Transformer pressure relief

[glendale]

We are installing an 1800kVA 11kV transformer bellow deck in a concrete floating structure. The transformer room is force ventillated to provide cooling. In the event of an internal fault causing the pressure relief device to lift (8 p.s.i), would there be an increase in room pressure?
Does anybody have any experience regarding the possible effect of change in room pressure during this type of event?

 

[ScottyUK]

Depends on how restricted the ventilation to the outside world is, but yes you would expect to see an increase in pressure. Most faults which are energetic enough to lift a PRD will be evolving gas, sometimes at a disturbing rate. The actual volume and rate of evolution of gas is difficult to quantify because a fault could range from a small 'pop' to a catastrophic failure. In the latter case it isn't unknown for the PRD to be overwhelmed and the tank to rupture although this sort of failure is thankfully rare. Is this room sealed?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[glendale]

The room is force ventilated (fans at the room inlet with ducted air outlets) and is fitted with fire dampers that will automatically close on detection of fire/smoke, effectively sealing the room under fire conditions. My initial concern is that an increase in room pressure could possibly cause failure of the ventilation system itself (collapse of ducting etc). My HVAC consultant has no idea how to deal with the concept of a transformer pressure relief occuring!

 

[davidbeach]

You don't want to close off the exhaust from the room in case of fire. Unless you can dump in enough fire extinguishing gas (probably CO₂) to quench the fire as fast as the fault is cleared (and there has to be some place for the displaced air to go) you will have so much volume from the products of combustion that you'll blow something out. Best to have unrestricted exhaust to a point safely away anything else that might be damaged.

 

[glendale]

The transformer room has been designed with a "burn out" policy, i.e. if a fire is detected the room seals and in principle the fire is starved of oxygen, there is no fire extinguishing agent used. The room is 2 hour fire rated, and I am told any fire will self extinguish in minutes (if the room remains sealed). This is when I start to get nervous about the possible increase in room pressure due to the pressure relief device lifting.
An unusual design I hear you all say!!

 

[ScottyUK]

The photo in the link below is transformer which suffered an overpressure due to a massive internal fault. The PRD operated but couldn't pass enough flow to prevent catastrophic failure of the tank. The tank lid is well over an inch thick.

You are right to be concerned about HVAC ducting bursting but frankly you should be looking the integrity of the room itself if there is not free ventilation through a large capacity duct to allow rapid depressurisation in the event of a fault.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[davidbeach]

I doubt that you can build a room to contain the resultant pressure. Probably what will save (?) you is that all of the pressure will be generated before any mechanical damper can begin to close. Until the electrical fault is removed, the evolution of gasses will proceed violently. Check the relative volume of gaseous copper vs. solid copper and gaseous iron vs. solid iron. Consider that the arc is burning either or both of these at a temperature on the order of 35,000K (hotter than the surface of the sun). Now, you want to contain that? Once the fault is removed you are down to only dealing with the oil fire, but the oil may well be everywhere.

 

[TurbineGen]

Can we asuume that you have adequate transformer protective relaying? Surely in the event of a fault your transformer should trip within a few cycles. I doubt your ventilation system can seal off that quick in which case you only have to deal with the potential of burning oil.

It is my understanding that the largest rate of gas production during a fault happens in the first few cycles. This can rupture the tank with a violent force which even the pressure relief valve may not be able to handle. In this case, the oil may no longer be contained within the transformer and may be all over your room and on fire.

Perhaps a better solution is to install a rupture disk system? Check Transformer Protector Co (

http://www.transproco.com)

for information about this. We installed one of these TPC systems on a 129 MVA GSU transformer as a fire protection system. If an ounce of prevention is worth a pound of cure, this might be your best option.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[itsmoked]

Also dumping a bank of large CO₂ tanks into a room requires a vent or the room will burst.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[cbor]

Does your transformer have a conservator or is it a hermetic (sealed) type transformer with a nitrogen "layer" above the oil? What is the volume of free air in the room? Does your PRD close when the pressure in the transformer goes below 8 psi? If you are so worried about the fire, why didn't you ask for a transformer without mineral oil (with silicon fluid)?

 

[prc]

Since the trasformer is a small one with low system input capacity,the energy release will be small on a fault. Hence the type of pressure increase that you are fearing will not be there in the room.On many occasions,the pressure relief device itself may not act due to low pressure build up.However it can be a different situation if protection system ,including back up completely fails.