There are times when life puts a wall in front of us, when no medicines work, when all available tools are of no use, when all that we know is irrelevant, when none of our experience comes in handy. At moments like these, all that goes on in our mind translates in to one simple question - Now what? This simple but profound question stirs our emotions, baffles our intelligence, makes us helpless, forces us to reboot and relearn and leaves us with memories for a life time. Times when we came to face this question are the turning points in our lives. They shape our lives, build our characters and make us who we are.
But those are also the moments when innovation triggers in some corner of the human brain. New concepts, new instruments, and new methods evolve. The initial spark of imagination in the mind of one passionate and fearless individual appears absurd at first but slowly gets refined and gains acceptance. Over the years it spreads to impact everything - companies, industries, communities, societies and even countries. Thus the old appears irrelevant and often incorrect and gives way to new.
The life cycle of large infrastructure projects is no different. They also go through this inevitable process of evolution. When major components of large projects cannot be interfaced correctly to work properly because wrong components have been delivered at project site, this question comes back to hunt everyone - Now what?
Whenever items delivered to a project are incompatible, site Field and Commissioning engineers take on the challenge. If possible and acceptable, they try to modify the interface in such way that the item can be used without extensive rework and modification to avoid costly reorder. But once in a while the sheer physical size of a component is so large that it is just not possible to carry out any physical modification whatsoever in any manner because suitable interface changes are just not possible. This is exactly what happened at one NTPC power station, where during the commissioning of the generator transformer it was found out that the generator bus ducts were of wrong design. This story describes what the impacts were and what was done at site to resolve the problem to go ahead with the project.
Large utility generators like those of 500 MW capacity, have three bus ducts, one for each phase that connect the generator to the generator transformer. These are aluminum tubes of about a meter and half in diameter, more than a few centimeters thick with an aluminum conductor centrally supported by insulators. These tubes come in pre-fabricated transportable spools and are welded together at site. When the spools are welded together as designed, the bus duct takes the shape it is intended to and connects the generator, generator transformer and the unit auxiliary transformers in a pre designed manner.
The main GT at the project comprised of three single-phase transformer units (connected externally through the generator bus duct) to act as a three phase transformer. On the generator side, the low voltage windings of the three single-phase units are connected as a "delta". For forming a delta on the LV winding, each bus duct coming out of the generator is split into two smaller ducts. R&Y go to the first transformer, Y&B go to the second and B&R to the third transformer. On the grid side, the high voltage windings of the three single-phase units connect as "star". This particular connection creates a phase angle difference of 30 degrees between the HV and LV of the GT. Keeping this phase shift in view, the unit auxiliary transformers are so designed that this phase shift is compensated.
Why Generator transformer, unit auxiliary transformer and the startup or station transformer must be so chosen that the two voltage sources are in-phase? This is necessary to ensure that uninterrupted auxiliary power can be available to the power station through a station transformer when the generator transformer trips.
Read on for some more explanation.
To generate electricity in a power plant, a number of equipment have to run and all of them consume electricity. Typically, 5 to 10 percent of power generated by a power plant gets consumed in the very process of generating that power. This power is called - auxiliary power. Even when a power plant is under shut down (not running), it consumes some auxiliary power for equipment safety and maintenance.
When a power plant generates power, it takes its auxiliary power requirement from its own generation through unit auxiliary transformers. When it stops or is not running it takes this auxiliary power requirement from a station transformer that is always connected to the grid. This switching over of power from one source to the other is called "change-over". Change-over of auxiliary power to plant has to happen without interruption, otherwise, all the plant equipment will trip during the interruption. This means, there is a small time interval during which both the unit auxiliary transformer and the station transformer have to remain connected together to the auxiliary loads, after which the source not required can be switched off.
For two power sources to be connected together, their voltages have to be in same phase angle. For this to happen, the phase shift due to the GT must be compensated by the unit aux. transformer. This requires that the vector group of the GT and the unit aux. transformers must complement each other in such a way that the unit transformers should bring back the phase angle of the voltage by the same angle that is shifted by the GT.
When the bus duct was and assembled at site, we observed that Y phase of generator was going to the R phase of the transformer and R phase of the generator was going to the Y phase of the transformer. Similarly, B&Y (in place of Y&B) and R&B (in place of B&R) was going in to the second and third transformer. This was applying a voltage with 180-degree phase shift to each transformer. This meant the phase angle between the high voltage and low voltage of the transformer would now be (180+30) or 210 degrees in stead of 30 degrees as designed. The phase shift a transformer introduces is conventionally represented on a clock face - e.g 30 degrees represents 1 O clock, 150 representing 5 O clock etc. By that convention, the GT was connected as a YdN7 and not as YdN1 as intended.
The massiveness of the mistake was such that virtually nothing was possible to rectify the problem. As far as the GT and the bus duct were concerned, there was no possibility of any modification at site. Period.
But so what if the phase angle is 210 degrees instead of 30 degrees?
Will there be any problem in synchronization to the grid? The answer is - No. The generator is a free running source and once synchronized, will keep generating happily without any problem. But after synchronization, the output voltage of the unit auxiliary transformers will be out of phase with the station auxiliary bus voltage coming from the station transformer. Because of this, the two sources cannot be connected in parallel and switching from one source to the other source will not be possible without interruption of power to the plant. This will pose serious problems in reliable and safe operation of the plant. There was nothing that could be done with bus duct. There was nothing that could be done with the generator transformer. Therefore, on everybody's face there was that question we began with - Now What?
The only way to match these two voltages was by changing the vector group of the unit auxiliary transformers in such way that the new connection will shift the voltage by another 150 degrees so that the total shift becomes 360 degrees which is equivalent to 0 degrees. 150 degree phase shift is possible with a DyN5 connection. Thus the only option was to reconnect the unit auxiliary transformers as DyN5 from their existing configuration of DyN11. This could be only possible by changing the connections of the winding inside the transformers.
We did not know if this was possible and practical. We contacted transformer manufacturer and discussed the problem. The designers of the manufacturer agreed to carry out connection changes inside the transformer at the site to change the vector group from Dyn11 to Dyn5.
The design practice for generator transformers of this service are generally of YdN1 type. However, with the wrongly manufactured bus duct, this was not possible and the change described above was the only solution possible. Nevertheless, using GT of YdN7 vector group has not resulted in any problems that can be attributed to this rather unconventional choice of the vector group.
The generating unit and the transformer are running satisfactorily since over 25 years.