Control room as refuge in emergency - wise? 1

[JOM]

Sometimes control rooms are the designated mustering point for plant personnel in case of emergency. (I don't know how common this practise is.)

Has the wisdom of this practise been examined and written about? Does anyone have examples of a control room providing refuge during a real emergency? Did things work well or were there problems? Do emergency drills reveal anything interesting?

I know of one case where it didn't work well.

J.

J.

 

[swn1]

No direct experience but I've been in a lot of plants.

First, the control room needs to be a hardened location to preserve as much control as possible in an emergency. So it is a natural refuge.

Second, during normal operations everyone in the plant has to check in at the control room and most of them "office" there. So it is familiar.

On the other hand, control rooms are often deep in the plant and if a secondary evacuation is necessary you will have lost time.

Finally, most incidents occur during turnarounds, not normal operations, and the plant is full of contractors. The control room is a poor muster point under those circumstances.

 

[JOM]

Useful comments, swn1.

In the case I know of, acidic fumes from a plant fire filled the control room, making it untenable. A poor refuge, especially for the injured personnel.

Another problem that you also aluded to, was the influx of numbers of personnel disrupting the operators' plant control functions.

I'm looking for any literature or experience on this. Does anyone play out "war games" to see what problems might occur.


J.

 

[schnipp]

In my past experience at a Chlorine plant, it worked well. The site was large (1700 acres, 1700 employees), and the chlorine plant was a small portion way in the back. The day staf included three engineers, manager, his secretary, and maybe two or three others (including me as the co-op). The break room was visible through a window, so non-essentials (like the co-op) mustered there, while ops and engineers worked out of the control room to regain control.

Air supply was regulated, positive pressure, and sourced from a high location on top of the building, away from gases that might cause issue.

Contractors went to the maintenance area which was considered out of the danger zone and their normal working area anyways.

If secondary evacuation was needed, we were all trained in the use of 30 min SCBA and 5 min escape masks.

 

[Guidoo]

If there is a possibility of having toxic smoke or gases near the control room, I normally specify that air intakes of HVAC system are provided with toxic/smoke/flammable gas detection that close dampers in the air intake, so that HVAC goes into internal circulation. As a rule of thumb, one person needs 1 m3 of air per hour (as explained in Lord Cullen Report for Piper Alpha investigation). You also need locking doors and a "air tight" design of the building to prevent toxic gases entering the building.

 

[JOM]

schnipp,

So, in your case, I think you're saying the personnel take refuge in the control room building but not in the actual control room. That would allow operators to continue with their control duties unimpeded.

With the case I mentioned, the plant fire burned instrumentation wiring and the PVC insulation produced HCl gas that travelled along the ductwork and entered the control room, which contained the operaotors, field personnel taking refuge and injured personnel. Not a happy situation.

Thanks.



J.

 

[JOM]

Guidoo,

What about fumes or gases entering through the instrumentation dusctwork? I know cables should have glands and seals but there are cases where gases entered the room. Do people validate the integrity of these entry points a matter of routine?

(Incidentally, the Browns Ferry nuclear plant fire resulted from technicians testing for smoke ingress to the control room through cable entries. They used a candle.)



J.

 

[Guidoo]

Good point. I have not seen this being considered in control room design (although this may be covered in control room design standards).

For modern plants I would anyway prefer to have the control room at some distance from the actual plant, also because you do not have to make it blast proof. With modern instrumentation and CCTV systems there is no real need to have the control room located inside the plant. Locating it at some distance is an inherently safe approach that I (and some reputated operating companies) prefer above mitigation measures such as the blast proof design etc.

 

[JOM]

G,

"With modern instrumentation and CCTV systems there is no real need to have the control room located inside the plant. Locating it at some distance is an inherently safe approach"

In which case, it can no longer be a mustering point, can it? Where would field personnel muster in an emergency? Would you provide a special building for the purpose?

Do you know if anyone has studied the performance of control rooms in emergencies? Anything published?



J.

 

[Guidoo]

Not having the control room inside the plant doesn't mean that it is at an unreachable distance. It can be ~200 m from the plant.

You could have 2 or more mustering areas (primary and alternatives) in the open air at opposite locations. Personnel moves to one of these areas depending on the location of the incident and wind direction. There should be sufficient elevated wind socks so that personnel can easily see wind direction.

There are no simple rules on how to do this. It depends on so many things (e.g. type of hazards, size of the plant, offshore/onshore, plant layout).

The war games you mention are actually done and named Escape, Evacuation and Rescue Analysis (EERA), especially offshore. They are scenario based. See

http://info.ogp.org.uk/RiskManagement/Process/Tools/EERA.html

or do a Google search on "escape evacuation rescue EERA"

The Lord Cullen Report is also a reference, because at Piper Alpha the muster area (dining room in accommodation block) turned out to be a vulnerable location (toxic smoke ingress). Recommendations are given for providing a Temporary Safe Refuge (TSR).

 

[Guidoo]

As a starting point for a literature search you could consider Frank P. Lees "Loss Prevention in the Process Industries". It covers topics such as survivability of escape routes and shelters with many cross references.

 

[JOM]

G,

Thanks for the lead on EERA. I see there are many publications about this. Good material. Mostly focussed on offshore. I'm more interested in land-based facilities. Offshore is a very special kind of workplace.

I'll have a look at Lees.

Do you think it would be useful to have reports on how control rooms performed during real emergencies? Could it be important enough to be made a standard component of the accident investigation?



J.

 

[Guidoo]

I agree that offshore is a special kind of workplace (mainly because of the expensive and therefore limited space available). Onshore best approach may be to simply move away from the hazard (taking wind direction into account). Offshore this is obviously not so simple.

However, there is a trend that the vast amount of safety related studies (safety cases) that are commonly done for offshore developments are now also required for onshore plants. I've seen this happening for example in Norway and in Abu Dhabi.

Maybe it is something for the UK Health and Safety Executive to commission studies to be done on this subject? On the other hand, I think there is sufficient knowledge available to design a control room that can be used as a safe refuge, e.g. based on offshore experience.

 

[Guidoo]

A search with keywords "control room" on

http://www.hse.gov.uk

gave some very interesting hits. For example:

http://www.hse.gov.uk/comah/sragtech/techmeascontrol.htm

,

http://www.hse.gov.uk/comah/sragtech/casepasadena89.htm

and

http://www.hse.gov.uk/comah/sragtech/caseflixboroug74.htm

 

[schnipp]

Its been 10+ years since I was in that plant, so some details are fuzzy (and some were never learnt).

The plant was built in the mid to late 70's, so PSM wasn't a standard yet. The control room building was built pretty stout, but I don't know how well it was sealed off from the atmosphere.

Knowing the way that company deals with risk assessment, they spent quite a bit of time discussing these scenarios (sp?). Other locations on site did muster outside, but had back up locations inside in case a green gas cloud went traveling down the road (which did happen every great once in a while).

In my opinion, it can be done, but you need to build the control room as a muster location and seal off the building as well as possible.

 

[JOM]

Lots of interesting comments on this topic. Thanks.

Can I reduce this "issue" to something simple?

What's the purpose of a control room? To control the plant.

Does that change in the case of an emergency? Course not.

So why use the control room as a mustering point? Why? Someone has said this is the place that people are familiar with (they "office" there). But in time of emergency, surely the control room should not be disturbed.

Maybe the rule should be: never disturb the control room during an emergency.




J.

 

[epoisses]

Sure. But your control building will be a blast proof building no matter what. If people are in danger in the plant and there is a blast proof building nearby, they will go theingre no matter what beautiful escape route signs you put up outside. At least I would. So you might as well make it an official muster point and save a couple of lives.

Just make sure the control room itself is not entered by other people unnecessarily, by its design (at the end of the corridor or I don't know what) and by everyone's everyday habits ("you don't enter the control room without a damn good reason").

 

[unclesyd]

In our situation the last place you would want to be is a control room. There control rooms are all located in the middle of the operating areas as most were built when the control systems were pneumatic. The one "hardened" control that we have sits in the middle of 2 reactor trains containing 100,000,000 gals of cyclohexane mixed with air, the same as Flixborough, England only a hundred times bigger. After that incident the building was no longer considered "hardened".

We use the control room as sign-in point for everyone entering the battery limits of any of our process buildings. In the case of an excursion there are designated assembly areas for each processing area where the sign-in books are supposed to be carried for personnel check off. Works during drills but so far hasn't worked very well in under an actual emergencies. The sign-in/checkoff book was in the assembly area but the people were elsewhere.

Of the several events that I've had the opportunity to participate in none have occurred during a turnaround or overhaul, all were process upsets under normal operating conditions. We had one medium size fire that occurred during a turnaround.

As alluded to above in the event of an process excursion people are going to move and not necessary in the proper or correct direction. Don't get in s their way.

Early in my career while doing some RT work there was a large NH3 leak next to the area where were working. We had a helper that was working with us and we had mentioned that the camera should never be left unattended. As the NH3 cloud was evolving my partner and I left telling the helper to get going. Running as hard as I could the helper came by me like I was standing still, as he passed he said that if I was going mess around I was going to get killed. At the time he was carrying the 90 lb Budd Camera.

 

[JOM]

epoisses,

I'm not advocating that doors be barred and entry prohibited during an emergency. Heck, no.

I'm thinking in terms of design and planning. Maybe the control room function and safe refuge should be separate design goals.

I've found an incident where 50 people came to the control room which was not helpful.

Unclesyd, that's a very funny story. When did your cameraman stop running?



J.

 

[Guidoo]

Looking back at my previous posts in this tread, I see that I used the word "control room", where I should have used "control building". If you would design the control building to act as a safe refuge, I agree with epoisses that there should be another room apart from the control room that has that function. Sometimes there is a small canteen or office rooms etc. in the control building, that can be used to provide shelter in case of an incident.