Ammonia Containment

[AndrewU]

Hi, I'm new to the ammonia refrigeration and just would like to get some comments regarding this. The ammonia equipment room at present is located on an open space adjacent to an existing building and an open lot on other side. There is a roof to shelter it from the rain but the area is otherwise open. We are planning to construct containment around the vessels to contain the liquid in case of leak, etc. Also we initially thought of constructing walls around the area and install fans on the roofing to disperse the gases to higher elevation. However, there was also an idea to install mists nozzles to absorb the gases in the room in case of gas leak.
I am not sure as to the correct practice or codes (if any) apply here. Also, if we use water to absorb the ammonia gas, how do we get rid of the water ? Can we just dispense this into the sewer ?
Can the more experience guys help by giving some comments and especially references where I can find the correct design / procedures ?

[AndrewU]

Hi, finally decided to go ahead with the construction of the walls around the machine room area and install mist nozzles also based on the suggestions from others outside this forum site.
As we will of course install exhaust fans for evacuation of the ammonia in case of leak, just wanted to ask if I can use this as normal ventilation fans for the room. Or does current practice require these to be separate and I have to provide separate fans for ventilation ? Also for fresh air, can I just install grilles on the walls or does it necessarily have to be a supply fan ?
For the roofing, I was thinking of some polycarbonate sheets to allow more natural light in. However, it seems polycarbonate and ammonia is not very compatible. If anyone has used this before, is there very appreciable deterioration ?
Would welcome what the others think regarding this idea.

[Magoo]

AndrewU
Hi, do you fully understand the product you are dealing with.
Forget any misting sprays anywere closely associate with electricity, this will create sparks and an explosive reaction, plus the fans should also be explosion proof.
Suggest an alarm monitoring system that isolates all power to plant area, and triggers release of remote power supply. That also starts extract fans [EP ] to clear vapour.As for spraying water on liquid ammonia in a bunded area ,say around vessels etc., you will create a bigger cloud of vapour than you can handle.
Check your local health and safety chemical hazard safety requirements.

[US Iceman]

AndrewU
Hi, do you fully understand the product you are dealing with.
Forget any misting sprays anywere closely associate with electricity, this will create sparks and an explosive reaction, plus the fans should also be explosion proof.
Suggest an alarm monitoring system that isolates all power to plant area, and triggers release of remote power supply. That also starts extract fans [EP ] to clear vapour.As for spraying water on liquid ammonia in a bunded area ,say around vessels etc., you will create a bigger cloud of vapour than you can handle.
Check your local health and safety chemical hazard safety requirements.

Agreed. People seem to think spraying water on liquid ammonia is good because the ammonia is anhydrous and will absorb the water. To some degree this is true however, the water will be warmer than the liquid ammonia laying in the containment area. Once the water hits the ammonia the ammonia will begin to flash off and create a vapor cloud. That's the dangerous part if the liquid has been contained.

You need to reevaluate your goals...

[AndrewU]

Sorry for the misunderstanding. I was thinking more of the water spray absorbing the ammonia gas leaks rather than liquid. There should be manually operated valves to open the water when there is gas leaks (purpose is to absorb the gas). This would be near valves areas that might release gas leaks and not on electrical power. There is no point in spraying water unto liquid ammonia. If there is liquid ammonia leak, this would be contained by bunded wall containment around the vessels and pumped to containers.
Thanks for the suggestion on the alarm monitoring system. We were initially thinking of installing ammonia sensors on selected areas to trigger an alarm and the exhaust fans. I will include your comments when I discuss with the supplier and installer here.
What about my questions on the air intake and polycarbonate, would you have some comments there also ?

[jaysephus]

personal stuff sorry

[hookster]

Hello Andrew
Our company successfully fitted misting sprays to the extract fans of plant rooms for the UK food industry for discharged vapour. The discharge causes an exothermic reaction with the small misting droplets thereby increasing evaporation and air stream is delivered at a high velocity to project ammonia vapour into the air for natural degradadtion. There is no water pooling and on tests tell tale ammonia releases reduce the panic factor associated with the refrigerant.

All the plant rooms are fitted with low and high level sensors with remote alarming. At low levels of detection fan only air change is instigated and should (highly unlikley on a well maintained system) liquid ammonia be released. The high level sytem will activate the misting sprays on the discharge fan.
As stated fan is designed with motor out of the discharge air stream.

Your poly carbonate sheeting will be fine with well designed ventilation as you should never achieve the levels of ammonia for incompatibility with good design.

The bunding of the plant will contain liquid releases and ventilation will remove evaporated ammonia.

[AndrewU]

Thank you for your suggestions Jaysephus and Hookster. I used the EN 378 (Refrigeration Systems & Heat Pumps - Safety & Environmental Requirements) formula to calculate. Ventilation Requirement, m³/h = 50.4 x (Refrigerant Charge, kg)^⅔ or 15 x (Room Volume, m³), whichever is higher. The result based on room volume is lower. There are two ammonia system (operated separately but interconnected) and I used the larger refrigerant charge. The capacity come up to about 12,500 m³/h.
Could you clarify on the "misting sprays on the discharge fan" remark ? We are installing them now on the compressor area and valve stations.
I agree with your comment on the polycarbonate and will be giving it a try.

[sterl]

Simply adding humidity to the air is not going to do much for the "panic factor" associated with a substantial discharge concentration; and depending on the resultant buoyancy of the gas mix, could cause the gas mix to go to ground instead of rising. Ammonia vapor will certainly dissolve into a water spray; so then you have to able to deal with the consequential aqua-ammonia.

The IIAR bulletin 111 deals directly with the ventilation of machine rooms: and covers both emergency and non-emergency ventilation. Non emergency ventilation is largely established by the size of motors and extent of "hot" surfaces within the machine room; targeting a temperature rise from ambient less than 18-deg. F. That same criteria is used by ASHRAE, IMC and EN378.

Emergency ventilation rates for R-717 room vary somewhat....It is usually determined by the largest refrigerant charge of a single circuit though there is a bare minimum aimed at avoiding (the surprises of) localized high concentrations.

In cold climates, conventional ventilation needs to be managed or the machine room temperature will drop too low and all sorts of things will quit working....

There is no reason not to combine the two under conventional ventilation conditions so for larger machine rooms, the last step of conventional ventilation could be one of the emergency fans....

In an all-air arrangement, emergency fans should be Spark Proof with drive motors out of the air stream and high plume discharge arrangement. (So Called "Chem Fans") Should the overall arrangement incorporate some form of air washer as AndrewU indicates, the high discharge velocity is contravened but the advice in Bulletin 111 about proximity to door openings etc. should be considered....As well as what happens to the aqua.

After working on a short valve train, I had a bucket of low concentration Aqua sitting in sunshine on a rooftop for about 4-hours. About a half gallon got spilled at the top of the ladder as we packed up that day...which was also near a downspout from the metal roof. We calmly went down the ladder but were contacted almost immediately by the plant electrician who had triggered the alarm to evacuate the plant and now wanted to know how long the repairs were going to take....

[brian_chapin]

Spill some of that aqua ammonia on a black rubber roof - you'll be amazed how quickly the locals panic!

[liamriley86]

Hello, I work on Ammonia quite often almost daily. If you feel you have to go to these high lengths to control a leak of the system you are obveously expecting a leak and dont trust its condition. I personally will not work on a system that is serverly detereated as it doesn't take much to kill you. I would stand back if I were you and concentrate on repairing the system to a trustworthy standard. Or just condem it if its badly run down beyond repair. If you have to ask what you should be doing with ammonianated water should you even be working on it? You should never through it down a drain some of these sewers you think are sewers are river drains for rain water, it will cost thousands in clear up and kill allsorts of wildlife. If you did throw it down the sewer you may be looking at fly tipping crimes and a massive fine. Controlled waste disposal is the way mate. As for throwing water on the ammonia well that would make it boil and if you didn't have an escape mask you gonna know about it. Not to mention the explosion hazard at 17% -22% concentration. Hope this helps and never work on it alone if you do prob think about a draeger unit to measure your exposure limits!

[HallsEngineer]

Simple, Detection system should shut off all incoming, non - EX certified electrical items. Vent should be to air as high as you can get it. There should be no un-bunded drains in the plant area as this is highly dangerous. The key is not to spray water all over the nicely subcooled liquid as it gets rather aggressive and starts spitting at you. It shoud be contained and pumped to a tanker if the leak is that bad. There is a system where you can pass the leaking gas through a "scrubber" which will clean the air going out of the plant room during a leak. This will need to be fitted with a bund and a solenoid that activates on a leak situation. Whereas if you leave the plant room "outside" non of this matters! as there are no safety rules goverining external ammonia plants as the Regs state "Special Plantrooms".

[AndrewU]

Thanks for info, HallsEngineer. I am not very sure on the term "outside". If the area I am referring to here occupies the space (boxed in you might say) between three concrete buildings (warehouses with different temperatures) and one side adjacent to the perimeter fencing and has a roofing would this still be considered outside ? There is a small alleyway between the buildings that serves as entrance to this "plant room". Anyway, there has never been any leaks that occurred. This was just a company decision to contain it considering the risk with our neighbors.