Tap-Changers in Generator Transformers 4

[prc]

In my country Generator transformers are provided with DETC (de-energized Tap-changers). In case of transformers for thermal sets (50-800 MW) tap-changers with +- 10 % tap range(9 tap) were used to be there. Later this range was reduced to 5 tap with +- 5 % range as these tap-changers are never used once the unit is commissioned. From transformer manufacturer's angle, it is better to reduce the tap range and if possible eliminate it altogether for higher reliability. I understand in UK and Germany, GSUs are provided with OLTC with wide tap range. My understanding is they use high impedance transformers and reactive power control is done by operation of OLTC. When DETC is used, reactive power control is by adjusting the AVR of generator.

In Hydro power stations, our major utility eliminated DETC and now procuring GSU without tap-changer. Is there a difference in reactive power contribution by thermal and hydro sets? What is the practice in other parts of the world? My point is if an accessory is never used in its life time, why it should be provided in the first instance itself? Requesting guidance from operating engineers in the field.

 

[crshears]

Never used? Not necessarily true...

From what I've been told, going without GSU off load tap changers [OLTC] / DETC may or may not be a good idea, depending on where the generating station is and the likelihood of any significant changes to the grid in the area; the thermal plant I worked at a number of years ago was provided with DETCs in all MOTs, and I don't recall a tap change ever being done during the six or so years I was there...but I do recall that virtually every more senior operator and power system engineer I spoke to about them said the extra expense up front was well worth being able to tweak the MOT ratio to match the prevailing grid conditions over the course of time...

For instance, one of the transformer stations that was only a few km from the GS was initially supplied exclusively by OHL, but in later years only one OHL remained, the rest having been transferred to new HV XLPE cables, which produced a significant amount of additional reactive power in the area and improved the local voltage profile. It's my understanding that if the Province had not decided to decommission all coal fired generation within Ontario, the DETCs in the MOTs would have been moved to a different tap in order to re-balance the reactive production/absorption capabilities of any running units.

As far as I've learned, differences in reactive production and absorption capabilities between thermal/nuclear high-speed hydrogen cooled turbo-generator units and slower speed air-cooled hydraulic units are not inherently better in one or the other; initial design commonly determines these capabilities, so, essentially and all else being equal, you get what you pay for.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[RRaghunath]

Large generators have AVRs with +/-5% capability to vary the voltage. Whereas the grid has allowable voltage band of +/- 10%. In such a case there will be situations when the generator AVR will not be able to match the voltage with that of the grid for synchronising the generator.
Another benefit with OLTC is the possibility of maintaining generator power factor on lagging side, despite the grid conditions, by simply altering the transformer OLTC position. The generator field is stronger when the machine is on lagging and thus can withstand grid disturbances better.

 

[bacon4life]

GSUs around here tend to have DETCs rather than LTCs. We have adjusted almost all of our GSU DETCs at least once in the last 15 years. Kind of hard to imagine getting rid of both.

Distribution low side LTCs have to periodically run through neutral to eliminate coking on the reversing contact. Without a DETC, an LTC may not cross neutral under normal operation. Not sure if this is an issue with generation LTCs.

 

[prc]

Thank you all for the inputs. It seems the situation has to be decided by individual utilities. But at least, the tap range should be limited as small as possible for higher reliability. Remember that Three Gorges 800 MVA 3 phase GSUs had only 2.5 % tap range (DETC).Yes coking is one problem but probably this may not be a problem with highly refined transformer oils used in new transformers. I used to suggest to run the DETC or OLTC from one end to other during planned shut down to wipe the contacts. Is it a good maintenance practice? Followed any where? Of course it has to be done manually and winding resistance at all taps to be checked after this operation.

 

[rcw retired EE]

There are some good IEEE papers that explore how the Generator Step UP (GSU) Transformer's tap range and impedance affect the ability of the generator to provide or absorb MVAR's without exceeding generator or station service voltage limits.

IEEE C57-116 "IEEE Guide for Transformers Directly Connected to Generators" has some good graphs showing system voltage versus MVAR output for different impedances, tap ratios and generator voltages.

My favorite is a 1995 paper "Estimation of Reactive Power Export and Import Capability fro Non-Utility Generators", IEEE Transactions on Industrial Applications, Vol. 33 No. 6, Nov/Dec 1997. The authors presented graphs showing the MVAR capabilities of a multi-unit generating plant versus the utility voltage with varying generator voltages, parasitic loada and GSU impedancea.

At high system voltages an OLTC helps boost voltage to push MVAR's out. I never understood why a utility demands MVARS when system voltage was 107% except when it was an attempt to prove that the plant did not meet contract requirements for 0.95 lead to 0.90 lagging power factor over the full +/-10% range of system voltage.

If the system voltage is relatively stable and the system voltage can be predicted accurately when the transformer is specified, the plant can operate well without an On Load tap changer.

Of the 120+ generators we installed in independent power plants around the world in the last 32 years maybe four of them had On Load Tap Changers. Most had the standard 5x 2.5% taps with the nominal tap set at 102% system voltage. We offered tap changers but clients never wanted to pay for them.

The fun part was testing the plant's MVAR output as the utility swung the system voltage. To achieve maximum export, the system voltage has to be low to help suck the VARS. System voltage would rise at the plant interconnect as more MVARs went down the line. Twice, large areas blacked out when the system operators lowered their remote generators' MVAR output too far trying to help the plant. (That was pre-NERC).

 

[ScottyUK]

At the plant I worked at for 10 years each GSU transformer had a 33-position reversing on-load tapchanger. It rarely moved far outside the 13-15 range, with 17 being the nominal tap, and our reactive export was low because the grid voltage at our connection was typically quite high and injecting an excess of VARs would have worsened the problem. The requirement for the tapchanger was imposed the TSO.

At that station one OLTC failed soon after initial commissioning and the plant owners decided to accept monetary compensation and a reasonably quick repair to link out the OLTC completely, rather than a lengthy repair which would have left half the CCGT station unable to run without its steam turbine to take the steam from the HRSG's. The TSO found out years later and were reportedly very unhappy about the decision, although it hadn't made any noticeable operational difference other than the AVR was working pretty much flat out just to keep the machine VAR-neutral with the high grid voltage.

 

[prc]

Thank you Scotty. I was waiting for your inputs.It is surprising that in UK also OLTC is rarely operated in GSUs. My understanding was that OLTC requirement in UK and Germany was due to relatively high % impedance provided in the GSUs. Planners provide DETC or OLTC with wide tap range while in field it is rarely used or within a narrow tap range. This is the feedback I too received from the market I am familiar with esp with strong grids where grid voltage change is narrow.

 

[RRaghunath]

Erstwhile CEGB of UK had 'Modern Power Station Practice' in number of volumes explaining the engineering and operation of power stations. In one of the volumes there was good explanation on OLTC requirement for Generator Step-up transformers (Volume-4, if my memory is serving me right).
That's where the need for running the generator always on lagging power factor mode (irrespective of grid conditions) to ensure generator stability is discussed.
NTPC of India had technical understanding with CEGB those days (1970s and 80s) and used to specify generator step-up trasformers with OLTCs.

 

[FPelec]

The common practice here in Italy is to avoid OLTC on large (thermoelectric and hydro power plant, as well as synchronous condensers) EHV step up transformers, mainly due to reliability/short circuit withstand issues.
Underexcitation is rarely performed due to the widespread adoption of shunt reactors on the EHV grid.
Voltage regulation is performed by a centralized hierarchical system, operating on the AVRs of each plant.
However, renewable energy power plants, when connected to the HV network, use an OLTC, with a large (about +/- 10%) range. To my knowledge, these OLTCs are rarely used, let's say 100 manoeuvres/year.

Si duri puer ingeni videtur,
preconem facias vel architectum.

 

[prc]

Thanks Fpelec. Are you providing DETC (de-energized Tap=Changer) in step-up generator transformer?

 

[FPelec]

prc,
no, DETCs are not used in large step up units for power plants and synchronous condensers: those units are simile 2-winding transformers without regulation and a reasonably low impedance (12.5 % tipically); short circuit withstand is thus much less an issue without a regulating winding and the attendant connections to the tap changer.
PS
Actually, DETCs are seldom used here on HV and EHV units. HV/MV distribution transformers and renewable energy trasformers are instead almost always equipped with an OLTC.

Si duri puer ingeni videtur,
preconem facias vel architectum.

 

[prc]

Thanks FPelec !