 230/12/12 kV transformers for big city distribution stations to solve the LV breaker size problem.
I scanned a list of all their breaker purchase contracts for one year. In the 12 & 25 kV classes 600A to 2000A CB prices were all within about 20% of each other. 2500A to 4000A were likewise within 20% but were twice that of the 600A-2000A group. Back to the 230/12/12 kV transformers, the CB pricing info meant that building the stations with two 12 kV rings was not a cost issue.
