I need possible explanations on quench tank oil fire

[Dchris5212]

I am a foreman at a heat treating plant and I was wondering if anybody can come up with an explanation for a serious situation I ran into a few days back. I was running a can-eng fluidized bed (small load of 8620 steel) 54 lbs divided into two baskets. The baskets sit on a long shaft that we lower into the heating zone full of aluminum oxide and after x amount of time take out and proceed to quench into a very large oil tank (roughly 500 gallons). We have a above oil agitation system (small pump that recycles bottom oil back to top through a hose that sits about 5inches from top of oil). We use AAA quench oil. Upon quenching we had a large flame come up through the quench area which is typically common, though this time the flame proceeded to engulf the entire top layer of the quench tank oil and continuously burn getting stronger and stronger. This has never happened and we have quenched out loads much much larger and heavier. Fire department was called in to resolve. Only thing we keep coming up with is possibly the oil completely broke down though I find that to not be to realistic a situation. Also I think the aggitation system also played a role in feeding the fire. Any help or explanation why this might've happened would be much appreciated.

Thank you

 

[Compositepro]

The oil in the tank has to be kept well below its flash point. As you say, it is common for flames to be present when a hot load is lowered into the tank. But the fire should go out as soon as the load is submerged in the oil, unless the top surface of the oil in the tank warms to above the flash point. Recirculating the oil from the bottom of the tank will help a little. But that cool oil that you are pouring in at the top will sink right back down the bottom of the tank without cooling the top layer of oil much

I have not worked with quench tanks, but the only method that makes sense for cooling the top layer of oil is to have the hot oil overflow a weir around the top lip of the tank into a sump where it is pumped back to the bottom of the tank. There it will mix with colder oil as it rises in the tank. Sort of like a swimming pool skimmer.

 

[arunmrao]

In order to avoid oil fire, keep it covered, so that you cut off the supply of oxygen. Also, if you could purge Nitrogen, it helps. But always select an oil with a higher flashpoint.

Ensure that oil is well circulated and that your heat exchanger is well designed and working well.

These are a few basic precautions to be taken.

 

[LittleInch]

Interesting.

AAA quench oil appears to have a flash point of >340F / 170C.

Really not that high when dealing with large lumps of metal at??temperature.

So what is your "recirculation" rate? You have a 500 gallon tank which recircs from the bottom to 5 inches below the top of the oil? My suspicion is that it's really quite low and not enough to adequately mix the oil and obtain a near uniform temperature and in fact is pumping the wrong way around as composite pro says. Even if this is above the oil, it may not be enough to mix very well.

Does it visibly agitate the top 5 inches or allow a layer of hot oil to persist above it?

Do you take any temperature measurements from top to bottom? Can you take measurements before quenching?

How does the oil cool down over time? Are there any oil coolers?
Do you have any other standard process for mixing the oil prior to quenching?

Did this particular quenching follow a previous one at short notice?

Having an inert gas flood mechanism in your tank (CO2?) would seem to be a good plan to me and I'm surprised this possibility (top skin fire) wasn't already thought about.

Any pictures or drawings helps a lot for us to visualise what you can see in front of you but we can't.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MJCronin]

You have water at the bottom of your quench tank that amplifies your problem

At what temperature do you maintain your quench tank ?

https://en.wikipedia.org/wiki/Boiling_liquid_expanding_vapor_explosion

MJCronin
Sr. Process Engineer

 

[mfgenggear]

would it help to have a circulating system to control oil temperature?