LEL measurement within a process tank

[leakyseal]

Does anyone have experience installing LEL and O2 sensors such that they are measuring the environment within a tank? The tank is blanketed with Nitrogen (Fisher ACE controller on the inlet side, Protectoseal PVSV on the outlet) and we require the sensors to confirm the blanket.

The further twist is that the tank is filled with agitated polyester resin and there is a chance of splashing and contamination. Most of the vendors I have spoken to are being pretty cagey about supporting this application -- I can't be the first person to need to instrument a tank in such a fashion?!

 

[Compositepro]

This would typically be done with a sampling line that would have a continuous small flow (with appropriate filters and conditioning)to remotely located sensors.

 

[MJCronin]

leaky...

You might not be the first person, but you may be within the first ten...

It is my understanding that a plant operator purges the tank, establishes an atmosphere that is non-explosive, then bleeds N2 into a tank as necessary. The environment inside the tank is controlled and there should be never be a source of ignition present.

The requirement for continuous monitoring of LEL and UEL assumes (in my understanding) a changing environment where a source of ignition might be present. ( Such as outside of the tank)

Remarks anyone ?

A "chapter and verse" quote from some NFPA standard probably exists that addresses this matter....

My opinion only....

-MJC

 

[leakyseal]

Per CP's response, this looks like the approach we will end up taking. Is it verboten to talk about vendor rec's on this board?

MJC, it's the changing environment that's the issue, plus failure detection of the blanket gas. The tank inventory is constantly changing due to different batching formulations etc., sometimes there is a solvent rich environment, sometimes not. The risk is especially present during spraydown operations with e.g. styrene. If area nitrogen goes down, or the local regulator fails, it is intended that the LEL sensor will provide a warning.

 

[JLSeagull]

We have considered something like you describe. Few LEL detectors are suitable for a tank. Liquid exposure would be bad. The measurement levels are not adequate for our needs. I think that catalytic bead sensors require oxygen in the atmosphere.

We are applying extractive analysis and knocking moisture out in the sample system for some tanks. The analysis includes oxygen and including IR hydrocarbon detection for component groups. This is a few orders of magnitude more costly than an LEL detector. The tanks are nitrogen purged.

 

[jamesbanda]

Leakyseal,

You are probably too far down the line to do this.. but

could you do the following.

1. Operate at slightly positive pressure say 0.1 barg and use pressure control to ensure the atm.
and
2. Have a high MAWP of the vessel, therefore reduce the severity of an explosion significantly. (you should do this anyway if operating in such a manner above flash point).

Other comments.
> Could you do the same as the o2 monitor by flow and level. My experience with 02 monitors on process duty is variable. in one case they were horribly expensive and we used for 1 month fought poor reliability and eventually put in another engineering solution. in another case no problems..

 

[schnipp]

We have a system here on an enclosed filter and dryer that measures for O2 only. I wouldn't worry about measuring for LEL since if there in no oxygen, no kaboom.

We have a slipstream off of the positive side of a blower that measures for oxygen and returns the stream to a lower pressure area. I'm sure you could use a small diaphragm pump or something of the sort to pull off of the tank (if not pressurized). We have had some trouble with the existing unit, mainly the reactant/catalyst inside has spilled a couple of times making it necessary to replace the unit.

 

[JLSeagull]

Some materials such as Ethylene Oxide may explode in the absence of external oxygen. This is a potential component in the tanks that were of interest where we considered LEL detectors and oxygen analyzers. The IR analyzer can detect the presence of EO and similar material groups.

 

[schnipp]

Yep, you're right. EO has the ability to explode from 3% to 100% concentrations. There are specific "probes" that can detect EO only, or am I daydreaming?

 

[JLSeagull]

We plan to use an infrared analyzer. I am more knowledgeable about gas chromatographs than IR analyzers. I think that the wavelength band that would detect ethylene oxide also detects olefins like ethylene. However, I should likely take the same interlock action upon detecting the interference compounds too. If you know something EO specific let me know too.

Thanks.