How Not to Obtain a Tank Sample

[bimr]

What are these people thinking about?

Steve Mathy, president of ALM Holding Co., which acts as the parent company to Midwest Industrial Asphalt said the worker was following protocol when the explosion happened. "The operator checked the gauge inside the tank, which indicated there was two feet, two inches of product," he said. "He then proceeded to draw a sample, since the gauge indicated the sample port was submerged in the liquid."

When the worker attempted to draw a sample, nothing came out of the port. Mathy said the worker then lit a torch to unclog the port, which is proper protocol. Instead, the gauge was faulty and not submerged within the mix of asphalt and diesel fuel. Mathy said the torch ignited vapors which traveled through the port and into the tank where the explosion occurred.

 

[SNORGY]

I can't imagine any scenario in which this was a good idea.

 

[marty007]

Some simple redundancy may have been able to prevent this. With two gauges, if one doesn't match the other, you have a problem. Better yet, have the two level measurements use different technologies, so that both may not fail at the same time. There are many technologies out there to measure liquid level:
[ul]
[li]float type[/li]
[li]radar[/li]
[li]ultrasonic[/li]
[li]differential pressure[/li]
[li]sight glass[/li]
[li]level switch on low-low[/li]
[li]many many more...[/li]
[/ul]

How much more would a second gauge have cost? How much has this accident cost?

Also, as SNORGY pointed out... who thought it was a good idea to make putting fire to a flamable product nozzle into proper protocol?

 

[JStephen]

I'm not sure how many of the level gauges normally used are compatible with heated asphalt service.

Taking a torch to a nozzle that is assumed to be plugged with asphalt wouldn't be my first thought. But then again, you can weld onto and hot-tap a gasoline tank in service, and that's not something that would occur to me to try, either.

Some of the heaters they use for asphalt tanks are combustion heaters, with a fire tube extending into the tank and a stack right up the outside of the tank. It's not something you'd do with a normal flammable-products tank, either.

 

[MJCronin]

With about five minutes of internet search time, you can find out that radar based gauges and level switches own asphault applications.

Yes, these devices are expensive....... too bad...!

No, you cannot let the maintenance department beat on them with hammers when there is a problem...

http://www.durablecontrols.com/applications/other/CS0143 FTL71 in asphalt tanks.pdf

http://www.tankgauging.com/whitepapers/AsphaltTankGauging_WhitePaper_1109.pdf

This torch "proper protocol" that you mentioned.... is that written down anywhere ?

Is there a signature at the bottom of this torch protocol document ?

Or... is this just something everybody in the maintenance department has been doing for a long time...???????



MJCronin
Sr. Process Engineer

 

[bimr]

One would think that this facility should have a hot work permit program and the area should have electrical classification. Yes, diesel fuel is not nearly as flammable as gasoline or other common fuels (such as ethanol or propane) but it can catch fire and can be very difficult to extinguish.

Diesel fuel MSDS sheets carry warnings.

Storage Procedures
Keep away from flame, sparks, excessive temperatures and open flame. Use approved vented containers. Keep containers closed and clearly labeled. Empty product containers or vessels may contain explosive vapors. Do not pressurize, cut, heat, weld or expose such containers to sources of ignition.

http://www.hess.com/docs/us-safety-...udingultralowsulfur_diesel(ulsd).pdf?sfvrsn=2

 

[SNORGY]

I did think of GWR (Guided Wave Radar) level detection in this application. But I didn't think of using a torch o unclog a nozzle. I guess I am not innovative enough to try something that smart.