Temperature problem with liquid NH3

[energeia]

Hello.
I'm a new member here. I recently got my first job as a chemichal engineering in a factory with french fries. There is a problem though with the freezer that cools the fried potatoes. The system uses liquid ammonia, and the problem description is that we can't cool the potatoes enough. I have a few questions, and i'd be more than happy if someone could enlighten me into this matters.
First of all i have a simple (i guess) question. The compressor displays the suction pressure in units of -40°C/R. Why is that? I can't figure out the math between the units of pressure that equals °C/R.

What else should i look at in order to understand the source of the problem? If ammonia leaves the freezer in order to go to the compressor at -40 degrees, it seems that there ins't a problem with the flow of ammonia or the comporessor system. In the freezer there is a big fan, circulating the air, and there is also the coil that if filled (?) with ammonia.
How can i find out if the coil is fully filled with ammonia? If the temperature of the suction in the compressor is -40, i should have temperature in the coil at least -40 degrees celcius. Am i right? Does anyone have any guidelines, or ideas about what things to check?

[josef]

Hi energeia, the time french fries in the freezer?

[Oregon Jim]

It sounds like your -40 temperature should be just fine, but having a good coil temperature is just a small part of freezing product. You must have good air circulation through your coils and through your product. Anything that reduces this air flow reduces your temperature transfer from the product to the coils. You might also be having a problem with NH3 feeding into your coil caused by a poorly adjusted expansion valve, bad solenoid valve, or even a pump problem.

Problems with air flow can be caused by fan problems, ice or snow build-up on your coils, fries piled too deep to allow air flow through them, or many other things. How long will the fries be in your freezer? Are there belt speeds you can adjust? What is your product temperature entering the freezer? Could you be having a problem with pre-cooling?

More information would be helpful. Tell us about your freezer.

[energeia]

First of all, thanks for the attention and the replies.
The fries stay in the tunnel for about 20 minutes. The belts that tranfer them, can't go slower. I think that there isn't a problem with ice building on the coil, because there isn't much ice, even after 6-8 hours. The fans are adjustable, and we have the same result if we reduce the speed of the fans to half. so i suppose it's not a "wind" problem either.
Jim, i don't think there is an expansion valve in the system. Are you talking about the kind of valve that exists in systems that use *****, like home refrigerators work?
About the pumps, there are two pumps available in the system. Each pump produces the same result. We tried running both pumps together, in case the one wasn't enough, but the results were the same.
One day i did a test, running the system, without the fans and one door open. I was watching the coil and i noticed that the 1/3 upper part of the coils didn't have ice on the pipes. I touched the pipes and i could leave my hand on it. It didn't feel cold, like the lower pipes of the coil. I measured the temperature of the cold coil with a simple thermometer, and it was around -32 degrees. But the upper part had definately higher temperature. How do these systems get filled with ammonia? I'm thinking that, perhaps it hasn't been filled properly?Is it possible to have air in the system?

[josef]

Hi, energeia, surely you do not have air in the freezer, I am convinced that no problem in the system you do not, the problem is somewhere else, your refrigerator is a small, 20 minutes can not be frozen french fries to a more-5až-7 ° C, provided that the frying shells into the freezer fries max +45 ° C. For me it is a factory for chips with the same problem. You slow-starting conveyor larger meeting that same nezmrzne. The only help you produce a small amount of fries, do not like to be the owner. The merchant who offers many mrazničku suggests, but the reality is different. On your question why the pipe above warmer?-Air temperature in the freezer and ammonia is equal. Excuse my ignorance of English, many of my thoughts turn when translate. Many of time I spent with the same problem as you, will be better if the email. Greet Josef

[US Iceman]

... because there isn't much ice, even after 6-8 hours.

...

I was watching the coil and i noticed that the 1/3 upper part of the coils didn't have ice on the pipes. I touched the pipes and i could leave my hand on it. It didn't feel cold, like the lower pipes of the coil.

If the pumps did not have enough liquid sitting in the vessel above them, they would be cavitating.

It sounds like you have a problem with the liquid feed to the evaporators. Something is either causing the liquid to be used before it gets to these evaporators, or...

the hand expansion valves are not adjusted properly
or
a strainer in the liquid line has a lot of debris in it, blocking off the major supply of liquid to the coils.

If the coils have sufficient liquid feeding into them and the suction pressure is constant the coil temperature should be constant from top to bottom.

[VGV]

Sounds like you may have a couple of problems:

A tunnel freezer coil should have realtively even temperatures top to bottom. These coils are usually designed for liquid overfeed..meaning that two to six times the liquid being evaporated is being fed to the coil. Leaving coil temperature should be close to entering. A difference means lack of flow.

1. Liquid flow.
The reason for two pumps on a liquid recirculation system is to provide redundancy. One pump should be sufficient for operation if it performing up to design capacity. Lack of flow could be as US Iceman said....cavitiation, plugged strainers, hand expansion valve setting.....OR maybe there is a flow regulating valve in liquid supply line to coil to prevent pump from deadheading if coil flow control solenoid valve shuts off... If so, flow regulator could be jacked open, or otherwise not working properly.

2. Ammonia quality.
A supply liquid temperature of -32C at a compressor suction of -40C says there is 8C superheat....thats a LOT!

Does the vessel from which pumps get suction have an oil pot?? Is it frosted?? Maybe your ammonia is contaminated with oil?? Or maybe water?? In the US, ammonia suppliers will test for water ...no charge.

Hope this helps.

VGV

[Plank!]

energeia, a little more info would go a long way here. Several times you mention -40c are any of these values measured temeratures or are they all related to the suction pressure?
If your evaporator is starved of ammonia, the suction pressure would be low ie -40c but with a high dry suction temp.
Is the compressor loading up or limiting duty on suction pressure?

edit - I just noticed you mentioned the bottom of the coil is -32c, thats 8K from suction pressure, on a flooded system this should be much closer to -40c. You have a blockage or flow restriction.

What is the temp of the wet vapour leaving the evaporator? The actual vapour temp entering the compressor? Even the liquid temp leaving the surge drum/pump would be a good indicator. Should be in the region of -36c
As for air in the system, its always possible in a system running in vacuum to draw in a small ammount of air, however it would be found in the condensor not the evaporator - look at the condensing pressure and compare it with the liquid temp leaving the condensor. The pressure here should be almost the same as the liquid temp.

It certainly sounds like the system is under feeding the coil. Check everything mentioned by US Iceman.

What kind of defrost system do you have?

[Oregon Jim]

"The fries stay in the tunnel for about 20 minutes. The belts that tranfer them, can't go slower."

Perhaps the belts are moving too slow and your fries are too piled deep on your belts. This would dramatically reduce your air flow through the fries and across the coil, resulting in a good suction pressure and NH3 temperature, but inadequate freezing of product.


"Jim, i don't think there is an expansion valve in the system. Are you talking about the kind of valve that exists in systems that use *****, like home refrigerators work?"

Follow the NH3 line from your pump to your evaporator coil and you will find an expansion valve at or near the coil inlet. It is used to control the amount of NH3 feeding into your coil. I suspect you will find this valve is not opened far enough and your coil is starved of NH3. This would also give you a good suction pressure and NH3 temperature, but would greatly limit your ability to freeze product.

[J.F.C.]

About the degrees C question.
In the vessel your compressor is sucking from there is a mix of liquid and vapor therefore the pressur can bi reed in deg. C/R.
I know it is used in UNISAB controlers(pres "set" to change to BAR)

[chilldis]

just a short note of agreement with Oregon Jim, if the top section of your coil is not frosted the coil is starved for liquid. open your hand exspansion valve 1/2 turn and monitor the coil. keep doing this until your coil frost pattern is uniform. this approach is low-tech but it works.

[aphun]

I make french fries with an IQF tunnel, 20 minutes is a lot of time I think; we need 8.5 minutes for freeze them to -18°C, 500 kls/hr.
My system dont have recirculation pump for NH3, is by gravity from LPS to coil, and when I had a similar problem was: 1. the feed of liquid to the coil is small, we turn the expansion valve 1/2 - 1 turn more; 2. Need to defrost the coil have a lot of ice; 3. Have oil in the bottom of the coil, we have to put a purge valve in the pipe before the liquid enter to the coil in order to remove the oil of the bottom; 4. The layer of the product is to deep and the fries leave out blocked, in this case the suction pressure rise in the compressors; 5. the second belt have a lot of ice and the wind cant pass through the product; 6. The rotation of the fans was inverted.
I hope this could help you to try to find the source of the problem.

aphun

[energeia]

If the pumps did not have enough liquid sitting in the vessel above them, they would be cavitating.

There is enough liquid in the vessel above the pump. There is an electrode (i think it is called like that) that monitors the lowest level of the tank.

1. Liquid flow.
The reason for two pumps on a liquid recirculation system is to provide redundancy. One pump should be sufficient for operation if it performing up to design capacity. Lack of flow could be as US Iceman said....cavitiation, plugged strainers, hand expansion valve setting.....OR maybe there is a flow regulating valve in liquid supply line to coil to prevent pump from deadheading if coil flow control solenoid valve shuts off... If so, flow regulator could be jacked open, or otherwise not working properly.

2. Ammonia quality.
A supply liquid temperature of -32C at a compressor suction of -40C says there is 8C superheat....thats a LOT!

Yes. The reason for having two pumps is only for redundancy. We use one of them each day. We have checked all valves, and they are fully open. I don't think there is captivation in the pumps. The temperature of -32C was taken with a low quality thermometer and at the outside part of a pipe of the coil. So i'm not sure it was just -32. I guess it was lower than that. The -40C is the temperature that the compressor measures while sucking ammonia. The vessel doesn't have an oil pot.

Is the compressor loading up or limiting duty on suction pressure?

edit - I just noticed you mentioned the bottom of the coil is -32c, thats 8K from suction pressure, on a flooded system this should be much closer to -40c. You have a blockage or flow restriction.

What is the temp of the wet vapour leaving the evaporator? The actual vapour temp entering the compressor? Even the liquid temp leaving the surge drum/pump would be a good indicator. Should be in the region of -36c
As for air in the system, its always possible in a system running in vacuum to draw in a small ammount of air, however it would be found in the condensor not the evaporator - look at the condensing pressure and compare it with the liquid temp leaving the condensor. The pressure here should be almost the same as the liquid temp.

It certainly sounds like the system is under feeding the coil. Check everything mentioned by US Iceman.

What kind of defrost system do you have?

The compressor at the beggining it is working on 100% load, as expected. After a few hours, it drops it's load to 60%-70% i think. That means that we aren't cooling anything, so the demand on liquid ammonia is getting lower. I haven't measured the temp of the wet vapour leaving the evaporator. The actual vapour temp entering the compressor is the one that the display of the compressor states (-40C)

Perhaps the belts are moving too slow and your fries are too piled deep on your belts. This would dramatically reduce your air flow through the fries and across the coil, resulting in a good suction pressure and NH3 temperature, but inadequate freezing of product.


Follow the NH3 line from your pump to your evaporator coil and you will find an expansion valve at or near the coil inlet. It is used to control the amount of NH3 feeding into your coil. I suspect you will find this valve is not opened far enough and your coil is starved of NH3. This would also give you a good suction pressure and NH3 temperature, but would greatly limit your ability to freeze product.

We've put the belts so slow, because we wanted to increase the cooling of the french fries. There is also a mechanism that moves french fries around the belt, so that they don't get stuck to each other and they get equal cooling. I've seen the valves that you mention, but they are fully opened. One techkical guy suggested that we close them a little, in order to have a better flow. We tried that as well, but with no difference.

My system dont have recirculation pump for NH3, is by gravity from LPS to coil, and when I had a similar problem was:
1. the feed of liquid to the coil is small, we turn the expansion valve 1/2 - 1 turn more;
2. Need to defrost the coil have a lot of ice;
3. Have oil in the bottom of the coil, we have to put a purge valve in the pipe before the liquid enter to the coil in order to remove the oil of the bottom;
4. The layer of the product is to deep and the fries leave out blocked, in this case the suction pressure rise in the compressors;
5. the second belt have a lot of ice and the wind cant pass through the product;
6. The rotation of the fans was inverted.

We have checked case 1, 2, 3, 4, 5 and nothing like that is hapenning. I'll monitor the rotation of the fans but i think it's right.



I was thinking that, when all things look right and though the result isn't the expected one, something big and simple must be happening. So i tried to look for other things, like having heat added to the freezer, leaks on the walls of the freezer etc. Nothing seems wrong though, except for one thing: The liquid ammonia enters the coil on the upper part of the pipes, and the outfeed of the coil comes from the lowest pipes. In my opinion that seems wrong. But then i asked the guy (engineer) from the company that installed the freezer and he said that this is normal, and the right way. What do you have to say about it? I think it would be right if we wanted to heat up the circulating air with some other liquid (e.g hot water) and not trying to cool it down.

[VGV]

I worked with Lewis when they manufactured IQF freezers.. Almost all coils were top feed so they wouldn't pocket oil. As aphun mentioned, he had to install an oil purge valve...but you say this has been checked.

However, you also state no oil pot on the pump vessel... Is the pump vessel free of oil??

If suction line is short (minimum temp. gain), suction lines not blocked (valves open and no strainer blockage) pumped liquid temperature should be very close to compressor suction temperature....If not, oil or water contamination are suspect..

[Plank!]

The temperature of -32C was taken with a low quality thermometer and at the outside part of a pipe of the coil. So i'm not sure it was just -32. I guess it was lower than that. The -40C is the temperature that the compressor measures while sucking ammonia. The vessel doesn't have an oil pot.

Ok, this differs from your original post - there you said the suction pressure was -40°c.
I think you may be misinterpereting the -40°c shown on the compressor.
In your first post you refered to the suction pressure as -40°c, this is the kind of value i am familiar with, we use °c as pressure equivelant as its a more logical way to work.
For example suction Pressure -40°c with dry suction Temp of -36°c gives us 4°K superheat. Much easier than comparing Kpa/Bar/PSI etc

Suction pressure and temperature cannot be the same, unless you have a machine capable of compressing liquid NH3

The compressor at the beggining it is working on 100% load, as expected. After a few hours, it drops it's load to 60%-70% i think. That means that we aren't cooling anything, so the demand on liquid ammonia is getting lower. I haven't measured the temp of the wet vapour leaving the evaporator. The actual vapour temp entering the compressor is the one that the display of the compressor states (-40C)

what does the suction pressure do when the compressor backs off?
It sounds like suction limiting. ie Compressor off loading due to low suction pressure.

What is your suction control setpoint?
And the suction limit setpoint?

[energeia]

Ok, this differs from your original post - there you said the suction pressure was -40°c.
I think you may be misinterpereting the -40°c shown on the compressor.
In your first post you refered to the suction pressure as -40°c, this is the kind of value i am familiar with, we use °c as pressure equivelant as its a more logical way to work.
For example suction Pressure -40°c with dry suction Temp of -36°c gives us 4°K superheat. Much easier than comparing Kpa/Bar/PSI etc

Suction pressure and temperature cannot be the same, unless you have a machine capable of compressing liquid NH3





what does the suction pressure do when the compressor backs off?
It sounds like suction limiting. ie Compressor off loading due to low suction pressure.

What is your suction control setpoint?
And the suction limit setpoint?

The suction pressure is always -39.5°c or -40°c independent of the compressor load. There is a low pressure vessel of ammonia from which the compressor sucks vapor ammonia in order to compress it. From the same vessel the pump, pumps liquid ammonia to the evaporator, and finally the vapour ammonia that returns from the evaporator goes into the same vessel.
So the temperature that the compressor displays as pressure (°c/R) is the temperature of the vapour of ammonia coming from the evaporator?

I've read some information on the web also, and i'm thinking maybe we have a subcooling problem. If the ammonia in the vessel i described above is in saturated state, when the pump increases the pressure in order to move ammonia around, the state becomes subcooled. Am i right? After the pump there is a pipe that goes to the freezer, and just before entering the evaporator coil, there are two manual valves on/off (sva type). If i'm having subcooling effect, will i be able to regulate the pressure with the sva valves?


I'm sorry if i mix up things, but i'm trying to understand better the whole thing.

[VGV]

Energeia,

Yes, you have the correct picture.

At compressor suction temperature of minus 40°F, what is the suction PRESSURE, and what is the pumped liquid TEMPERATURE leaving the pump??

VGV

[NH3LVR]

It certainly sounds to me like you have a problem with lack of NH3 being fed to the coil.
But just a thought. It may not apply in your case. I worked in a plant on a seasonal basis a few years ago. One winter they moved the Spiral Freezer to a different location. When they restarted it built little ice on the coil. With the light loads they were running it worked well. It did not work as well when we put it into heavier service.
What had happened is that the fans were running in reverse. There was lots of ice on the coil, but where we could not see it.
Again I do not think this is your problem, but something I always check.
Did the Freezer work alright at one time?
Are you using Cornell Pumps?

[Poodzy]

Are you running butterfly valves on the suction header from the freezer? It could be stuck partially closed.. Just another suggestion. What kind of freezer is this anways? Spiral, IQF, ? multi evap set up? Could be a problem with just one evaporator if is multiple. I would also check the suction valves we had a problem with some of the ones we have here sticking and not fully opening. But like Oregon Jim and some others if they are trying to produce more then you can freeze it is out of your hands.

[Plank!]

Energeia,

Yes, you have the correct picture.

At compressor suction temperature of minus 40°F, what is the suction PRESSURE, and what is the pumped liquid TEMPERATURE leaving the pump??

VGV

Not so.
The -40°c displayed IS the pressure.

The suction pressure is always -39.5°c or -40°c independent of the compressor load.

I'm concerned that the pressure is only changing by 0.5°C. Every system I've seen changes by more than that in normal use. If the system is starting to freeze then warming up I'd expect the suction to change much more than that. Has the pressure transducer been calibration been checked? Vent the transcucer to atmosphere, it should then read -33.6°c

As others have suggested measure the suction temperature close to the surge drum, and the liquid temperature after the pumps.
Suction temp should be 2K or more above suction pressure (in °c) Pumped liquid temp should be the same as suction pressure.

[VGV]

°C = TEMPERATURE
BARS, PSIG, IN MERCURY = PRESSURE

Answers to 3 questions will solve this.
1. Suction PRESSURE = ??
2. Suction TEMPERATURE = ??
3. Pumped Liquid Temperature +??

What is suction

[Magoo]

If there is a temp difference form bottom to top of coils, equals low liquid overfeed rate. [ should be only a few degrees C']
If low airflow, coil temp same out as infeed temps. What is pump differential pressure ?, is there sufficient liquid in surge vessel?, is the pump cavitating [ making rattly noises ]?. Have you checked settings of hand expansion valves, feeding coils ?. Has system operated since installed and now not working as well.
Has anyone drained oil off separator vessel when shut down, check oil logs versus whats been drained out.
Are the fries fluidizing on belt.
Many possibilities.
The suction temp at compressor seems correct, but is compressor/s fully loaded, has client changed something prior to freezer like by-passed the pre-cooler, ramped up hourly tonnage rates.

[gwynn jones]

Hello.SNIP..First of all i have a simple (i guess) question. The compressor displays the suction pressure in units of -40°C/R. Why is that? I can't figure out the math between the units of pressure that equals °C/R.
?

I don't know what sort of coldroom/spiral freezer you are running here.
-40c equals about -0.3 Bar pressure [NH3] at the compressor, which sounds normal for a spiral freezer but you could be short of gas, have a choked strainer or poor air flow over the evaporator giving you the low suction at the compressor.
And of course, as someone has already said, how long is your product in there and at what temperature is it going in at?

[AmmoniaMike]

From what I've been reading it sounds like the problem is brining. The fact that the coil is top fed and no frost at the top indicates that the NH3 is subcooled and it has to warm up to saturation temp before it can do much work. Back off a lot on the hand expansion valve (try starting at 1/2 turn). The plant I'm operating is all liquid recirculation aka flooded and our hand expansion valves are all between 1/8 and 1 1/2 turns open. The biggest problem I find in flooded systems (pump fed) is overfeeding ammonia liquid.

[Segei]

From what I've been reading it sounds like the problem is brining. The fact that the coil is top fed and no frost at the top indicates that the NH3 is subcooled and it has to warm up to saturation temp before it can do much work. Back off a lot on the hand expansion valve (try starting at 1/2 turn). The plant I'm operating is all liquid recirculation aka flooded and our hand expansion valves are all between 1/8 and 1 1/2 turns open. The biggest problem I find in flooded systems (pump fed) is overfeeding ammonia liquid.

Typically, liquid recirculation plants have overfed evaporators not flooded. Setting of hand expansion valve is important for overfed coil and less important for flooded coil. What do you have flooded or overfed coils?

[AmmoniaMike]

If I read energeia's problem correctly he can't seem to get the cooling he needs from his pumpe supplied top fed coil in his spiral freezer. You are correct in that what he is describing is a recirculating system and not a 'flooded' coil but people keep calling them flooded. If he was starving his coil for ammonia you would expect frost at the top of the coil and superheated ammonia at the bottom but he has warm at the top and cold at the bottom. This leads me to conclude that he is overfeeding subcooled liquid into the coil. He should not have the hand expansion valve wide open. I would start at 1/2 a turn open and go up from there if needed to get the full capacity of that coil. A great paper on this can be found at NHTres.com in the download section. Sorry I can't put the link directly but I don't have 15 posts in here so I can't post it

[Segei]

If I read energeia's problem correctly he can't seem to get the cooling he needs from his pumpe supplied top fed coil in his spiral freezer. You are correct in that what he is describing is a recirculating system and not a 'flooded' coil but people keep calling them flooded. If he was starving his coil for ammonia you would expect frost at the top of the coil and superheated ammonia at the bottom but he has warm at the top and cold at the bottom. This leads me to conclude that he is overfeeding subcooled liquid into the coil. He should not have the hand expansion valve wide open. I would start at 1/2 a turn open and go up from there if needed to get the full capacity of that coil. A great paper on this can be found at NHTres.com in the download section. Sorry I can't put the link directly but I don't have 15 posts in here so I can't post it

There is significant difference between flooded and recirculation system. In flooded coil, liquid delivered to the coil and inside the coil circulation goes by gravity. For recirculation system, liquid pushed trough the coil by liquid pump. I believe that we discussed on this forum NHTres.com paper. I don't agree with all conclusions in this paper. Majority of freezer coils are bottom feed overfed coils. Energia's coils probably the same. Most likely these coils underfed. It is difficult to help people if they aren't refrigeration guys.

[AmmoniaMike]

I also don't agree with all the conclusions in that paper however his section on what happens in a top fed coil in an overfed system is correct. No frost at the top of a top fed coil in an overfeed system would indicate to me that the liquid is subcooled so the top third of the coil is only gaining sensible heat and the real work is only starting in the last 2/3 of the coil where ammonia is starting to boil. I believe that this coil should be looked at very closely and probably should be switched to bottom feed. This would take advantage of all the surface area of the coil since the coil would then be flooded and reduce the subcooling problem.

[Segei]

I also don't agree with all the conclusions in that paper however his section on what happens in a top fed coil in an overfed system is correct. No frost at the top of a top fed coil in an overfeed system would indicate to me that the liquid is subcooled so the top third of the coil is only gaining sensible heat and the real work is only starting in the last 2/3 of the coil where ammonia is starting to boil. I believe that this coil should be looked at very closely and probably should be switched to bottom feed. This would take advantage of all the surface area of the coil since the coil would then be flooded and reduce the subcooling problem.

Why do think that this coil is top feed?
Typically, top feed coil is designed for high temp. coolers which don't require hot gas defrosting. These coils have air defrost. During defrosting, liquid solenoid will close and liquid ammonia will drain from the coil and coil will be defrosted by air(fans on).
About subcooling and sensible heat. Bottom feed coils sometimes have more frost on the some front tubes than on other tubes. This is liquid oversupply. It means that cold liquid enter the coil, warm up a bit and start evaporate. So it doesn't mean that this initial part of the circuit will be frost free. In opposite, it will have more frost because entering liquid has lower temperature.

[AmmoniaMike]

energeia says it is top feed from his post on 03-02-2009.

[Segei]

energeia says it is top feed from his post on 03-02-2009.

I'm not sure that this is correct information. It is doesn't matter top feed or bottom feed. The only way that coil won't have a frost if superheated vapor will be inside the coil. You can test it on bottom feed coil. Close hand expansion valve. Defrost coil and turn it on. Slightly open expansion valve. Bottom part of coil will have frost. At top will be superheated vapor and no frost.
Subcooled liquid is cold and more frost will be formed on the part of the coil where this liquid presented.

[US Iceman]

If you don't have liquid, you will not have frost. It's that simple.

Coil circuiting (top fed or bottom fed) is more concerned with how someone thinks coils operate. If you get liquid into them and the liquid is cold, the coil will frost.

Top fed coils (circuited properly) work better than what the paper in question indicates. That paper was written shortly after the author listened to a revised lecture I did.

[subb-zero]

This system if flooded might not have an expansion valve the pump would be rated for the amount of liquid/vapor evaporated, if it has an expansion valve it should not be fully open, the valve (expansion) valve should be 1.5 to 2 maybe 2.5 turns open, reason being to have a constant liquid supply to the evaporators, if not and the valve is all the way open then the liquid would just be dumping?

[subb-zero]

Us ice is right, depending on application, top fed works better ????? think.

[RefrigNoob]

Never worked on a NH3 system, but I havn't seen Energeia report back that they have checked all the strainers in the system for blockage/debris...