What should transformer Oil and Winding Temp Trip?

[veritas]

I have generally followed the belief that the oil temp trip (OTT) and winding temp trip (WTT) trip only the LV breaker of the transformer. Should the LV breaker not open within a certain time, then a trip is sent to the HV breaker.

Reasoning has always been that OTT and WTT are not inidcative of trfr unit faults (for that there is Buchholz, diff, pressure relief valve, REF, etc.) but rather overloading (OL). The remedy is thus to shed the load if the trfr is overheating.

This also cures the hotspots as these are related to the current flowing in the windings. With current gone, it is safe to have HV breaker closed and leave trfr energised.

Interested to hear other views/practices.

Thanks.

 

[pwrtran]

It is common to use oil and winding temperature for cooling control and caution alarms rather than to trip the LV breaker.

 

[davidbeach]

Often all there is to trip is the transformer lockout relay. That lockout relay will then trip both the low- and high-side breaker(s).

 

[veritas]

I generally do not route the OTT and WTT signals to the lockout relay as they do not indicate a faulted trfr. Usually there are OT and WT Alarms as well.

Even if the trfr has a hot spot, the remedy is to reduce/remove the load. Should the hotspot degenerate further localised gassing will eventually collect in the Buchholz leading to a Buchholz Alarm which then should initiate further investigation.

Regards.

 

[ppedUK]

veritas,

I'm with you on that one (tripping the LV that is), as high OT or WT is usually an overload condition and tripping LV saves calling out a SAP in the middle of the night to reset a HV trip, but some installations don't have the luxury of an LV breaker on the secondary of the TX, so then it has to be a HV trip.

Depending on the criticality of the installation, some companies alarm only on high oil or winding temp and allow time for the operators to reduce load.

 

[dpc]

We trip the lockout relay. The high temperature trip contacts at the transformer are latched in so we can tell what caused the lockout trip. In many utility distribution substations there is no low side main breaker, so tripping the high side is the practical option.

 

[slavag]

Only trip of both sides

 

[Marmite]

I'm with you Veritas, but I'm from a UK background. One of the guiding principles of the design of a protection system is that it should be selective, so if you have a transformer overheating from an overload condition that can be cleared by opening the LV breaker, then just open the LV breaker. More often than not in the UK the main transformer has a tank mounted auxiliary transformer or an earthing transformer associated with it with an auxiliary winding providing the station LV supplies, which would be another reason not to trip the main transformer unnecessarily.

Regards
Marmite

 

[pwrtran]

Folks, what size of transformers are we talking about? Are transformers scada monitored? Trip HV or LV breakers? well, do you let the transformer overloaded and the temperature goes up to the ceiling then deal with it?

 

[slavag]

From my point of view, size or type of transformer are not important.
Same issue in case of some distribution transformer with DGPT/DGPT2 relay.

Overload situation must be solved on the low level, overload protection, thermal image protection, etc of the LV breaker.

 

[bacon4life]

In our distribution substations, instead of trip either the HV or the MV bank breaker, we trip one of the 12 kV feeders nameplate temperature. This scheme was developed before we had SCADA: Dropping just one feeder reduced the loading by about 25% and resulted in customers notifying the dispatcher about the loss of power. Upon investigating, field crews would check drag hands to discover the overloading. Now that we have SCADA, dispatchers take action to reduce loading well before the overload reaches tripping levels. Tripping the HV side also removes station service power in some cases. We usually retest a transformer whenever a lockout rolls, so hitting the lockout on an overload may result in extra downtime.

 

[veritas]

I guess htere's somethng to be said for the oile and winding temp alarms as well. I have seen instances where these are not monitored and other instances where these are relayed to the operators. Intention is that the operator(s) is able to initiate corrective action before an OTT or WTT occurs.

Thanks for all the replies. They confirm that I've not strayed off the straight and narrow regarding OTT and WTT.

Regards.