NH3 recirculation systems

[lraja]

Hi Guys

For a moderate capacity NH3 recirculation plant, say about 300 kW cooling, what is the best arrangement for feeding evaporator coils?, liquid pump based system or gas/compressor driven system …, could anyone guys reason things out for me

lraja

[US Iceman]

I like the mechanical pump version the best. It is more flexible for energy conservation.

The gas driven systems are dependent on constant discharge pressure for pumping the same volume of liquid in the same time frame. If the discharge pressure is allowed to reduce, the transfer rate can decrease.

Secondly, the check valves required in the gas driven system are more prone to problems due to the constant cycles.

I also think the mechanical pumps are more cost effective to operate.

[TXiceman]

I vote for a liquid overfeed or recirculated system using pumps. A number of mmanufacturers make very good canned or semi-hermetic refrigerants. The gas powered or CPR (control pressure receiver) systems are touted as the savior to refrigeration. The controls are more complicated and prone to problems (check valves and solenoid valves). If is also less efficient due to the re-expansion os the gas in the pumper drum or CPR.

On your pump recirculator drum you will need a level control and an oil recovery still or pot.

Ken

[taz24]

liquid pump based system or gas/compressor driven system …, could anyone guys reason things out for me

lraja[/quote]

The heat transfer with the liquid is better han with gas so I would go for the pumped liquid.

Cheers taz.

[nh3simman]

liquid pump based system or gas/compressor driven system …, could anyone guys reason things out for me

lraja

The heat transfer with the liquid is better han with gas so I would go for the pumped liquid.

Cheers taz.[/quote]


Hi taz24
I wonder about this statement. The fluid comes in as a liquid that boils vigorously to a gas. The liquid heat transfer is purely turbulent, so why should this be much better than boiling.

[US Iceman]

The heat transfer with the liquid is better than with gas so I would go for the pumped liquid.

Cheers taz.

I just noticed this comment after nh3simman replied to it as well.

The heat transfer is not appreciably different with gas pumped or mechanically pumped liquid to the evaporators.

The biggest concern is the cost of flash gas on the gas pumped systems and the check valves slamming all of the time. These systems are also dependent on a constant discharge pressure to maintain a constant pumping rate, so this could limit the potential for discharge pressure reduction.

Sorry taz.

[cha09]

I have 2 unit ammonia refrigeration system, each has capacity 300 kW, all of the systems use liquid pump, it is looks good.

[taz24]

I just noticed this comment after nh3simman replied to it as well.

The heat transfer is not appreciably different with gas pumped or mechanically pumped liquid to the evaporators.

The biggest concern is the cost of flash gas on the gas pumped systems and the check valves slamming all of the time. These systems are also dependent on a constant discharge pressure to maintain a constant pumping rate, so this could limit the potential for discharge pressure reduction.

Sorry taz.

It just goes to show I'm not too old to learn

I was of the understanding that flooded evaporator swere used because the heat transfer was greater than saturated liquid vapour evaps (DX).
The reason behind this was the whole evap is used on a flooded system because the need to superheat the vapour was not nessesary.
My understanding of a wet evap was that liquid entered it and saturated liquid left it. Thats why I thought they were more effiecent.

Oh well back to school

taz.

[US Iceman]

I was of the understanding that flooded evaporators were used because the heat transfer was greater than saturated liquid vapour evaps (DX).

The reason behind this was the whole evap is used on a flooded system because the need to superheat the vapour was not nessesary.

My understanding of a wet evap was that liquid entered it and saturated liquid left it. Thats why I thought they were more effiecent.

All true...

The original question related to the type of liquid delivery method to the evaporators.

A mechanically pumped liquid overfeed system simply uses a pump to transfer saturated liquid (under pressure) to the evaporators.

A gas-powered system uses high-pressure gas to displace liquid from a tank. Since this liquid is now at a higher pressure, it flows to the evaporator.

There was another thread in the last few days that discussed the coil capacity issue, but I can't remember where it is right now.

[taz24]

The original question related to the type of liquid delivery method to the evaporators.

Should have read the question then shouldn't I .

Back to School twice for me .

Cheers taz.

[Sinke]

...The simply things is the best things...
...just buy mechanical hermetic pump and keep on...

[cha09]

I have 300 kW ammonia refrigeration system, I can give you all the spec of equipments if you want it

[Sinke]

I have 300 kW ammonia refrigeration system, I can give you all the spec of equipments if you want it

...I work on system over 1000kw....

[John Hunter]

I For what its worth
Being an ammonia re-circulation Plant your liquid flow to your evaporators will be either via a Flow Control valve or fixed orifice . Assuming the valve is set to handle the total load a fairly constant differential pressure across the valve or orifice is needed. A pump system can deliver a fairly constant delivery head to your distribution system. Gas systems require constant gas pressure to do the same , but to achieve this there will times you would be running compressors to maintain head. The economics doesn't justify this. So use pumps , there are some very reliable Semi- hermetic canned pumps on the market , but I still like my Cornells.

[US Iceman]

...but I still like my Cornells.

Me too! Just started up a new set last week.

[GrantD]

Hi Guys,

Been reading about pumps vs. "pumperdrum" systems. I too would rather go for pumps. Unfortunately for me one of our clients has a "pumperdrum" system and thus have to work on it. A high maintenance animal to say the least!! The system uses H.P liquid subcooled through a closed flash tank to pumpwith majority of the time and when the level in the surge drum rises either when a room goes on defrost or so on pumps on hot gas once the float switch closes.

A few questions for the guys in the know:
Im sure they have a place somewhere.
1.) What are the energy implications using them on low stage compared to high stage systems? How would or what would the formulas be to equate these differences?
2.) What is the general NRV design for inlet/outlet on the pumping vessel?
3.) How safe are they really? they are constantly ranging between 0 bar - 3 bar every few minutes. Surely the vessel undergo stress.
4.) How would you design these vessel sizes on the required flow rate for the required load with a bit extra for future extensions? And.... Where would you fit your float switch on the drum for this?
5.) Horizontal or vertical and difference other than space requirements?

Your opinions would be greatly appreciated.
Out of intrest. Recently we found a problem with one drum not pumping efficiently. The problem being a bypassing vent valve on that drum. This caused a bigger problem for the compressor at the end of the line.. A bit of liquid carry over!!!!!
On further inspection we found the common vent line from the pumping vessels tied into a nozzle on the surge drum literally 200 - 300mm from the dry suction. Thus with enough velocity the subcooled pumping liquid vent through the bypassing vent valve into the surge drum and carried through to the dry suction line...
Luckily we had recently installed a "knockout pot" at the end of the line wich took most of the liquid...

We must always remember to keep our eyes open on site!!!!

[US Iceman]

We must always remember to keep our eyes open on site!!!!

Sounds like a lot of fun!

I absolutely detest these systems. Knocking, banging, jumping pipes.... something for everyone!

You ask a lot of questions....

Let's try one first.

What are the energy implications using them on low stage compared to high stage systems?

Less impact at high suction pressures, more at lower suction pressures. It depends on where the pressurizing gas comes from and where the gas is vented to. Some changes minimize the impact, however pumps do not have these negative side-effects.

[HallsEngineer]

The pumper drum systems are the rarity yes? There are many more overfeed systems outhere so somebody has worked this out before. Pumped systems rule!!!!

[US Iceman]

Pumped systems rule!!!!

I agree...

[Frank Day]

Hi Guys

For a moderate capacity NH3 recirculation plant, say about 300 kW cooling, what is the best arrangement for feeding evaporator coils?, liquid pump based system or gas/compressor driven system …, could anyone guys reason things out for me

lraja

Rather stay with mechanically pumped liquid recirculation than gas powered. Pumper drum systems can be very problematic and frustrating to work on, ask GrantD. Mechanically pumped systems are more cost effective in the long run providing all components are sized correctly.

[LoveNH3]

Liquid Recirculation

Liquid overfeed systems use a pumping mechanism to force more liquid refrigerant than can be vaporized through one or more evaporators. By pumping much more refrigerant through the system than is evaporated, liquid recirculation or overfeed systems effectively increase the wetted area of the evaporator coil and in some applications can increase the cooling rate over that of flooded evaporators and always that of DX evaporators with comparable surface area and internal volumes.

The operational difference between flooded evaporators and liquid recirculation systems is that an electric motor driven or gas operated pump system delivers overfeed liquid to the evaporators. Liquid feed pump pressure and flow control devices at the evaporator inlet(such as a hand expansion valve) control the flow rate through an evaporator of a liquid recirculation system. Multiple evaporators can use the same low pressure receiver and the flow rate through each evaporator is not directly dependent on the evaporation rate within that evaporator.

Recirculation Ratio

In the liquid recirculation system, a prime consideration is to overfeed the evaporator or coil circuit so that regardless of load variation, all internal surfaces of the evaporator are fully wetted and effective in heat absorption. Once the evaporator or coil circuit is fully wetted, little is gained by increasing the amount of liquid refrigerant. In the liquid overfeed system the recirculation ratio is defined as the relationship (by weight) of the amount of liquid pumped - to the amount of vapor formed in the evaporator or coil circuits in the heat transfer process. A 4:1 recirculation ratio means that 4 pounds of refrigerant is fed to the coil and 1 pound is vaporized.The return line would contain 1 part gas and 3 parts liquid by weight Pumps provide the pressure and flow for the required recirculation ratio.

[Frank Day]

GrantD

3.) How safe are they really? they are constantly ranging between 0 bar - 3 bar every few minutes. Surely the vessel undergo stress.

In answer to your third question.

Yes the pumping vessels are constantly under stress and
are considered pressure vessels.They are therefore manufactured to strict engineering codes of practice such as ASME 1X , BS 5500 or what ever code is Government approved in the respective country of manufacture. Essentially specific calculations have to be made relative to the working conditions of the vessel under manufacture. Pressure, temperature, size, refrigerant and safety all form part of the equation. The out come is that the vessel materials and thickness together with manufacturing procedures are established. Through out the manufacturing process Government Approved Inspectors are used to carry out inspections at mandatory hold points, such as material identification, vessel fit ups, welder qualifications, welding procedures, are all checked and documented. On completion the vessel is hydrostaticaly pressure tested, stress relieved and re - pressure tested before final approval and certification. Only then can the vessel be put into service. Every pressure vessel is issued with a data book and carries a name plate with manufactures details and design data.

Because the pumping force used on these vessels is by way of either regulated high pressure liquid or high pressure vapour the design pressure is therefore higher than the refrigeration systems normal head pressure. In other words it is possible for the vessel to be subjected to the systems head pressure therefore, that sets your design criteria.

These pumping vessels should be protected by safety relief valves set to lift at maximum allowable working vessel pressure and not design pressure which is higher.

The bottom line is that providing the vessels have been manufactured to the correct standards they are safe. I trust this answers your question.

[GrantD]

Hi Frank,

Thanks for the reply on that. We have spoken privately, but for the sake of the forum.. It definately makes me feel safer that the guys on the ground are in good hands.

On the same site the pumping drums make funny or should I say scary sounds. If you arent expecting them they can be startling! What Ive pinned them down to is this:

First of all the sound starts off with a low hum and gradually gets louder towards the end of the cycle. Therefore as the liquid level in the drum drops it changes the sound the drum makes. As the drum empties it gets hollow. It also tends to only happen when pumping on hot gas. Bear in mind that whether pumping on liquid or hot gas the pressures are the same. Does the velocity play a role here?
Having opened up the inlet NRV to inspect. I found it to be a 10mm plate with a pipe end through a flange as a seat. A very crude seal. If the valve doesnt seal 100% , the liquid would rather want to bypass the NRV but it battles as the NRV is being held down by the pressure hence it vibrates like a humming bird on its seat.
Solution possibly thicker plates maybe 25mm, spring close assisted?, wider surface seat surface area. Current seat surface area is shed 40 pipe wall thickness.
Going back to the "making a noise on hot gas pump statement". The system pumps on hot gas when the level in the drum rises and float closes. If you watch the pumped liquid pressure gauge towards the end of the line.. It pulsates less when pumping on hot gas and Ive noticed the drum cycle pump time is longer on hot gas. Have we pushed the recirc. rate for this breif period of time( 5- 10 minutes till the level in drum settles) so much that we cannot fill the wet return anymore?

Dont know if my thinking is right? opinions guys would be aprreciated This thread on recirc. rate, pump pressure could be intresting to keep going.

GrantD

[Magoo]

Back to original post, I definitly prefer the pump fed /overfead system.
magoo

[jklydr]

if the gas powered pump system is optimised, it can be also very effective and efficient like a mechenical pump system. did any one have the experience for the design of such systems?

[Sridhar1312]

NH3 liquid pumping is most energy efficient .We had designed similar system way back in 1980 for Minus 25 deep freezer for marine products, where we used low pressure reciever as additional in addition to the high pressure reciever. From Low pressure reciever the Liquid ammonia pump, pumps liquid ammonia to various evaporators and the return line is back to low pressure receiver with baffles for draining the liquid to bottom.
Suction to the compressor is taken to the compressor.We had imported Vilter ammonia pump for this purpose.

[Mark II]

iraja,

You got a great respsponse for such a vague question..

The key benifit to liquid recirc is capacacity.. If there is such a thing as 300kw systems with liquid pumps, I've never seen one..

How you guys going with my auto Vi issue ??

[chilldis]

liquid pumps by all means, not only are pumper drum systems high maintenance with multiple check valves and three-way valves I have seen a higher rate of external vessel and pipe corrosion due to constant frost/defrosting during the life of the system. this will happen no matter how well the insulation contractor does their job. and as always power consumption is always a factor( with pumps the need to recompress the dump tank hot gas is eliminated.)

[mbc]

In very brief .

It depends on how big is your system?

If you system is large so you must have separators vessel to avoid and protect liquid comes to compressor during working system and rooms come on and off and also during defrost time (hot gas), to control NH3 level in separator you have electrical system ( low level turn on and off solenoid valve is supplying liquid from receiver to separator) and mechanical system (float valve).
So to use liquid NH3 pump is only way you have.

But in small system .
If your system is small and you have one room with one compressor (or one cooling system)
As you know we have 3 kind of evaporation system in our evaporators.
1- Dry system 2- wet system 3 – flooded system
In dry system you can feed evaporator direct from receiver with putting solenoid valve and expansion valve and… also your defrost system should be electrical heaters or you should have Re evaporator in your system to use hot gas for defrost and it is better you have suction separator to protect your compressor from returning liquid .
In wet system and flooded system are same as dry one except your receiver and suction separator should be too large or use a hot water for defrost
Also pumping system has more efficiency