Integral Transformer Protection Elements - Alarm vs Trip

[rockman7892]

I am currently reviewing a new transformer control panel drawing and was curious which of the integral transformer protection functions are typically used for alarming and which ones are used for tripping upstream protection. Looking at the control drawings the following protective functions are supplied with the transformer and are avaliable with NO and NC contacts:

26Q1 - Top Oil Temp Alarm (60deg C)
26Q2 - Top Oil Temp Alarm (90deg C)
71Q1 - Liquid Level Low
63P - Pressure Relief Tank
63VP-V - 3.5 Vacuum Alarm
63VP-P - 7.0 Pressure Alarm
63 - Rapid Pressure Rise Relay Tank

For this application the transformer is protected on the primary by a fused switch as well as an upstream 13.2kV breaker that is dedicated for the transformer feed. The customer has stated that he only wants the 63 (Rapid Rise Relay) to trip the upstream breaker through an 86 and just simply alarm back to SCADA for the rest of the elements.

I'm curious to hear from others if this is typical, or if other elements besides the 63 are typically used to trip upstream protection.

Thanks

 

[rustyjvan]

Normally, we common all of the 63 together and send them to the 86 (how your customer specified.) With the others we will send them to an HMI, or Annunciation panel to provide alarm indication. Depending on if your client has a DCS system the alarms may split and go to the DCS.

 

[dpc]

A fairly common practice. Tripping on the vacuum/overpressure switches is generally not done. Liquid level low - we usually have two levels - alarm and trip. Top oil temp - be good to have a trip from something if the oil gets hot enough. If the alarms are tied back to a SCADA system that someone actually responds to 24/7, I don't see much problem.

The sudden pressure relay is an incipient fault detector, so you definitely should trip on that.

Also have to consider the trade-off between transformer protection and down-time.

 

[KuanYau]

To rockman 7892

Your question "which of the integral transformer protection functions are typically used for alarming and which ones are used for tripping upstream protection".

1. 26Q1 - Top Oil Temp Alarm (60deg C)
26Q2 - Top Oil Temp Alarm (90deg C)
71Q1 - Liquid Level Low
63P - Pressure Relief Tank
63VP-V - 3.5 Vacuum Alarm
63VP-P - 7.0 Pressure Alarm
These protection functions are [typically for alarming].[No immediate tripping is needed}. Actions may be taken after a cup of coffee (minutes later), usually can be tolerated. Transformers usually take hours to reach the next [dangerous stage] where [immediate tripping is needed], hours later.

2. However, 63 - Tank Rapid Pressure Rise Relay is different from the above schedule 1.
When 63 operates, [an immediate trip action must be taken without any delay].
Attention: 1. [The immediate trip] signal shall (MUST) trip the [up-stream transformer primary winding breaker].
2. Tripping the [down-stream breaker alone/only], is (NOT) effective; and is (NOT)acceptable.
3. Whether an inter-trip is needed [from the up-stream breaker to the down-stream breaker] is a different matter.

3. Caution: No consideration for trade-off between transformer protection and down-time [when 63 operates]. [An immediate shut-down is strongly recommended] to save the transformer before a likely fire and explosion.

 

[prc]

63 P also must be connected to trip as usually this will involve oil spilling.

 

[ScottyUK]

63 and 63P are almost always triggered by a major internal fault. Trip primary and secondary sides of the transformer.

26Q2 is typically overload-related so in principle tripping the secondary breaker should be an adequate response, but if the conventional 51 inverse time relay hasn't already operated then there is likely a problem with the secondary breaker or its protection: I would trip primary and secondary on a 26Q2 signal.

The others are generally used as alarms to initiate human investigation.

 

[DTR2011]

I agree with others on what should alarm and trip. One point about the "63 - Rapid Pressure Rise Relay Tank" and perhaps it is wired into the scheme already. I've generally seen the "63 - Rapid Pressure Rise Relay Tank" tripping a 63X (Aux Seal In) Relay, which then trips the transformer lock out relay.

The purpose is several fold.

#1 - The 63 - Rapid Pressure Rise Relay Tank device trips the 63X, via both NO and NC contacts, which is for security.

#2 - The 63 - Rapid Pressure Rise Relay Tank contacts are generally not rated for tripping duty, while the 63X has contacts that are.

#3- The seal in reset is located in the transformer control cabinet, which forces someone to go out there and inspect the unit!

There are some discussions on this forum on whether to supervise the 63X trip with an O/C relay element. I have not seen this in installations in my area.

In newer installations, I have seen both a local Annunciation panel at the transformer control cabinet and some kind of fiber optic connection that takes the individual alarms back to SCADA.

Older installations have a local Annunciation panel, which has a set of contacts (any / all alarms) back to SCADA, being a general alarm.

I've seen some trip on pressure relief, and some alarm.

The temperatures stated for 26Q1 - Top Oil Temp Alarm (60deg C) & 26Q2 - Top Oil Temp Alarm (90deg C) seem close to what is used for stage 1 and 2 for cooling on a unit I recently dealt with.

 

[ScottyUK]

You absolutely do not want to alarm on PRD operation. If the PRD operates then something is desperately wrong inside the tank and the PRD must trip the transformer off the system. Anyone who contemplates assigning a PRD as an alarm should look at the aftermath of a transformer explosion and then decide whether it is an 'alarm' event or a 'trip' event.

Aside: a PRD with a high oil flow rate forms a crude atomising nozzle which makes the ejected oil much easier to ignite. I think Qualitrol provide some sort of pipe or duct to direct ejected oil away from the transformer and it exists as liquid rather than droplets. Something to consider if specifying larger transformers which can generate large oil flows.

 

[dpc]

If the PRD operates then something is desperately wrong inside the tank and the PRD must trip the transformer off the system.

Or it operated for no apparent reason. We have seen this. We normally only alarm for operation of PRD. But the vast majority of these transformers are non-conservator type. If the transformer has exploded, something else is going to trip anyway.

 

[prc]

PRD must be wired for trip. It is required atleast to check whether operation is due to internal fault or due to no apparent reason or due to some other reasons.It is true that PRD may not operate at all even with severe internal faults and or tank explosions. I had experienced such cases.

Some of the reasons for PRD operation with out internal fault- with conservators with air cell: more oil filling in to conservator than indicated by ambient temperature,over loading, valve between tank and conservator kept closed.

 

[DTR2011]

The OP has not stated if the transformer is a sealed type or conservator type, which I suspect is why there is debate over the tripping issue on PRD. I gathered from the
63VP-V - 3.5 Vacuum Alarm
63VP-P - 7.0 Pressure Alarm

with the tank mounted RPR and lack of Bucholtz relay that this was a sealed (N2 Blanketed) unit. As has been mentioned by others here in North America, on these types of transformers, and in my experience in a utility application, PRD is an alarm function. This is generally a decision made by the utility, based on company standards, which many times are based on user experience. Having SCADA alarms and a local maintenance crew available to dispatch may also influence these decisions, as well as other indicators / alarms available from SCADA.

As part of a substation upgrade project I am on now, I noticed that one of the 84MVA (1970 Vintage) transformers indicated tripping the 86T on High Liquid Temp (26Q) on the single line diagram. Further inquiries to the client revealed that this was a practice years ago and the wiring / tripping practice (company standards) had long ago been modified. In other words, the utility had determined through operational experience with this type (Sealed unit) of transformer that Liquid & winding Temp alarms were sufficient.

 

[rcw retired EE]

One Canadian utility in British Columbia only alarms on sudden pressure on their transformers. They have a history of multiple false trips duirng siesmic events. The false trips required manual resets at remote sites, delaying return of service.

Some engineers attributed teh flase trips to the pressure fluctuations caused by oil sloshing during a quake. Others said it was the relay trip contacts slapping closed.

But most of our clients would rather trip than have a transformer go from square tank to round tank or no tank. I might accept a sudden pressure trip supervised by differential or overcurrent.

 

[dpc]

A lot of industrial facilities disconnect the sudden pressure trip if it ever misoperates. We have seen a lot of them taken out of service. But most of them are not properly tested or maintained so the nuisance trips are partly the result of that.

But supervising a 63 with an 87 or 51 gives up a lot of the reason for having the 63 in the first place. It can be much faster and more sensitive than the differential trip for low current faults that are below the pickup level of the differential (and far below the OC).

The sudden pressure device is the one thing we pretty much insist on being a trip function.

 

[stevenal]

[URL unfurl="true"]https://www.beckwithelectric.com/docs/tech-papers/xfmr_en.pdf
[/url]

Good discussion beginning on page 6. He is speaking of big EHV transformers, though.
Our sole misoperation occurred when we found the valve in front of an SPR in the closed position, probably since the day the transformer was installed. The proper remedy for this situation is not to simply open the valve. Oops.