Liquid Overfeed Systems [Archive] - Refrigeration-Engineer.com forums

ecclesk

24-03-2006, 02:10 PM

I look after some systems in which the pumps are installed in this way, one vessel with two pumps, common discharge,

i only usually run one pump and keep one as stand by, i do have cavitation problems on occasion with one system and have tried everything we could think of but i think it's a design issue, also found that one of the two pumps on this plant in question seems to deliver a more stable pressure and doesnt cavitate as much as the other, even after changing the pump & motor for a spare one we still have the same issues?

i believe it may be to do with the way the vessel is designed/piped. the suction lines of the pumps are at different ends of the vessel as to are the wet return and the compressor suction, the pump situated under the wet return end of the vessel is the one we have problems with.

I'm also aware of another site within our company that has the same set up but with three pumps, it was designed to have two pumps running and one stand by but they have never been able to run two pumps due to cavitation problems.

Tycho

24-03-2006, 02:19 PM

with non returnvalves on the discharge it works like a charm.

Also, if three pipes are connected to a circular vessel, you need a vortex stopper inside the vessel so you dont get cavitation.

What kinda pump is it?

on our systems you have to manualy start and stop the pumps. On systems with two or more pumps we have diff.pressostats to start the pumps in sequence.

but on the majority of the systems we use our KS-E5 ( kvaerner pump), and usually one is enough to supply 8-12 vertical platefrezers with a stable pressure of 2 bar.

Renato RR

24-03-2006, 02:55 PM

Try to contact pump suplier to find out what he recomends and how he advice to install the pumps.

R:rolleyes: enato

SteveDixey

24-03-2006, 06:07 PM

I recently investigated a large liquid overfeed system (ammonia). The system was using multiple refrigerant pumps piped in parallel.

Each refrigerant pump was installed with a separate suction line and suction service valve dropping down from the vessel. The pump discharge pipes were installed into a common main pipe for the liquid supply to the evaporators.

What has your experience been with systems like this?

Key thing is getting the suction piping right and keeping the wet suction return away from the section of the vessel feeding the pumps.

Most now recognise pumps need 3" or even 4" suction lines and take as direct a route as possible. If the evaps have no load, and so reduced liquid feeding in, pumps can start to cavitate in low pressure \ off-load conditions. If the design is wrong, you can suffer no end of problems and it is costly to fix.

Quite common on large industrial plants, and have come across up to 3 pumps feeding in, but most use 2 with one as stand-by.

Steve

Steve

US Iceman

24-03-2006, 06:18 PM

Hi guys,

I wanted to see what sort of problems the rest of you have seen or had.

i only usually run one pump and keep one as stand by

This is what I would recommend. One pump for 100% of the requirements with one standby pump for redundancy. Either of these pumps are sized for the design flow rate and head.

...the pump situated under the wet return end of the vessel is the one we have problems with

As you probably suspect the splashing of the liquid into the vessel is creating problems for that specific pump. When the entering liquid disturbs the liquid around the pump suction connection some gas bubbles are probably being entrained in the liquid.

If liquid has bubbles in it the density of the liquid is decreased slightly. This reduces the static head on the pump suction. Any decrease in static head is a decrease in net positive suction head for the pump.

A pump curve will provide the minimum Net Positive Suction Head Required (NPSHR) that is needed at some flow condition. If the Net Positive Suction Head Available (NPSHA) is less than the requirement of the pump (NPSHR), the pump will cavitate.

One of the trouble areas is solving for the correct NPSHA and making sure the pump operating point does not change.

...it was designed to have two pumps running and one stand by but they have never been able to run two pumps due to cavitation problems

If these are centrifugal pumps they are probably fighting each other during operation. One pump is more than likely pumping more liquid than the other pump. The pump that has a higher volume flow is probably the one cavitating.

Check the motor amps to see if they are the same or different. Also compare the running amps to the motor nameplate full load amps.

US Iceman

24-03-2006, 06:27 PM

What kinda pump is it?

Hi Tycho, they are Cornell refrigerant pumps (open drive, not the hermetic ones) See the link below for additional information.

http://www.cornellpump.com/industrial/documents/HermeticHansen.pdf

...you need a vortex stopper inside the vessel so you don't get cavitation

These certainly help, but they don't prevent all of the cavitation issues. Without a vortex eliminator (stopper) you will have all sorts of problems if the operating conditions are close to the minimum pump requirements.

The vortex eliminators do provide a very much needed operational requirement.

US Iceman

24-03-2006, 06:38 PM

Most now recognise pumps need 3" or even 4" suction lines and take as direct a route as possible.

I agree. I think the suction lines should be full size (same as the pump suction connection). The pump isolation valve on the suction side should be either a full size angle valve, or a high performance butterfly valve (Much less friction loss than an angle valve).

I also believe it is important to have about 6-10 pipe diameters down stream of the angle valve to allow the liquid to straighten out before it enters the pump suction.

If the evaps have no load, and so reduced liquid feeding in, pumps can start to cavitate in low pressure \ off-load conditions. If the design is wrong, you can suffer no end of problems and it is costly to fix.

Yes sir. The minimum flow situation can be just as bad as the full load condition. Worse if you are using hermetic pumps that are not set up properly.

This system had four pumps in parallel. All pumps were running at 3550 RPM also!

Tycho

24-03-2006, 07:00 PM

Now that is the oddest looking pump I've ever seen :D

Here's how we do it
http://www.pbase.com/kimmo98/image/57683060.jpg

Also note the gas trap on the suction line.

The pump we use is the same one as in the drawing, One of the engineers that worked with us before have one running a water fountain in his garden :D

Hi Tycho, they are Cornell refrigerant pumps (open drive, not the hermetic ones) See the link below for additional information.

http://www.cornellpump.com/industrial/documents/HermeticHansen.pdf

These certainly help, but they don't prevent all of the cavitation issues. Without a vortex eliminator (stopper) you will have all sorts of problems if the operating conditions are close to the minimum pump requirements.

The vortex eliminators do provide a very much needed operational requirement.

US Iceman

24-03-2006, 09:40 PM

Beauty is in the eye of the person looking. :p

The pump link I posted shows the pumps that are similar to most of those used here in the US.

Some people build their packages very similar to what you show Tycho. Others will use two independent pump suction lines instead of the one large drop leg with the pump suctions connected to it.

Why does the piping for the gas trap in the pump suction line only project a short distance into the vessel?

Do you actually run the liquid level this low in the vessel?

Tycho

24-03-2006, 10:02 PM

I'd say that our pump looks like a Rolls Royce, next to a Yugo if it was placed side by side with the one you showed :D

It's only projecting (poking :)) a short way into the vessel, so that if the customer choose to run with a minimum liquid level it wont work as an injector (speed of liquid passing the gas trap sucking gas into the pump).

"drop leg" is that what you call the the thing sticking out from under the tank? here in Norway we call it "udder" like the ones you have on cows :D

No, we dont run with that low liquid level, it's usually a few inches higher :D
seriously, the number on the drawing is the minimum liquid head needed for the pump to maintain it's pressure.

the good thing about having the drop leg is that with this kind of pump you can maintain a liquid head with a minimum of Kg's of refrigerant instead of having to keep a ton of extra refrigerant in the seperator, you can do it with a 100kg instead.

On our Spray chillers, all of the charge is in the drop leg of the chiller, nothing up in the shell.

we used to be world-wide when we were a part of Kvaerner, but now we are somewhat limited to Norway, Denmark, Sweden, UK, Scottland, Iceland and Faroe Islands.

(Btw, if any of you UK engineers are situated close to Peterhead, and know your way around RSW, we might be interested in a service partner over there... If I get my way, you will be taken to Norway for education on our plants :))

do you mean that they use two separate suction lines on each side of the vessel, instead of the central drop leg?

we use the drop leg on all our pump systems, no exceptions.

We used to have a department in Seattle 5 years or so back... are you saying they failed in converting the U.S to our ways? :D

We now co-operate with Highland refrigeration (I seem to remember they are in Seattle also)

US Iceman

24-03-2006, 10:18 PM

I'd say that our pump looks like a Rolls Royce, next to a Yugo if it was placed side by side with the one you showed

No need to get nasty. :D

the good thing about having the drop leg is that with this kind of pump you can maintain a liquid head with a minimum of Kg's of refrigerant instead of having to keep a ton of extra refrigerant in the seperator, you can do it with a 100kg instead.

Absolutely right and no argument from me.

do you mean that the use two separate suction lines on each side of the vessel, instead of the central drop leg?

Some do, some people do not. Everyone has a personal preference. For the separate lines they connect into the bottom of the vessel. I'll try to find some pictures to show you.

are you saying they failed in converting the U.S to our ways?

You know how it is... some people are slower than others. :p

There has not been much research on these vessels since Prof. Lorentzen wrote some articles back in the 60's and 70's. I am still trying to find some copies of these.

Do you know anyone who may have some English copies of his work?

Chopper

24-03-2006, 10:22 PM

Tycho

24-03-2006, 10:55 PM

Neener neener, your pump sucks neener neener :D

personal preference? we dont go that way when it comes to the construction of the plant, Kvaerner had some pretty good engineers (the office kind, and the working kind) and they came up with some pretty good solutions back in the days, so we have stuck with them and we have some pretty sturdy plants, not saying they are the best, but they are way up there :D *floats away on a white sky* :)

Oh, we'll whip you into submission soon enough :D
*Whip crack*

(if you need a place to host the pictures, let me know)

We do struggle with (working)engineers having different minds abround the regulating, alarm and warning settings on our PLC load control system :)

I'll talk to a guy who knows a guy who is a friend with someone who has heard of a guy that has a friend who knows someone who once heard of someone who had seen some of Lorentzen's work in English somewhere :D

Kidding, I'll shoot a mail to the Lorentzen foundation and hear if they have anything, or know about any of his work in English :)

No need to get nasty. :D

Absolutely right and no argument from me.

Some do, some people do not. Everyone has a personal preference. For the separate lines they connect into the bottom of the vessel. I'll try to find some pictures to show you.

You know how it is... some people are slower than others. :p

There has not been much research on these vessels since Prof. Lorentzen wrote some articles back in the 60's and 70's. I am still trying to find some copies of these.

Do you know anyone who may have some English copies of his work?

Tycho

24-03-2006, 11:07 PM

Tycho, what is the name of the company you work for? I work for the largest and best refrigeration company in the world:D and am based in Peterhead, Scotland. I persume you are looking for someone to do service work on the pelagic vessels based across here with their RSW plants.

Kind Regards

Chopper

You mean you work for the largest company... yes? :D

I work for MMC Kulde (translates to MMC Refrigeration), we are the continuation of Kvarener Kulde, or Kvaerner Fish Process Technology (KFPT), our sister company MMC Tendos (translates to MMC Tendos :)) delivers vacuumpumps and vacuum systems for unloading fish from pelagic vessels.

we have an engineer heading over there on monday or tuesday to do some service on a vessel.

Seen this one before:
http://www.pbase.com/kimmo98/image/49607781

************Edit**************
Called him, he's going to Fraserbourgh, Fraserburoh or something like that (sorry, but the rough ending is hard on us lesser ones :))

He didnt know yet if it was monday or tuesday, but I can hook you up if you want...

Chopper

24-03-2006, 11:19 PM

Tycho

24-03-2006, 11:38 PM

I know it very well, It normally fishes out of Fraserburgh.
Did it not burst the cooler? I thought it was an optimar install.
Do you have many systems across here?
Coincidently its York i work for and we have a just recently opened a new branch in the area.:)

Regards

Chopper

First of all, the cooler did not burst.. very important :)

What happened was that the company that produce the coolers for us didnt roll (extrude? you know, when you put the pipe in the cooler and use a special tool to widen the pipe so it is "widened" to fit the hole in the endplate) all the pipes on the cooler, my guess is the cooler was made on a monday or a friday :)

They use a Loctite compound on the pipes before they roll them, but seeing as they didnt roll all the pipes, the loctite hardened and kept it tight for a while, but with the water on one side and ammonia on the other side it wore off and the cooler started leaking. Nothing wrong with the cooler apart from that.

Yes, it was an Optimar install, we used to be called "Optimar Kulde" we seem to be changing name so often, so when anyone asks anyone of us engineers out on site about the name we usually give a reply along the lines of "Yeah, we got to be known out in the market, so we decided to change names... You know, so people wouldnt recognice us" :D

We have a few over there, but as you work for York... a certain place will freeze solid before our higher up's will allow you to know anything about our systems :/' as we are competing with York in our area,

By the way, did you know that the branch in our district (Aalesund) doest work overtime? :) dig that, I clocked in 600 hours of overtime last year :) and all of their 6 engineers clocked in maybe 100 together :)

Tycho

24-03-2006, 11:48 PM

I'll just throw in one that show's our work site :)

MMC Kulde is situated in the building infront of the wheelhouse, while Optimar, the company that makes factory processing equipment is in the larger buildings to the right.

http://www.pbase.com/kimmo98/image/49607769

(you can click the link called "ships" in the top left to get to a gallery showing some of the ships we service)

Chopper

25-03-2006, 12:13 AM

For years all the vessels were kitted out with teknotherm chillers but the norm appears to either York or Optimar these days. Service on these systems is very minimal at the moment due to cut backs on fishing quotas and the chillers only clocking up 2'000 running hours per year! Alot of plant but not a good turn around o service work.
My previous employer used to be Teknotherms prefered contractor for warranty and service work on their installations. Wasnt too impressed!
If you are looking for a contractor to carry out such work across here i can give you a list of reputable contractors and there contact numbers if it would be of intrest.
Beautiful place Norway, was across working last summer up in Kristinsund on a dive support vessel. Scotland has some cracking veiws but Norways got some breath taking veiws, especially of the women. Beers too expensive though!!!:p

Tycho

25-03-2006, 01:08 AM

Teknotherm seems to be a swear word among 95% off fishingboats around here, I find it odd, as I know the people working there and they are competent people, but as soon as I enter a vessel that has had teknotherm onboard I'll be going "whoooah, it cant be that bad... NOOOooo, they didnt... really, they did that? no way... you were there yourself? and they said that?" I dont know what the deal is, but around here teknotherm is pretty much the devil...

good for us, but bad for them, it's kinda sad tho, as the people working for them are good guys and they know what they are doing.

But teknotherm, in my eyes, are stuck in the last century with their equipment... it's obsolete, end of story.

As R.J Mitchell said "It is not good enough to follow conventional methods of design. It is essential to invent and involve new methods and new ideas."

The service on our plants would be limited to the occasional solenoid calling it a day, but every now and again a relay in the controlcabinet going tits up and it would be nice to have someone over there be able to drop by and take a look at it.

If you have numbers and contact adresses for local companies that you feel are competent, it would be great if you could shoot them over, as the only ones we work with in GB is KER refrigeration in Ireland (and it seems like we only work with them, so they can get discounts on our spare parts :))

Beautiful country my arse
http://www.pbase.com/kimmo98/image/57691781.jpg
This is me driving home from work last week... I'll agree with the beer price tho... costs an arm and a leg to get drunk in Norway :) and you'll have to sign you firstborne over to the government.. and I''m not kidding :D

For years all the vessels were kitted out with teknotherm chillers but the norm appears to either York or Optimar these days. Service on these systems is very minimal at the moment due to cut backs on fishing quotas and the chillers only clocking up 2'000 running hours per year! Alot of plant but not a good turn around o service work.
My previous employer used to be Teknotherms prefered contractor for warranty and service work on their installations. Wasnt too impressed!
If you are looking for a contractor to carry out such work across here i can give you a list of reputable contractors and there contact numbers if it would be of intrest.
Beautiful place Norway, was across working last summer up in Kristinsund on a dive support vessel. Scotland has some cracking veiws but Norways got some breath taking veiws, especially of the women. Beers too expensive though!!!:p

US Iceman

25-03-2006, 02:36 AM

...not saying they are the best, but they are way up there

I do agree with you the plants should be designed and built to last for a long time without problems. This is always my goal to "do it right the first time". It is not much harder to design it right and does not cost much extra to do it right. What I like to see is a system that you can start up without much trouble and walk away from knowing it will work properly.

Oh, we'll whip you into submission soon enough

Sure you will. ;) I have been know to change my mind if presented with facts.:cool:

Control logic and safety and alarm points are always a fun discussion topic. Everyone has an idea for the best way. Myself, I like two layers of controls (at least). One layer is a fail-safe when all else goes balls up. The second layer is the upper-most layer for all of the automatic controls, PLC, computers, etc.

An example would be a mechanical float switch for a high level shutdown, while a capacitance probe can be used for normal operations and safeties. If one fails, the other is sure to catch it.

I do appreciate you inquiring about the Lorentzen papers. He wrote about some of the most interesting topics and was always based on solid evidence. Unfortunately, I suspect most of it has been forgotten by most.

Oh, by the way... The picture you took is very cloudy and out of focus.:p

And what is this talk about beautiful women??? You didn't answer that one.:D

US Iceman

25-03-2006, 02:41 AM

Andy

25-03-2006, 09:35 AM

US Iceman

25-03-2006, 04:57 PM

Thanks Andy. I reviewed the website, but it appeared they only listed a picture of a chiller without much additional information.

SteveDixey

25-03-2006, 07:26 PM

If you click on this line in the downloads section,

"installation manual GP" it is a pdf file, and look at pages 23 and 24.

Steve

aawood1

25-03-2006, 08:54 PM

Hi US Iceman, I have a 500 ton per hour surge drum that has 3 Douglas 4A pumps freeding a Freezer that needs 400 tons of Ammonia per hour in the 7 evaporators. The pump main is 6" about 230 ft. in lenght with a working pressure of .5 Bar,with a wet return of
10" the same lenght that feeds back in at each end of the surge drum. It was about 2 days untill we got it right on each coil, but now works well at -38oC on the coils and -42oC off the surge drun to the compressors.
( Got called to a Ashworth Inc. freezer today it had Frigoscandia on the door but the plate inside said Ashworth Inc. USA, Mini gryo belt freezer as the belt had stoped. found it had come over the top of the drum. Not bad it was made in 1972. The freezers we get now are the Frigoscandia Gryocompact units self-stacking belt, We had the first one in the UK 3 years of over time untill it was right and lots of nights out with the Frigo boss men.( Good old days)
all the best Arthur.

US Iceman

25-03-2006, 11:42 PM

"installation manual GP" it is a pdf file, and look at pages 23 and 24.

This is for the pump installation manual. I did look at this but i did not see anything about spray chiller requirements.

I would assume the pump suction design is similar for both liquid overfeed or spray chillers. The inside of the spray chiller is what I'm interested in.

Thanks Steve.

US Iceman

25-03-2006, 11:46 PM

...It was about 2 days until we got it right on each coil...

Was this time for the adjustment of the hand expansion valves on the coils?

It does not really surprise me that it took this long. These adjustments can really eat up time.

What type of methods did you use for the hand expansion valve adjustments?

aawood1

26-03-2006, 02:22 PM

Yes it was time taken to adjust the hand expansion valves. When the freezer was moved we had 18 probe pockets welded in to the coil infeed and return pipe work and the liquid line header and wet return header, all feeding back to one digital thermometer. Then with some good luck it all started to get a good balance over the coils. Down side was the main contractor did not work out the right Ammonia charge for the new pipe work, Had to charge 5 Tons of Ammonia into the old and new system, 4 Tons more that they worked out. I have adjusted the liquid feed pressure down on the RS A4A pressure regulators on each pump to .5 Bar from 1 Bar.

US Iceman

26-03-2006, 06:18 PM

I have adjusted the liquid feed pressure down on the RS A4A pressure regulators on each pump to .5 Bar from 1 Bar.

Where are these regulators installed? Are they at the end of liquid line and connected into the wet suction lines?

If I am following your description then these regulators are for the minimum flow pressure control?

The four extra Tons of ammonia is a fairly big mistake. Any ideas on how they were this far off of the system charge?

aawood1

26-03-2006, 07:38 PM

Hi Iceman, The RS valves are fitted next to the pumps before the right angle non-return/ stop valves where the 3 pumps connect into the liquid line. As for the Ammonia charge they said it was the fact that the main system was low to start with. But if you take 50 ft of 10" pipe that makes up the wet return header on the freezer and the coils that have about 1/4 a ton of Ammonia in each one with the 25 ft. up lift onto the roof. (They did not have to put the cost of charging into the cost of moveing the freezer it made the bill look good. The job was put out with us doing all the pump outs and venting of the pipe work and the pressure testing/ recharging of the plant.) What would say.
Arthur.

US Iceman

26-03-2006, 07:38 PM

Here are some pictures from one system.

One picture is of the low-pressure receiver and pump looking down the engine room.

The other picture shows a single pump installed before insulation was applied.

I apologize for the small pictures, but the file size was too large to post directly. They were printed into a PDF to get the file size down to the limit allowed by the RE site.

US Iceman

26-03-2006, 07:53 PM

Hi Arthur,

The RS valves are fitted next to the pumps before the right angle non-return/ stop valves where the 3 pumps connect into the liquid line.

OK, I think I see how they are installed. These are the bypass valves that maintain a constant head on the pumps. If all of the solenoid valves are closed and the pumps are still running, the pressure is bypassed back into the vessel or wet suction line. Is that right?

As for the Ammonia charge they said it was the fact that the main system was low to start with.

That is a lot of ammonia to be short of. When the system was operating it must have been right at the lowest minimum level to still run.

They did not have to put the cost of charging into the cost of moving the freezer it made the bill look good. The job was put out with us doing all the pump outs and venting of the pipe work and the pressure testing/ recharging of the plant. What would say.

Sounds like a way to make the first cost look better.

I think it is a good idea for the facility people to do the pump outs and venting since they probably know more about the system though.

aawood1

27-03-2006, 04:51 PM

Hi Iceman, Yes the pressure valves return into the surge drum at just over the liquid high level control about 1/3 up at one end. The Freezer was made by Hardy Lucas of the USA and is about 30Ft wide 50 Ft long and 30Ft high with a 6Ft 6ins wide belt.
Arthur

US Iceman

27-03-2006, 10:21 PM

Tycho

28-03-2006, 06:08 PM

Tycho,

Do have any information you can share about the spray chillers? I think these are very interesting, but this is one item I have limited experience with.

I understand the principles and the benefits but have not been able to find any technical information related to design aspects.

Thanks,
US Iceman

We dont have much information about the Chillers ourselves, as it's not our product.

The design is very simple, looks alomst like a normal chiller, tho on the inside only the lower half has tubes, in the top half there are two pipes with nozzles, side by side running the length of the chiller. The two top pipes are fed by the ref.pump.

No charge in the shell, all the charge is in the dropleg.

------------------

And on the girls, I'm not sure... the wintergear makes it kinda hard to determin :D

US Iceman

28-03-2006, 07:06 PM

Tycho

21-04-2006, 10:24 PM

Thanks for the information Tycho. The arrangement you describe sounds very similar to the water spray headers used on evaporative condensers.

Do you know the spacing on the spray nozzles, nozzle type, and flow rate through the nozzles?

I think the idea of using the large drop leg for the pump makes a lot sense. The volume of liquid required for the chiller is fairly small and the piping circuit is really short. The amount of refrigerant in suspension could be contained within the drop leg very easily.

I'm getting a pretty good picture of these now. Thanks again.

OK you got me with that comment. Almost fell out of the chair. :D

I (My company, as far as I know) Have no information when it comes to the design features of the chillers, as I said, it's not our product :) we are only informed when it comes to the waterflow, cooling effect and so such.
we do all the calculation on the piping and compressor/condencer capacity needed and ship it all off to SES in Holland and they build the units for us.
By the time we touch it, it's all placed on a skid allready welded to the hull or placed on vib.dampers. :)

I was looking through my papers just now, but couldnt find the flow chart... from the top of my head I'm gonna say ummmm... cant remember... was gonna say 653 cubic yards an hour... but that just sounds wrong in my head :)

When are you 'muricans gonna submit and start using the metric system anyways? :D
It all goes in 10's... how hard can it be :)
5280 feet to a mile? I mean c'mon...
wouldnt 10 Km to a metric mile be better? :D

Oh and the ladies are starting to dress down in anticipation of summer... I'm still wearing my winter jacket while at work... but dang... some of those girls HAS to be freezing with their bellies uncovered like that... makes me wanna jump out of my car, run over and drape my jacket around them and go "here... come home with me" :D

Tycho

21-04-2006, 10:45 PM

On the overfeed systems you guys work on do you install the pressure control valves at the pump discharge or at the end of the liquid line (as shown in the attached drawing)?

For those of you who have done it both ways, any complaints or suggestions with either method?

I am interested in learning more about the installations you work on and the problems you frequently encounter.

Thanks,
US Iceman

Our systems is just like the one in your drawing, except we have a 5 bar overflow valve connected to a 3/8" line between the pump and the non return valve. other than that we have nothing.

to paint a flow diagram:
Pump - ofv - flashgas trap (or whatever you call it:)) - nrv - main liquid tine - reg.valve - freezer/evaporator.

Not many problems on our installations... we also have 0.2" nozzles on each plate on our freezers. the reg valves we use to ensure a stable pump pressure to the last consumer on the liquid line.

Must say your piping looks very clean, and I envy you for gettion to work on land installations :) when we work on ships, the only line we refuse to yield on is the return line :) and every time we have shipbuilders tearing their hair out while we stand there saying "Thats the way it is"

US Iceman

22-04-2006, 12:18 AM

When are you 'muricans gonna submit and start using the metric system anyways?

Sorry about that.:D We are kind of used to it by now.

I will say talking to people on this forum has forced me to remember things I had forgotten.

...some of those girls HAS to be freezing with their bellies uncovered like that

Perhaps you should try that also. Maybe the girls will ask you to come home with them.:p

the reg valves are there to ensure a stable pump pressure to the last consumer on the liquid line.

I think this makes a lot of sense, but I've only seen this a couple of times here. I intend to use this on the next system for the reason you said.

I can well imagine the issues you have to deal with on marine systems. A lot more issues to deal with than just level and stationary.

bruceboldy

29-04-2006, 12:03 AM

having built recirculator packages for many years i have found that the majority of the problems have been not following the pump company guidelines.

Pumping volatiole liquid like ammonia or ***** is easy if you have enough npsh and keep the pd down with proper inlet sizing i think Iceman was all correct in his observations

I mostly saw npsh problems causing the most difficulties.
I also like the seperate suction lines in lieu of the common drop leg.

Sometimes i have found engineers overthinking the problems.

i never tried it but i want to put in some of the witt pumps simply hanging from the vessel. with the suction straight into the pump inlet. with a valve of course.

Have seen a few sketches of this but never seen it in practice.
Would really save us on installing cost. Anyone do it ?? and what problems or success have you had.

David in texas

US Iceman

29-04-2006, 02:46 AM

Hi David,

I mostly saw npsh problems causing the most difficulties.

I agree. These are normally the result of too much inlet loss, insufficient static head, or the pump is not setup properly allowing it to run out on the pump curve.

i never tried it but i want to put in some of the Witt pumps simply hanging from the vessel. with the suction straight into the pump inlet. with a valve of course.

I have a friend who told me was thinking of this for some Cornell pumps. He was planning on using high-performance butterfly valves for the suction valve instead of an angle valve.

If the suction valve was a globe style in the vertical suction line, I think you would loose more than you can gain because of the inherent extra length of the globe valve.

Using a high-performance butterfly valve would of course drastically reduce the equivalent pressure loss.

I should check in with him and see what the results were. Thanks for reminding me of that.

Nico

19-06-2006, 04:36 PM

I installed Witt pumps type HRP 3232. They hang directly under the separator (with only a ball valve in between if you want). Works superb, easy to install, insulate and to service. As we always use one stand-by pump, the piping as shown in the attached file is standard. It clearly shows the advantages of the pump directly fit to the separator. However, only one capacity available at Witt(for now that is). Find info in the WITT HRP manual at th-witt.com.

As for the pipe inserted into the dome; as to my understanding that was invented to prevent falling oil from the separator to enter the pump, but being collected at the bottom of the dome.

We always use leg's as drawn in the attached file. Very little (never) problems with oil in the pumps.

Also standard is the use of a pressure regulator near the pump. That, with a good(!) discharge head of the pump makes it easy to set the regulating valves, and the system will be very stable. No need to go to the end of the wet return line.

Nico

19-06-2006, 04:38 PM

****, forgot the drawing.
This will be better I think.

US Iceman

20-06-2006, 03:14 PM

Very nice drawings Nico. Thank you for sharing these.

Tycho

21-06-2006, 01:32 AM

****, forgot the drawing.
This will be better I think.

If I'm correct, the pumps are shown on the left side of the drawing, and you have the suction pipe taken out on the bottom of the drop leg... am I correct?

On ***** R-22 this will work fine, as the oil will gather in a "layer" around the top of the liquid head...

but on ammonia, you will have no way of draining the oil before it enters the pump and is flushed out into the system along with the liquid ammonia.

As I have shown previously in this thread.. (I know might sound like a "know it all" in showing this again... but I know it works)
http://upload.pbase.com/image/62209774.jpg
(added some english text in the picture)

Notice that there is a ring around the drop leg preventing oil from draining down into the dropleg, there is a seperate dropleg for draining of oil... and as a precaution, there is an oil drain valve on the dropleg just in case :)

Just have to say that I'm not a big fan of having any important pipes going out of the bottom of a dropleg like that, cause any "contamination" in the plant that finds it's way to the liquid tank will go straight into the pumps.

AMARNATH JHA

21-06-2006, 08:49 AM

Generally in the large overfeed system has liquid feed pumps which supply liquid refrigerant to the evaporators.
I think your system must have multiple evaporators. I have read that your system has separate lines connected to pumps. This is not advisable.There must be a common header between vessel & pumps.I have handled very large overfeed ammonia system.If any querry pls write to me.
Thanks
AMARNATH JHA

US Iceman

21-06-2006, 02:21 PM

I have read that your system has separate lines connected to pumps. This is not advisable.There must be a common header between vessel & pumps.

Can you explain your comments please?

Thanks.

Nico

22-06-2006, 03:52 PM

Both left and right side of the drawing are of a separator with pumps. One with horizontal and one with vertical shaft. The top is a detail of the oil dome with ammonia down pipe to the pump(s) in the middle. The system is for Ammonia indeed! As oil is more heavy than Ammonia, it will fall down to the bottom of the separator, and than sort of fall off the edge into the oil dome. Ammonia is collected just 10mm higher than the bottom of the separator. That's why the vortex breaker is so important.

Nico

22-06-2006, 06:58 PM

This detail will be more clear I guess.

US Iceman

22-06-2006, 08:28 PM

Tycho

22-06-2006, 10:58 PM

This detail will be more clear I guess.

Ah, thanks Nico, that was the part I was unsure about :)

In the previous drawings it seemed like you fed the pumps from the bottom of the dropleg :)

as you see in my drawing, we take it out on the side, and the bottom of the vortexbreaker has a triangular opening cut into it, allowing oil to drain to the bottom of the dropleg if it happens to enter into the vortexbreaker...

The votexbreaker you use, is it just a "cross" of flatsteel welded across the top of the suction line?

cant say other that different solutions same result :)

tho I cant say other than your solution looks as good as any, I'm stuck in my ways and I'm gonna do it Myyyyy waaaaaaaaaaaay :d

Tycho

22-06-2006, 11:14 PM

Hi Nico,

Thanks for posting the drawing on the sump details.

Do you work for a contractor, manufacturer, ???

What type of liquid feed control valve are you using to maintain a constant liquid level in the vessel?

wouldnt a system be in a somewhat equilibrium, depending on how many "consumers" are in use, in regards to the liquid level in the LP vessel? less consumers = higher level and all.

When I calculate the ***** charge I go by rule of thumb :) 1 vertical freezer = 5 x 57 Kg (5 x 125 pounds) or 5 bottles of *****.
a normal cargohold on a ship is approx 2-5 bottles

Usually we charge from 800 Kg (17681 pound) tanks on a new system, and as it goes, I'll just open the consumers one by one till I loose pump pressure.

when I loose pump pressure I'll wait to open the next consumer till I have pump pressure again, then rinse and repeat till you have pump pressure with all consumers running.

Of course you gotta keep an eye out so you dont charge more than the capacity of the LP reciever/tank... I never charge more that 80% of the LP reciever volume, it's the law, and no way no how anm I going to be struck down by some insurance company for either destroying a ship, or wrecking a compressor :)

US Iceman

23-06-2006, 01:47 AM

Hi Tycho,

wouldn't a system be in a somewhat equilibrium, depending on how many "consumers" are in use, in regards to the liquid level in the LP vessel? less consumers = higher level and all.

Yep, if the system is something similar to a critically charged system, which may be what you are describing.

On some systems there might be a need to control how much liquid enters the vessel, which is how I see most of the liquid overfeed systems (except those designed for a critical charge).

I have seen some level controls using a simple float switch controlling a solenoid. Others might use a capacitance probe and a modulating liquid feed valve.

I was wanting to see what type Nico used.

I never charge more that 80% of the LP receiver volume...

Is this when the system is being commissioned? Hopefully, not 80% full when operating at full load.:eek:

Nico

23-06-2006, 04:22 PM

As a one man company I work for several big contractors and manufacturers,as well asfor larger users. I (try to) help with design / commisioning / training and trouble shooting. Always traveling. Lots of fun!

The level control of the separator depends on the country (regulations), the system and the budget.

When there are lots of small clients, than you can work with a simple level switch, solenoid valve and regulator valve from a liquid receiver.
If you have an analog level sensor, you can use the same valves and program a proportional pulse pause regulator or -if you have the monney- a regulator valve. I know that the Danfoss SV float regulator with PMFL valve works perfect for decades.
A high pressure float valve gives you the lowest NH3 charge (I like that), but most clients want to see a good level in a large Ammonia vessel (Middle East and Africa). Your separator will be exspensive (large) with a HP float valve, but you don't need a liquid receiver.

US Iceman

23-06-2006, 04:45 PM

Your separator will be expensive (large) with a HP float valve, but you don't need a liquid receiver.

I have done this on several projects and really like it. If the float valve is sized properly you can allow the compressor discharge pressure to decrease to really low pressures (if everything else is sized correctly).

Works great!

Tycho

23-06-2006, 10:58 PM

Hi Tycho,

Is this when the system is being commissioned? Hopefully, not 80% full when operating at full load.:eek:

Yes, on commisioning, and 80% with all the consumers defrosted :)

-I know that the Danfoss SV float regulator with PMFL valve works perfect for decades.

on all our plants older than 5 years we used the danfoss SV-1 HP float with a pilotreciever with about 40 liter capacity, worked like a charm

we still use the sam pilotreciever, but are now using danfoss AKS level transmitters... lots of trouble with them in the beginning, with them failing after a short time and so such... but with feedback from all the users danfoss has weeded out most of the problems and we now use this solution up against either AKV valves or PM valves.

Nico

25-06-2006, 08:39 AM

US Iceman

25-06-2006, 09:49 PM

Hi Tycho,

Yes, on commissioning, and 80% with all the consumers defrosted

OK, I'm assuming that when the vessel is 80% full the load is is very small. If the vessel is 80% full, I would think this would limit how much compressor capacity you could have ON during this time.

The separation space in the vessel would be very small. I'm also assuming you have really good controls to limit how fast the users can be activated and how fast the compressor(s) load up to prevent liquid slugging.

I suppose the reason I raised the question in the first place is... Someone who is not familiar with the particular requirements you normally see may assume this is standard practice for any overfeed system.

Josip

27-06-2006, 11:31 PM

Hi,

on attached drawing you can see the old fashioned way of level control, pump protection and compressor protection if customer does not like to spend to much money ;)

1033

All is working Ok.

Yes, one pump is stand-by.

Best regards,

Josip :)

US Iceman

28-06-2006, 03:37 AM

Hi Josip,

Are those strainers/filters in the suction line to the pumps?:confused:

Josip

28-06-2006, 08:24 AM

Hi, US Iceman

Hi Josip,

Are those strainers/filters in the suction line to the pumps?:confused:

Yes, and mostly on new plants during start up, we have to clean them often. Also they are bigger then the pump inlet connection. Pumps are Hermetic CAM 2/3 with Qmax and Qmin orifice.

That filter we install as the lowest part of the pump inlet pipe.

See you are :confused: , but we have to protect pumps (on ammonia plants with dirty black steel separators and pipes) on some way, but that is another long story ;) .

Best regards,

Josip :)

US Iceman

28-06-2006, 03:49 PM

Andy

28-06-2006, 06:29 PM

Hi Josip,

What you say makes a lot of sense. The carbon steel pipe can be dirty with the scale that comes in the pipe.

Do you normally find a lot of weld slag in these strainers?

Do you use weld rings (not sure what you might call them)? They fit inside the end of the pipe to help with alignment and help to prevent weld slag from blowing into the pipe.

Have you tried pickled pipe before? This is very clean, but must be protected until the insulation and paint are applied to the pipe.

Here in the US we normally allow the strainers before the solenoid valves to catch this junk.

The strainers in the pump suction line add pressure drop which decreases the net positive suction head on the pumps. This can cause cavitation.

You are correct. This could be a long story.;)
I have always seen a strainer in the pump suction line, pumps can be quite expensive and systems can be dirty:eek:
Probably a good idea to remove the strainer after a year or so use or fit a large mesh nut and bolt catcher at this point.
Small mesh strainers can become oil traps on LT systems, best to avoid smaller mesh strainers on these types of applications.

Backing rings were fitted in the past, but now we are more advanced:D
TIG welded roots are the norm with the filler and cap in stick:)

Kind Regards Andy:)

US Iceman

28-06-2006, 06:42 PM

Hi Andy,

I have always seen a strainer in the pump suction line, pumps can be quite expensive and systems can be dirty:eek:
Probably a good idea to remove the strainer after a year or so use or fit a large mesh nut and bolt catcher at this point.
Small mesh strainers can become oil traps on LT systems, best to avoid smaller mesh strainers on these types of applications.

Absolutely no doubt the systems can be quite dirty.

What size mesh screens do you use for low temp and higher temp. applications?

I'm not disagreeing with the use of strainer, just saying the pressure drop should be accounted for in the pump NPSH calculations. The pressure drop of this device could add a meter or more of head required for the pump suction. (just a guess, but it's probably something like this)

Backing rings were fitted in the past, but now we are more advanced:D
TIG welded roots are the norm with the filler and cap in stick:)

Are you saying I'm old fashioned?:p

I would prefer exactly what you are doing Andy. This makes a much better weld. I always suspected the backing rings could collect junk under the collar and create a potential corrosion site from inside. I also think the weld rings are harder to read in an x-ray.

Do you use consumable inserts for the first pass TIG weld? Or, just simply fit them up by hand? Or, use a pipe alignment fixture.

Andy

28-06-2006, 10:26 PM

Hi Andy,

Absolutely no doubt the systems can be quite dirty.

What size mesh screens do you use for low temp and higher temp. applications?

I'm not disagreeing with the use of strainer, just saying the pressure drop should be accounted for in the pump NPSH calculations. The pressure drop of this device could add a meter or more of head required for the pump suction. (just a guess, but it's probably something like this)

Are you saying I'm old fashioned?:p

I would prefer exactly what you are doing Andy. This makes a much better weld. I always suspected the backing rings could collect junk under the collar and create a potential corrosion site from inside. I also think the weld rings are harder to read in an x-ray.

Do you use consumable inserts for the first pass TIG weld? Or, just simply fit them up by hand? Or, use a pipe alignment fixture.

We use filters in the range 40 to 100 micons range, quite fine really:D I favour big bore long drop legs, put the surge drum on one floor and the pumps on the floor below with a straight line down into the pump (vortex breaker at the top of the pipe, ball valve at the pump, all say 6" on a 2 1/2" or 3" inlet pump.

TIG root is not used with a ring. The pipe is champhered, with the slope creating a vee for the weld to be built up in. The pipe or fitting is gapped by 2 to 3mm all the way round, to allow a weld bead to form evenly on the inside of the weld, much like a mini cap on the inside.

Kind Regards Andy:)

Josip

28-06-2006, 10:46 PM

Hi, US Iceman :)

Not to much of weld slag (nowdays mostly use TIG for root weld) but lot of particles remain after grinding to make a cone for weld. this you have even with sand or CO2 blasted pipes.

I have to admit about weld rings and backing rings I do not know, because, never used them (please, post a link or some picture ;) ). We used sometimes alignment tool for pipes (in case of X-ray) but mostly doing that by hands of experienced pipe fitters ;) , faster and cheaper.

Yes, installing filter in front of the pump will reduce NPSP but filters are of very big size and pressure loss is very small.

I think Andy, did explain almost all but in much better english than me :D , thanks Andy;)

We use filters in the range 40 to 100 micons range, quite fine really I favour big bore long drop legs, put the surge drum on one floor and the pumps on the floor below with a straight line down into the pump (vortex breaker at the top of the pipe, ball valve at the pump, all say 6" on a 2 1/2" or 3" inlet pump.

TIG root is not used with a ring. The pipe is champhered, with the slope creating a vee for the weld to be built up in. The pipe or fitting is gapped by 2 to 3mm all the way round, to allow a weld bead to form evenly on the inside of the weld, much like a mini cap on the inside.

best regards,

Josip :)

US Iceman

29-06-2006, 01:34 AM

Andy,

TIG root is not used with a ring

Depnding on the customers pipe welding specification it may be called for. Here are some links to suppliers of these. We used these when the specifications were very stringent in chemical plants.

They are ordered in the same material as the pipe, with the inserts being consumed during the fusing process. They make a really nice joint, which sometimes is hard to tell the joint from the main pipe.

http://www.robvon.com/html/proc_consum.html

http://www.robvon.com/html/consumables.html

Main page:
http://www.robvon.com/
Also see their link for backing rings. This is listed on the left hand side of the webpage.

http://www.imperialweldring.com/new/consumable.html

Josip,

...you have even with sand or CO2 blasted pipes.

I have used sand blasted pipe before but never CO2. Does this work well? I'm assuming they use dry ice for the blasting right?

Josip

29-06-2006, 09:21 AM

Hi, US Iceman

thanks for links, very good things (kind of VFD for pipe fitters and welders ;)

http://www.cryogenesis.co.uk/

Yes they use dry ice and complete work is very clean. I did not use it but got some info about from my friends.

Best regards,

Josip :)

Tycho

01-07-2006, 11:42 PM

Have not seen the AKS 41(?) yet.
Do you use the EKC 347 as an controller, or direct to the PLC?
What do you use as control valve?

I have used the RTK NI1321 a lot. Problem with this sensor is that they are density sensitive. If the evaporating temperature changes, than you get an offset.
We always controll through the PLC. With one click you program a controller, so why install others?
For regulator valves we are often directed by the client's standard; Samson, Worchester etc. We use RTK Kornwestheimer as our "standard".

We make our own control system with PLC and the AKS is wired directly to the PLC, with readouts and calibration made possible from the plc display.

we have also retrofitted older systems with an AKS working with a EKC 347... Works fine with either...

Had some problems with the EKC tho, not sure what was done, but the electrician doing the wiring also powered up the system before I got there, and managed to burn out the contacts on the EKC, so I had to use the alarm contact for either hi or lo level, cant remember... the customer is non the wiser and the system works :)

We use mainly Danfoss valves, ranging from ERV to PM (depending on system size)

Tycho

01-07-2006, 11:53 PM

Hi Tycho,

OK, I'm assuming that when the vessel is 80% full the load is is very small. If the vessel is 80% full, I would think this would limit how much compressor capacity you could have ON during this time.

The separation space in the vessel would be very small. I'm also assuming you have really good controls to limit how fast the users can be activated and how fast the compressor(s) load up to prevent liquid slugging.

I suppose the reason I raised the question in the first place is... Someone who is not familiar with the particular requirements you normally see may assume this is standard practice for any overfeed system.

To put it in other words, we design our plants so that we can fit all the liquid in the LP tank, incase the consumers need major service.

Mostly, getting all the liquid back to the tank is done by service personel from us or whatever other company is available at the time.

it is also safe for the onboard engineer to do this, close the pump inlets, shut of the pumps, have one compressor running in auto and our load controll system takes care of the rest.

The 80% Charge I said is also more like a rule of thumb :)

Lets say I have a system with 20 plate freezers, but the tank can only hold a charge for 18 of those (18 if the tank is topped off), I will then fill enough to run 15 freezers at one time, inform the ship owner and chief engineer that "your tank can only hold enough charge to run 15 freezers at a time yadda yadda yadda, live with it"

ofcourse there are some ships running with a 120% charge, cause the chief engineers are ordered by the ship owner to charge more *****, to get more consumers running...

But when we charge, we charge just so that we can fill the tank to the brim without getting liquid back to the compressors.

if that made any sense :)

Tycho

02-07-2006, 12:06 AM

We use filters in the range 40 to 100 micons range, quite fine really:D I favour big bore long drop legs, put the surge drum on one floor and the pumps on the floor below with a straight line down into the pump (vortex breaker at the top of the pipe, ball valve at the pump, all say 6" on a 2 1/2" or 3" inlet pump.

Kind Regards Andy:)

I always put ladies nylon stockings inside the pump outlet/ice filter (our pumps can handle the dirt, so we only have outlet filters :D brag brag brag :D), on new plants, and re built plants.

I supply the onboard engineer with some spares so he can change them if there is to much dirt, or if there isnt, I'll change them the next time they are in, I run the plant with the stockings for a minimum of a month, usually two months.

The reason for this is that you can run the plant with a bigger mesh filter afterwards.

but it's always embarrasing when you go to the store and stock up on nylon stockings like another perv :D

This one time I accidently bought ancle high when I needed thigh high stockings, and I had to go back to get the ones I needed... same very very handsome looking girl operating the register when I got there with my second armful of stockings.

Her: "Oh... I see you really like those"
(I turned lobster red in less than a second.. it was like *POOOF* I'm red)
Me: "oh.. hehe no, ummm filter... mesh and yeah, they... I use them... they are fine meshed... umm as a filter"

We use condoms to cover up the safety valve outlet when checking for small leaks in the safety valve, and one time we had a poor guy have to go to the store to get nylon stockings, condoms and dishwasher soap...

I'm just glad I wasnt that guy :D

oh, this should have been in the chit chat section, buy hey :D

US Iceman

02-07-2006, 12:12 AM

if that made any sense

Perfectly...

BTW, do you have a preference on the color of the stockings you buy?....for filters that is.:D

I have used and recommend the condoms over the relief valve vent lines also. It sure does get a lot of comments when one valve is leaking.:rolleyes:

corineramer

15-08-2006, 09:08 AM

guys do you know where i can buy equipments for liquid overfeed system online that has a brochure... nid help pls...

abdulazman

15-08-2006, 09:40 AM

Hi guys,

Anybody got any ideas or prices or website for Ebara Direct Fire CHILLERS-RAD-G028.

Could use any help.

slsammoniaman

15-08-2006, 02:30 PM

One location to find information on recirculation packages is to check out the Evapco website at evapco.com

US Iceman

16-08-2006, 01:50 AM

Hi corineramer,

What type of equipment are you looking for?

Pumps, evaporators, valves, pressure vessels, ???

If you can provide some more details, we can provide more information to help you.

corineramer

16-08-2006, 04:42 AM

well US Iceman im just a student... we have a plate that we need to finish.. we're to design one utility of an ice cream plant and we picked the refrigeration system and we used the liquid overfeed system our proffesor told us to use that system... we dont have any idea where we could find all the equips that we need the eveporator, condenser,separator, ammonia pumps, etc.. we have a minimum capacity of 50 TOR... can u help me?? is there a site where they offer all of the equips?? we only need the brochure or specs.. thnx for replying we badly need help.. also 1 of our problems is we havnt been to an ice cream manufacturing plant we dont hav any idea what the machines look like inside the plant how they are arranged, that sort of thing... we need it for our plant lay-out...

Josip

16-08-2006, 02:12 PM

Hi, corineramer :)

follow this link and enjoy ;)

http://www.refrigeration-engineer.com/forums/showthread.php?p=47621#post47621

Best regards, Josip :)

US Iceman

16-08-2006, 08:01 PM

Here is a link for evaporative condensers:
http://www.evapco.com/media/pdf/173B-EM-EvapCon-EngMan.pdf

Here is a link for liquid overfeed packages (vessel & pumps)
http://www.evapco.com/media/pdf/asme-recircular.pdf

Here is a link for the celing hung evaporators for the holding freezers:
http://www.evapco.com/media/pdf/nt-series.pdf

Here is a link to some ice cream production machines:
http://www.tetra-pak.com/ Look under the links for "Production".

Here is a link to some compressor information:
http://docnav.grasso-global.com/DocNav_Eng/Sources_e/Screws_e.htm

Here is a link for control valves + other good information:
http://www.danfoss.com/North_America/BusinessAreas/Refrigeration+and+Air+Conditioning/IR+Literature+and+Software+Selection.htm

That should keep you busy for awhile. :D

Josip

16-08-2006, 09:46 PM

Hi,

That should keep you busy for awhile. :D

Maybe for whole life like us ;)

Best regards, Josip :)

US Iceman

17-08-2006, 01:46 AM

Hi Josip,

I've been doing this for a long time and I still learn something new everyday. I bet you do too.;)

The next big project to figure out is process AC systems, something like your project you are working on now. I'm still going to school after all this time.:rolleyes:

corineramer

17-08-2006, 02:43 PM

hi josip and us iceman,

thnx

corineramer

17-08-2006, 02:43 PM

hi josip and us iceman,

thnx for the help... i hav seen the topics for the ice cream plant and im im currently

corineramer

17-08-2006, 02:55 PM

hi josip and us iceman,

thnx for the help... i hav seen the topics for the ice cream plant and im currently checking it out now... we hav a problem in finding the exit temperature of our refrigerant which is ammonia.. the temperature entering the evaporator is -30C and our only data is the temp of the ice cream which is the temp entering is 4degC and exits at -5degC... by the way this is the freezing of the ice cream which is a closed system... correct me if im wrong but we're doing the mass balance and energy balance.. is this correct? are we in the right track hope u reply because we're running out of time... thank you so much sir or sirs...

US Iceman

17-08-2006, 04:45 PM

Hi corineramer,

the temperature entering the evaporator is -30C and our only data is the temp of the ice cream which is the temp entering is 4degC and exits at -5degC... by the way this is the freezing of the ice cream which is a closed system...

If the liquid ammonia is being pumped into the evaporator at -30C, it also exits the evaporator at -30C on an overfed system.

Therefore, if your ice cream is entering at 4C and leaving at -5C, that is your heat load for the mass flow you have.

To do the energy balance, the heat load divided by the latent heat of ammonia at -30C will provide the vapor mass flow out of the evaporator. You have to be careful, as the ammonia liquid mass flow will be even higher.

The whole purpose of an overfeed system is to pump in more liquid than will boil off.

corineramer

18-08-2006, 08:29 AM

hi us iceman,

i thought that there would be a temp change in the ammonia because of the heat transfer through the evaporator... so, there would be not temp change in the ammonia... thnx for the help.. just post here for my progress and questions.. tnx again

US Iceman

18-08-2006, 09:18 PM

The refrigerant is boiling at -30C. As long as the refrigerant pump is supplying sufficient liquid into the evaporator, the refrigerant exits the evaporator at -30C.

If the pump is not selected properly, or the hand expansion valve is too far closed, you can get some superheating of the gas leaving the coil. With superheat, you would have a temperature change.

The boiling process is all latent heat, so the temperature does not change.

Good luck on the school project.

corineramer

19-08-2006, 10:41 AM

hi us iceman,

now its getting clear... our prof told us that there would be no temp change but the pressure will be higher so we need a back pressure valve before the vessel i think to minimize the pressure until it equals the temp of our vessel, right? then for the compressor, the suction pressure will be the pressure at the vessel and the discharge will be depending on the ambient condition in the condenser... pls correct me if im wrong... wow! it feels good now because im beginning to understand this system... the next will be the computations... im having fun in this design... hehehe

US Iceman

19-08-2006, 11:44 PM

...but the pressure will be higher so we need a back pressure valve before the vessel i think to minimize the pressure until it equals the temp of our vessel,

I'm not sure why he thinks you need a back-pressure regulating valve. These are used to keep the pressure higher in the evaporator. BY increasing the evaporator pressure with this valve, you increase the saturated evaporating temperature. This would reduce the evaporator capacity.

Unless, there is something I don't know about in your system design.

Without this valve, the evaporating pressure would be equal to the compressor suction pressure plus the suction line pressure drop. If your ice cream freezer is the only evaporator on the system, the -30C evaporating temperature will require a slightly lower compressor suction temperature.

The suction line pressure loss is equal to a specific value for the equivalent temperature loss.

Adding a back-pressure valve at -30C increases the suction line pressure loss and also decreases the compressor capacity.

corineramer

20-08-2006, 02:57 PM

hi us iceman,

now im getting confused... this is his explanation, when the ammonia leaves the evaporator there will be a slight or big increase in the ammonia pressure I think...then he said that the pressure of the refrigerant should be equal to the vessel thats why we nid this valve... he said that the compressor suction pressure should be equal to the saturation pressure at -30C... By the way he said that this valve decreases the pressure of the ammonia so that it will be equal to the pressuer of the vessel... what does a back-pressure valve do? is this correct? i have another question he added that we can use 1 condenser in our freezing, hardening and storage... he said that when the ammonia has been compressed by the 3 compressors we can join them in 1 line to the condenser... is this possible? then im thinking on how to distribute the ammonia that exits from the condenser... im really confused... can I send u the details of our design to your email? if its ok with you.. and have your comment on it.. thank you for the time..

US Iceman

20-08-2006, 04:26 PM

Hi corineramer,

Let's see if we can clear this up.

when the ammonia leaves the evaporator there will be a slight or big increase in the ammonia pressure I think

No. When the liquid is pumped into the hand expansion valve, the pressure is reduced to the evaporating pressure (the saturation pressure of the ammonia at -30C). The pressure does not rise. In fact, the pressure decreases some due to the pressure loss of the refrigerant flowing through the evaporator circuits.

then he said that the pressure of the refrigerant should be equal to the vessel thats why we need this valve... he said that the compressor suction pressure should be equal to the saturation pressure at -30C

If we neglect all pressure losses, the evaporator pressure is the same as the vessel pressure is the same as the compressor suction pressure.

By the way he said that this valve decreases the pressure of the ammonia so that it will be equal to the pressure of the vessel...

My guess is, he is making a common mistake on liquid overfeed systems. The refrigerant pump provides the pressure to move the liquid out to the evaporators. When the liquid flows through the hand expansion valves, the liquid pressure is reduced down to the evaporating pressure on the outlet of the hand expansion valve.

Once the liquid is in the evaporator, it is also at the evaporating pressure equal to -30C. The pump pressure is all gone. All of the gas and liquid exist at the evaporating/suction pressure (Again, I'm assuming we have neglected the pressure losses due to flow.).

The refrigerant flashes in the evaporator, not after the back-pressure valve.

[/quote]
what does a back-pressure valve do?
[/quote]

A back-pressure regulating (BPR) valve controls the pressure of the refrigerant upstream of the valve. If the BPR valve is set for a pressure higher than the evaporator pressure, the liquid will not boil at -30C. Adding BPR's to a liquid overfeed system creates a lot of additional problems that must be solved.

If you only have one evaporating temperature on your system (-30C) you do not need a BPR to make the system work.

he said that when the ammonia has been compressed by the 3 compressors we can join them in 1 line to the condenser

That's OK. This is done all of the time. You must make sure the piping on the condenser is absolutely correct to allow gravity drainage. Condenser piping is probably one of the most mis-understood areas of indutrial refrigeration systems.

The outlets of each condenser coil (most larger condensers have at least two outlet connections) must be trapped at the bottom. The equalizing line from the receiver to the condenser inlets must also be large enough to allow the liquid to drain properly.

Josip

20-08-2006, 08:55 PM

Hi,

seems to me that the teacher is more confused then students:)

Best regards, Josip :)

US Iceman

20-08-2006, 09:22 PM

HI Josip,

That teacher would not be the first either.:D

NH3LVR

20-08-2006, 09:44 PM

Andy

20-08-2006, 10:02 PM

I have not been on in a while. Seems Summer breakdowns have had us running. North Star Icemakers (My favorite). Compressor seals, etc.
So lets open a can of worms on Hand Expansion Valves used with Liquid Recirculation.
Just once I have installed a AFR3 Flow Control from Refrigeration Specialties. In all other cases we have used H.E. Valves.
I believe you can calculate the orifice size of any H.E. Valve from the Manufacturer's literature and from then calculate the approximate opening turns to start with, although I have never done it.
The proper method of adjusting after startup is to open the H.E. Valves until Evap performance starts ti fall off. (Using Air Temp measurements across a coil for instance). Once the coil is overfed ("Brining") the performance starts to fall off.
Of course if the Liquid pump loses pressure as the valves are opened you need to start over.
Last Summer I started up a Medium Temp warehouse with a dozen or so NH3 Evaps. It would have taken several days to go through the procedure, especially as product was being moved in as we were starting up.
So we make an educated guess as to the settings. That is made easier by the fact that this customer does not insulate the control valves, and that they are installed on the roof. You can see the effects of the liquid return and adjust accordingly.
Liquid recirculation is certainly more tolerant of underfeed and overfeed than other methods but I suspect many of the systems I work on are underfeeding.
Does anyone have a better adjustment system, or is this way most of you do it?

I suppose the way I go about it would be. Check my selection on the Dircal software, it gives valve openings.
Setthe valves up something less than that. Observe the frosting on the evaporator, should be less than fully fed. Keep opening the valves until the evaps are all frosted, all just fully flooded and all the same. Then give the valves another 1/2 turn or so open. I would then check the air on and offs on the central monitoring system, they should be all similar, the poor ones I would then look at more closely and adjust to get the best performance. To my mind you should have no big drop off when overfeeding, if you do you are just wasting pump power at the expense of evaporator and system performance.

Kind Regards Andy:)

US Iceman

21-08-2006, 02:35 AM

HI NH3LVR,

Welcome back.

I believe you can calculate the orifice size of any H.E. Valve from the Manufacturer's literature and from then calculate the approximate opening turns to start with,...

Yes, that is correct. A little number crunching will get you there.

Does anyone have a better adjustment system,...

I'm not saying this is better, it's just the way I would go about it.

The first thing to do is to check the refrigerant pump pressures and operation to make sure it is dialed in properly. If a small compressor has been started to pull down the suction pressure in the vessel, the pump will start to cool off. As the suction pressure approaches the normal operating condition, the pump should be very close to it's operating temperature.

It is a lot easier to do a cold start-up on a liquid overfeed system than a hot start-up.

The evaporators that are the farthest away from the pumps, or the ones that are at the highest elevation are the coils that need the most liquid pressure to HE valves.

I would start at one of these coils with the adjustment of the HE valves. Shut off the fans on the coil and begin to crack open the HE valve. As the frost pattern begins to approach the suction outlet header, you are getting fairly close. Turn the fans on and move to the next coil.

Continue to work your way back to the pumps.

Once you get this far, you can go back and check the air temperature drops across the coil. You can determine the exact number (TD) you should be trying to get. By the time you get ready to start measuring air TD's, the system is starting to stabilize.

I certainly don't think this question is, or will be a can of worms, but it is a very good question. I think the biggest problem is the HE valves are normally adjusted too far open, causing the coils to overfeed and nuisance problems with the pumps.

BTW, we have another thread going under English Language... . You should post this saying " a can or worms" in there.

corineramer

21-08-2006, 09:27 AM

hi us iceman,

our equips are not complete do you know a site where i can have brochures of liquid receivers?? tnx.. BTW, the hand expansion valve is located after the pump before the evaporator, right?

US Iceman

21-08-2006, 02:09 PM

If you go back to the Evapco web site, they will have a complete brochure for various pressure vessels. You should be able to find something in there.

The liquid feed valve train for the coil usually consists of these items:

Service valve
solenoid valve with close-coupled strainer
Hand expansion valve
Check valve (required only if hot gas defrost is being used)
Service valveHaving a service valve on the beginning and the end of the valve train allows you to service any of the valves in that valve group, without pumping down the evaporator coil.

Here is a link for some information about one type of refrigerant pump. It shows some of the typical installation details. These details vary somewhat if you use a hermetic pump.

corineramer

22-08-2006, 03:17 PM

hi us iceman,

wat link are you referring to? the evapco site or other sites? we're really confused.. we're not that familiar with this system because the cycle that we have studied is the vapor compression cycle... im sorry if our progress is slow... i thought we understand this cycle but not there's a lot of things we need to learn.. also we have 3 evaporators that has 3 different temp we're trying to figure out if we will use one condenser in this 3 evaps or provide them with their own condensers... tnx

US Iceman

22-08-2006, 04:11 PM

Hi corineramer,

Use this link to review vessel information.
http://www.evapco.com/media/pdf/asme-pressure.pdf

...the cycle that we have studied is the vapor compression cycle

Nothing has changed. This is still a vapor compression cycle. It's just not a direct expansion system.

If you try to add other coils which are operating at higher evaporting temperatures, you would need back-pressure valves on those coils operating at the higher temperatures (greater than -30C).

This really complicates a liquid overfeed system and the explanation of how it operates.

If this is supposed to be a school project your professor has actually given you one of the hardest problems to work on.:(

ipelatak

23-08-2006, 03:24 AM

Hi US Iceman,

I'm a groupmate of corineramer. We are really confused about the liquid overfeed refrigeration system. This is what we understand about it.

You will supply more liquid refrigerant flowrate in the evaporator than required to enchance heat transfer by providing a higher velocity in the coils.

In our subject, we need to design a refrigeration system(liquid overfeed) for ice cream manufacturing plant.

We have used a separate refrigeration system for each process, freezing, hardening, storage.

Our professor said that we can tap the discharge line in the compressor because it has the same discharge pressure.

Is it possible if we only have 1 to 2 condenser? is it possible if we only have one condenser and 3 expansion valve?

Back to liquid overfeed. for example, in our freezing process, we have a -10 degree C ammonia. The ammonia pump will pump the refrigerant to the evaporator and increase the pressure of the refrigerant thus increaseing the boiling point. So we need the hand expansion valve to bring back the saturation pressure at -10 degree C? In liquid overfeed, the refrigerant that comes out from the evaporator is a mixture of liquid refrigerant and vapor right? then it will go to the separator to separate liquid and vapor. Then the compressor will suck the vapor at saturation pressure of the -10 degree C. The discharge pressure will depend on the ambient temperature in the condenser. We have a 35 degree C. Then after passing in the condenser, it will pass to the expansion valve to bring down the pressure to the saturation pressure at -10 degree C then stored in the liquid receiver for use when the level of refrigerant in the separator is low. is this correct?

Sorry if we ask so many question, we are really confused and thank you so much for spending time and helping us. Thanks :)

rbartlett

23-08-2006, 09:39 AM

corineramer

23-08-2006, 02:43 PM

hi us iceman,

BTW ipelatak is my groupmate... his message is the one after your latest post pls read his message, tnx... actually our professor told us to pick an industrial plant and design one of its utilities so that it would be easier for us because if he would choose what we're gonna design then it will be very difficult but unfortunately we picked a hard one not the easy ones... we have problems in selecting the compressor because our reference is the suction pressure, is this correct? for example our suction is 42.3psig and discharge press is 190psig... our prof told us to base our compressor to this given data but when im looking for the compressor they are giving the refrigeration capacity not the pressure... is this correct? or convert the TOR to refrigerating capcity and use it in selecting the compressor.. we're really confused... sorry if we'er disturbing you..

guapo

23-08-2006, 03:24 PM

Hi kurimaw & latek,

Expansion valve before liquid reciever is WRONG!:eek:

I think you should go back from the very basic of refrigeration before jumping out for the design. You shoul know the function of every components of refrigeration.;)

Are you a first year college student? what school?:(

Don't rely too much in your prof. you should do your own researh & study.

Regards,
Guapo:)

US Iceman

23-08-2006, 03:48 PM

Hi ipelatak,

You will supply more liquid refrigerant flowrate in the evaporator than required to enchance heat transfer by providing a higher velocity in the coils.

You are supplying more liquid into the evaporator than the amount that boils off. The purpose of using a liquid overfeed system is that the excess liquid provides a complete wetting of the heat transfer surface of the evaporator coil.

In a DX system, you loose some of the capacity of the evaporator due to the need to provide superheat. If you look at the evaporator ratings you will see a DX coil has about 15% less capacity than the same coil for liquid overfeed.

We have used a separate refrigeration system for each process, freezing, hardening, storage.

OK, that makes the design process much easier then.

Is it possible if we only have 1 to 2 condenser? is it possible if we only have one condenser and 3 expansion valve?

You can have one condenser or multiple condensers. All of your compressors can be connected to a common discharge line, which is piped to a single large condenser, or smaller multiple condensers. The condenser(s) would drain down to a common high pressure receiver. From this receiver, you can feed liquid to your separator (the one with the refrigerant pumps).

The ammonia pump will pump the refrigerant to the evaporator and increase the pressure of the refrigerant thus increaseing the boiling point. So we need the hand expansion valve to bring back the saturation pressure at -10 degree C?

The pump provides the pressure to transport the liquid to the hand expansion valves. The pressure supplied by the pump overcomes the friction and other losses to ensure the liquid is getting to the hand expansion valves.

You are correct about the increase in boiling temperature. The added pressure generated by the pump actually seems similar to subcooling of the liquid. Therefore, the liquid will not boil until the pressure is reduced. The hand expansion valve is really a balancing valve. You are using it to adjust the flow rate into all of the evaporators.

In liquid overfeed, the refrigerant that comes out from the evaporator is a mixture of liquid refrigerant and vapor right? then it will go to the separator to separate liquid and vapor. Then the compressor will suck the vapor at saturation pressure of the -10 degree C.

OK, so far.

The discharge pressure will depend on the ambient conditions and the capacity of the condenser. If you are using an evaporative condenser, you are concerned with the condensing temperature and entering wet bulb temperature of the air.

Then after passing in the condenser, it will pass to the expansion valve to bring down the pressure to the saturation pressure at -10 degree C then stored in the liquid receiver for use when the level of refrigerant in the separator is low. is this correct?

In a typical system, the liquid will flow into a high-pressure receiver after it drains out of the condenser. When the level in the separator begins to decrease, a control system will allow liquid from the high-pressure receiver to flow into the separator to maintain the liquid level. The liquid in the high-pressure receiver would be at 35C.

I'm not sure whay you have an expansion valve after the condenser as you described. (Note: I have done something like this before, but quite different than what you are describing.

Sorry if we ask so many question, we are really confused and thank you so much for spending time and helping us.

My pleasure. I hope I have not added to the confusion.:D

ipelatak

23-08-2006, 04:41 PM

Hi US iceman,

I'm not sure whay you have an expansion valve after the condenser as you described. (Note: I have done something like this before, but quite different than what you are describing.

Uhm we thought we should need to use a expansion valve after the condenser to reduce the pressure back to the refrigerants saturation pressure at -10 degree C.

So after condenser, we should put a high pressure liquid reciever. The temp of the refrigerant is 35 degree C = to the ambient temperature of the condenser. When the level in the separator begins to decrease, a control system will allow liquid from the high-pressure receiver to flow into the separator to maintain the liquid level. The temp in the separator is -10 degree C and the temp in the high pressure reciever is 35 degree C. We thought we need a expansion valve to bring back the pressure and temperature of the refrigerant in high pressure reciever equal to the refrigerant's temp and pressure in the separator.

Do we still need a pump after the high pressure reciever or the pressure will transport the refrigerant back to the separator and bring back the temperature and pressure because of the pressure drop in the pipe?

thanks a lot masters, were really thankfull that you guys are helping us. thanks a lot :)

US Iceman

23-08-2006, 04:41 PM

Hi corineramer,

I had a long explanation of the compressor rating procedures written out and when I tried to post it, something happened and I lost the whole thing.

The short version is:

You have to include the suction and discharge pressure losses to correct for the actual compressor performance.

Discharge pressure = condensing pressure + piping pressure lossses

Suction pressure = evaporating pressure - piping pressure losses

Convert the suction and discharge pressure to their equivalent saturation temperatures. These are the temperatures you want to use for selecting the compressors.

US Iceman

23-08-2006, 04:52 PM

The temp of the refrigerant is 35 degree C = to the ambient temperature of the condenser.

No. The condensing temperature is not the ambient temperature. If you want to condense at 35C, then you have to select the condenser for a condensing temperature of 35C with whatever your local wet bulb temperature is for. In the Phillipines it is probably quite high.

We thought we need a expansion valve to bring back the pressure and temperature of the refrigerant in high pressure reciever equal to the refrigerant's temp and pressure in the separator.

You need an expansion device to feed the liquid into the separator to maintain the liquid level. The liquid comes from the reciever at 35C, goes through the expansion valve, into the separator.

Do we still need a pump after the high pressure reciever...

Ahh, crap... Know I see. You guys are designing a Hysave system. That is not a liquid overfeed system, it's a DX system.:mad:

The refrigerant pump in an overfeed system is under the separator and operates at suction pressure.

Josip

23-08-2006, 10:15 PM

Hi :)

Do we still need a pump after the high pressure reciever or the pressure will transport the refrigerant back to the separator and bring back the temperature and pressure because of the pressure drop in the pipe?
thanks a lot masters, were really thankfull that you guys are helping us. thanks a lot :)

No, from receiver to surge drum (-10C) you have only a pipe with liquid ammonia under condensing pressure and little subcooled. On this pipe you can have (because there are some other solutions) installed stop valve, solenoid valve, expansion valve and again stop valve just before entering into separator (surge drum -10C). At this expansion valve our high pressure and high temp ammonia expand giving us a mixture of liquid and vaporized ammonia with temperature what we have in our surge drum (-10C), liquid remain in separator and vapour we remove from surge drum with compressor. This liquid with temperature of -10C we can transport (pump) to our cooling units (homogenizers).

hi us iceman,

BTW ipelatak is my groupmate... his message is the one after your latest post pls read his message, tnx... actually our professor told us to pick an industrial plant and design one of its utilities so that it would be easier for us because if he would choose what we're gonna design then it will be very difficult but unfortunately we picked a hard one not the easy ones... we have problems in selecting the compressor because our reference is the suction pressure, is this correct? for example our suction is 42.3psig and discharge press is 190psig... our prof told us to base our compressor to this given data but when im looking for the compressor they are giving the refrigeration capacity not the pressure... is this correct? or convert the TOR to refrigerating capcity and use it in selecting the compressor.. we're really confused... sorry if we'er disturbing you..

Coming to compressor sizing first you have to calculate your heat load=compressor capacity (not exactly, but good for example) for defined evaporating temperature/pressure (-10C, -15C, -35C or -40C) and defined condensing temperature/pressure i.e. one compressor can run at each suction temperature/pressure from -15C to + 5C with the same (or different) condensing temperature/pressure but allways with different capacity.

To obtain needed capacity, compressor capacity is very important but power of motor also.

Hope this will help a little;)

Best regards, Josip :)

Dan

24-08-2006, 12:12 AM

Ahh, crap... Know I see. You guys are designing a Hysave system. That is not a liquid overfeed system, it's a DX system.

Are you sure, Iceman? Seemed to me they are trying to sort out an overfeed system. I was actually puzzled why Richard assumed they were looking at using a pump to increase liquid pressure. The mention of hand expansion valves had me puzzled until you sorted them out as balancing valves. But the original premise is returning liquid and vapor to a separation drum. This is not DX to my thinking.

guapo

24-08-2006, 02:38 AM

Hi,

I based my condenser designed to 45C, you will have a very huge condenser if you based on 35C.
Ambient is already 28-34C.

US Iceman

24-08-2006, 04:43 AM

Hi Dan,

Here is why I thought the discussion took a different turn of events:

Do we still need a pump after the high pressure receiver...

That to me says a Hysave system or something very similar. The only reason to have a pump after the high-pressure receiver is to boost the liquid pressure.

The post that Mr. Bartlett made did not really have anything to do with a liquid overfeed system, other than it uses a refrigerant pump. Different system, different reason.

But the original premise is returning liquid and vapor to a separation drum. This is not DX to my thinking.

I agree Dan. They did state they had a separator, and it certainly sounded like an overfeed system, but when they mentioned the pump after the high-pressure receiver, that is something completely different.

I guess I made some assumptions too. I should have asked what type of condenser they are using. I assumed they were evaporative condensers, since it was an ammonia system.

I just hope the students are beginning to understand this type of system. An overfeed system is not hard to understand, but it sounds like their professor has told them some conflicting information.

I think we can get them to understand this, but I'm not sure if they have enough time before their school project is due.

ipelatak

24-08-2006, 06:05 PM

Hi to all,

Sorry if we make some confusion in what we said in the previous posts.

This is what we understand so far, pls correct us if we're wrong.

New example:
Ambient temp = 34C
Condensing temp = 45C
Evaporating temp = -10C

The liquid refrigerant in the separator is at -10C. The refrigerant will be pumped by a ammonia centrifugal pump. The pump will supply more liquid refrigerant into the evaporator than the amount that boils off.

The pump provides the pressure to transport the liquid refrigerant. The pump will increase the pressure of the refrigerant thus increasing the boiling point so
we need a hand expansion valve to bring back the saturation pressure of the refrigerant.

The liquid refrigerant will pass through the evaporator and leaves at the same temperature of -10C but a mixture of liquid and vapor. The mixture of the refrigerant will go to the Separator to separate the liquid refrigerant and the vapor refrigerant.

The vapor refrigerant at -10C will be sucked by the ammonia compressor. The discharge pressure of the ammonia compressor will be the pressure of ammonia at the ambient temperature in the condenser.

The refrigerant will be condensed at 45C by a water-cooled condenser. The refrigerant still under high pressure leaves the condenser and passes to a high-pressure receiver where it is stored for later use for supply in the separator.

We expansion device after the high-pressure receiver to feed the liquid into the separator to maintain the liquid level. The liquid comes from the reciever at 45C, goes through the expansion valve, into the separator.

what are the valves we need and where to place them? sorry out professor didnt taught us how
to do this.
Uhm.. if you have some free time, can you pls tell us how liquid overfeed works?

In getting the mass flowrate in the system

mass flowrate of refrigerant (latent heat of vaporization ammonia) = Heat added by the
evaporator

Sizing of compressor
Suction pressure = Psat at -10C, ammonia
Discharge pressure = Psat at 34C(ambient temp), ammonia

work of compressor = m(delta h), mass flowrate (enthalpy at 34C - enthalpy at -10C)

Sizing of condenser
Condensing temp = 45C
Heat rejected = work of compressor + heat added by the evaporator

Cooling tower
Air temp in = 32 C
temp of water in reservoir = 34C
temp of water entering condenser = 34C
temp of water leaving condenser = 50C

water flowrate
water flowrate (sp. heat water) (50C - 34C) = Heat rejected by condenser

Sorry if we ask so many, thanks for all your help and thanks for spending time with us. thanks masters. :)

frank

24-08-2006, 07:23 PM

It's not one of my strong points (ammonia) but some of this needs explaining to be able to understand.

The discharge pressure of the ammonia compressor will be the pressure of ammonia at the ambient temperature in the condenser.

This equates to 34C?

The refrigerant will be condensed at 45C by a water-cooled condenser.

The liquid comes from the reciever at 45C, goes through the expansion valve, into the separator.

How does the ammonia vapour condensing at 34C get up to 45C?

US Iceman

24-08-2006, 07:54 PM

New example:
Ambient temp = 34C
Condensing temp = 45C
Evaporating temp = -10C

The liquid refrigerant in the separator is at -10C. The refrigerant will be pumped by a ammonia centrifugal pump. The pump will supply more liquid refrigerant into the evaporator than the amount that boils off.

The pump provides the pressure to transport the liquid refrigerant. The pump will increase the pressure of the refrigerant thus increasing the boiling point so
we need a hand expansion valve to bring back the saturation pressure of the refrigerant.

The liquid refrigerant will pass through the evaporator and leaves at the same temperature of -10C but a mixture of liquid and vapor. The mixture of the refrigerant will go to the Separator to separate the liquid refrigerant and the vapor refrigerant.

OK, so far... Except, the ambient temperature does not matter if you are using water-cooled condensers. I want you to clarify the condenser type. Is it a shell & tube condenser, or a plate heat exchanger?

The discharge pressure of the ammonia compressor will be the pressure of ammonia at the ambient temperature in the condenser.

No. The discharge pressure of the compressor is the condensing pressure (pressure at 45C) plus the discharge line pressure loss. Ambient dry bulb temperature only matters if the condenser is air-cooled.

The refrigerant will be condensed at 45C by a water-cooled condenser. The refrigerant still under high pressure leaves the condenser and passes to a high-pressure receiver where it is stored for later use for supply in the separator.

We expansion device after the high-pressure receiver to feed the liquid into the separator to maintain the liquid level. The liquid comes from the reciever at 45C, goes through the expansion valve, into the separator.

That part is OK.

sorry our professor didnt taught us how to do this.

He should have. If he did not teach this, then what is he doing?:confused:

The mass flow calculations are a little complicated for an overfeed system. It is different in various parts of the system.

In general terms, the liquid supply to the separator provides sufficient liquid to maintain the level. The refrigerant that boils off in the evaporator is this amount.

The refrigerant pump circulates about 300-400% this amount to the evaporator. The 300-400% mass flow is flowing through the liquid line from the pumps, to the evaporator, and back to the separator.

I think you are getting the idea.:)

Josip

24-08-2006, 09:02 PM

Hi, :)

what are the valves we need and where to place them? sorry out professor didnt taught us how
to do this.
Uhm.. if you have some free time, can you pls tell us how liquid overfeed works?

I think the best for you is to visit those links:

http://www.danfoss.com/Asean/Products/Categories/
http://www.danfoss.com/Pacific/BusinessAreas/Refrigeration+and+Air+Conditioning/

here you can find all products you need for refrigeration or AC plant. There you can find also some brochures and so on...;)

Here you can find almost all you need:) valves, systems&vessels, accessories

http://haphillips.com/
http://haphillips.com/products.html?pc=67#77

Pls, don't show that to your professor:D

Hope this will help to finalize your project on time:)

Best regards, Josip :)

ipelatak

25-08-2006, 03:10 AM

OK, so far... Except, the ambient temperature does not matter if you are using water-cooled condensers. I want you to clarify the condenser type. Is it a shell & tube condenser, or a plate heat exchanger?
shell and tube condenser.

No. The discharge pressure of the compressor is the condensing pressure (pressure at 45C) plus the discharge line pressure loss. Ambient dry bulb temperature only matters if the condenser is air-cooled.
thanks

The mass flow calculations are a little complicated for an overfeed system. It is different in various parts of the system.

for mass flowrate, pls correct us if we're wrong
the flow will be the same at the separator to pump, pump to hand ex to evaporator to separator.

In separator (vapor) to compressor to condenser to high-pressure liquid reciever.

High-pressure receiver to ex valve to separator.

In general terms, the liquid supply to the separator provides sufficient liquid to maintain the level. The refrigerant that boils off in the evaporator is this amount.

aahh so this is the flowrate in the high-pres receiver to separator? thanks

Pls, don't show that to your professor:D

Hope this will help to finalize your project on time:)

thanks for the link:)
haha dont worry, we wont:D
Thanks alot masters, ur our savior. thanks for spending time with us and for helping us. can we ask again if we have a problem somewhere again in our project? thanks master us iceman, master josip and to all:)

US Iceman

25-08-2006, 03:56 PM

for mass flowrate, pls correct us if we're wrong
the flow will be the same at the separator to pump, pump to hand ex to evaporator to separator.

Here is how the mass flow rate works.

From the refrigerant pump to the hand expansion valve, to the evaporator, and back to the separator the mass flow is higher. This is where the mass flow will be about 3 or 4 times higher than the other part of the system.

The mass flow from the separator (gas) to the compressor, to the condenser, to the receiver and out to the hand expansion valve on the separator (maintaining the liquid level) is based on the mass flow that boils off in the evaporator, PLUS...

The flash gas that is generated when the 45C liquid flashes off down to 10C, which is being fed into the separator to maintain the liquid level.

So, what you have coming out of the separator as gas (back to the compressor) is the flash gas + the gas boiled off in the evaporator coils.

The excess liquid that is not boiled off in the evaporator, just simply recirculates. Someone may say you need to recirculate more liquid than 3-4 times what boils off.

I say you can design the system for less, so you do not have to pump as much liquid. We will save that discussion for another time.

It sounds like you guys are getting close to being finished. I hope you get a big A++ on your project.:D

Best Regards,
US Iceman

corineramer

26-08-2006, 06:48 AM

The flash gas that is generated when the 45C liquid flashes off down to 10C, which is being fed into the separator to maintain the liquid level.

So, what you have coming out of the separator as gas (back to the compressor) is the flash gas + the gas boiled off in the evaporator coils.

how can we calculate the gas that flashed in the expansion valve??? we're computing it for our compressor capacity...:confused:

we're still confused in getting the compressor capacity because we have computed it using the enthalpy differencence between the discharge and suction times the mass flowrate(we assumed that the refrigerant that will vaporize in the evaporator is 25% of the overfeed massflowrate of the refrigerant that enters the evap. we havnt included the mass of the flash gas) is this correct??

BTW our overfeed is 400%

SteveDixey

26-08-2006, 12:57 PM

I'm stuck without my refrigerant tables but using steam as an example, I think this should help if I remember this right (sounds of scraping around in dusty corners of brain....)

Flash gas is the excess heat energy caused by the reduction in pressure. So for steam condensate:

7 bar = 721kj/kg
0 bar = 419kj/kg
Excess = 302kj

So the percentage of flash gas is:

(excess/enthalpy of evaporation kj @ 0 bar) x 100%

(302kj/2258kj) x 100% = approx 13%, or 13kg of gas and 87kg of liquid.

(Enthalpy of evaporation is the term that is now used for latent heat).

So if you know your mass flow in kg through the regulator, you should be able to work out the amount of flash gas your compressor needs to deal with.

Steve

corineramer

26-08-2006, 01:15 PM

hi to all,

pls correct me if im wrong... what we did is, we calculate the percentage or quality by this equation: h= hf+xhfg
this equation is from thermodynamics.. are we correct???

also how can we measure the mass of the gas that leaves the evaporator?

do the latent heat ammonia vary in different temp???

SteveDixey

26-08-2006, 01:57 PM

what we did is, we calculate the percentage or quality by this equation: h= hf+xhfg
this equation is from thermodynamics.. are we correct???

It is straight thermodynamics so the answer should come from the formula I posted up.

also how can we measure the mass of the gas that leaves the evaporator?

If you know the amount of liquid you are supplying to acheive a refrigerated affect, this volume of liquid will expand to vapour when it hits the evaporator.

In simple terms, you supply say 0.05 kg/second of liquid to the evap and it will expand to produce say 0.05m3/hr of vapour. That vapour density at the evaporating pressure can be found in your refrigerant properties table.

This means your compressor will need to deal with 0.05m3/hr from the evap plus flash gas generated at the hand regulator.

do the latent heat ammonia vary in different temp???

The answer to that should be in your refrigerant tables: http://www.engineeringtoolbox.com/ammonia-d_971.html

Steve

US Iceman

26-08-2006, 03:12 PM

also how can we measure the mass of the gas that leaves the evaporator?

That's easy. The evaporator mass flow for the gas is based on the evaporator cooling capacity divided by, hg-hf (this is also listed as hfg in the refrigerant tables)

Where,

hf = liquid enthalpy at evaporating pressure/temperature

hg = vapor enthalpy at evaporating pressure/temperature

hfg = latent heat at evaporating pressure/temperature

do the latent heat ammonia vary in different temp???

If you look at the tables, you will find your own answer to this.;)

Now, if you calculate the refrigerant mass flow into the separator, you need to use the conditions that exist for the compressor.

hf = liquid enthalpy at condensing pressure/temperature

hg = vapor enthalpy at evaporating pressure/temperature

When you find the mass flow for one kW (or 1 Ton) using the above NRE (Net Refrigerating Effect), hg-hf, you will find this mass flow is higher than the mass flow for the evaporator (for a 1 kW or 1 Ton basis).

The difference between the two mass flows (compressor mass flow minus evaporator mass flow - gas only) is the flash gas.

ipelatak

26-08-2006, 05:10 PM

Hi to all,

Example

evaporator cooling capacity = 178.2637 KW
Temp of ammonia in separator = -10C
Condensing temp of ammonia = 40C

For evaporator mass flow for gas
From ammonia table at -10C
hf = 135.4 KJ/kg
hg = 1433 KJ/kg
hfg = hg - hf = 1433 - 135.4 = 1297.6 KJ/kg
Mass evap gas = (178.2637 KJ/sec)/1297.6 KJ/kg
= 0.13738 kg/sec

For compressor mass flow for gas
From ammonia table at 40C (condensing temp)
hf = 371.9 KJ/kg
From ammonia table at -10C (evaporating temp)
hg = 1433 KJ/kg
hfg = hg -hf = 1433 - 371.9 = 1061.1 KJ/kg
Mass comp gas = (178.2637 KJ/sec)/1061.1 KJ/kg
= 0.16800 kg/sec

Mass comp gas = Mass evap gas + Mass flash gas
Mass flash gas = Mass comp gas - Mass evap gas
= 0.16800 - 0.13738
= 0.03062 kg/sec

Pls correct us if we're wrong
In getting the mass flowrate with overfeed of 400%
= mass evap gas x 4
= 0.13738 kg/sec x 4 = 0.54952 kg/sec
In getting the volume flowrate with overfeed
from ammonia table at -10C
vf = 1.5338 L/kg
Vol. flowrate = 0.54952 kg/sec x 1.5338 L/kg
= 0.84285 L/sec

In getting the work of compressor, Wc
Suction pressure, Ps = Psat at -10C
From ammonia table at -10C
Ps = 2.908 Bar or 290.8 KPa
hg = 1433 KJ/kg
Discharde pressure, Pd = Psat at 40C
From ammonia table at 40C
Pd = 15.54 Bar or 1554 KPa
hg = 1473.3 KJ/kg
Wc = Mass comp gas x (hg at 40C - hg at -10C)
= 0.16800 kg/sec (1473.3 KJ/kg - 1433 KJ/kg)
= 6.7704 KW
pls correct us if we're wrong.. thanks masters :)

corineramer

29-08-2006, 12:14 PM

SteveDixey

29-08-2006, 01:31 PM

hi to all,

i have a question in our ageing process... we have to maintain the temp of our product to 4C... we're confused in getting the load because there's no temp difference... but we're to maintain this temp... how can we get the load in order for us to select the equipment.. we will use a water chiller in maintaining this temp... pls help, tnx

Load comes from heat. Your product will heat up by respiration (biological activity) and heat will leak into the cooled space (heat transmission). Your refrigeration should carry away this heat. Heat will also come from any lights, fans, or people in the cooled space. Even if you load your product in at 4c, it will warm up eventually.

You have to work out this heat load and this information is readily available in text books and on the internet.

In simple terms, you have to work out the equipment needed to carry away this heat and from this you get your equipment requirements. Look for water chiller suppliers and download their information. A water chiller will be rated in kW and you need to supply it with enough refrigerant and enough water to move the heat.

Steve

US Iceman

29-08-2006, 02:33 PM

we're confused in getting the load because there's no temp difference...

Steve mentioned where the heat comes from for your load. The remaining issue to determine is the desired temperature range for the water chiller flow and the evaporating temperature of the chiller.

If you use a high temperature range (larger TD for the water), the flow rate will be smaller (smaller water pump).

A lower temperature range (smaller TD for the water), the flow rate will be much higher (larger water pump).

Depending on what you select for the supply chilled water temperature to the coil(s) and the return water temperature (from the coils), you are finding the TD required. This is a balance between the requirements of the cooling process and the costs of the equipment (operating and initial equipment costs)

When you have the temperature range and flow rate established, then you can select the evaporating temperature for the chiller.

Once this is completed, you have the load and the evaporating temperature, so now you can add this to the refrigeration capacity requirements and select the chiller.

corineramer

30-08-2006, 11:11 AM

hi us iceman,

im selecting the the compressor using the grasso site that you gave... what does refrigerating capacity at 2940 min^-1 mean??? min raised to -1, what does it stand for?

SteveDixey

30-08-2006, 11:29 AM

hi us iceman,

im selecting the the compressor using the grasso site that you gave... what does refrigerating capacity at 2940 min^-1 mean??? min raised to -1, what does it stand for?

Means revs per minute (rpm) as in say 160m3/hr of refrigerant @ 2940 rpm

Steve

NoNickName

30-08-2006, 11:30 AM

hi us iceman,

im selecting the the compressor using the grasso site that you gave... what does refrigerating capacity at 2940 min^-1 mean??? min raised to -1, what does it stand for?

min^-1 means 1/min
so 2940 min^-1 is 2940 / min or 2940rpm

US Iceman

30-08-2006, 02:29 PM

what does refrigerating capacity at 2940 min^-1 mean???

Others have addressed the engineering notation for the 2940 min^-1, so I believe we can move past this point.

However, I did want to clarify the statement itself.

When the Grasso literature says capacity at 2940 RPM, are they stating the kW (refrigeration) at 2940 RPM? Or, does the literature list cubic meters per hour?

Capacity is kW (or Tons), not m^3/hr. You may have already seen this, but I wanted to point out the fact that you must read the small notes sometimes hidden in the technical information.

The other issue that sometimes gets overlooked is the use of subcooling. Some manufacturers will use subcooling on ammonia compressors to inflate the performance values quoted.

Very seldom will you find subcooling from the condenser in an ammonia system, IF THE CONDENSER IS PIPED CORRECTLY.

Capacity is capacity, but you have to be careful on how the manufacturer states you will get that capacity at a specific operating condition.

NoNickName

30-08-2006, 02:56 PM

Capacity is kW (or Tons), not m^3/hr. You may have already seen this

Generally, capacity is directly proportional to swept displacement and (that is the same) to rotational speed.
2940rpm is the synchro speed for 50Hz asynchronous three-phase 2-poles motors (50/s x 60 s/min = 3000/min), but at 60Hz it is more likely to be 3520 rpm, with a net increase in capacity of 20%.
Which is why capacity is specified together with the swept volume or rotational speed.

US Iceman

31-08-2006, 12:51 AM

Generally, capacity is directly proportional to swept displacement and (that is the same) to rotational speed.

Not always... If the screw compressor speed is slowed down below about 50% of the full speed, the capacity drops off very quickly.

In general terms though, I agree with what you said. But in no way did I intend to make a point that the capacity does not change with speed.

corineramer, here is a little equation to help you with this...

(Hz X 120)/ number of motor poles = nominal RPM

Here is an example:

(50 Hz X 120)/4 pole motor = 1500 RPM

(60 Hz X 120)/4 pole motor = 1800 RPM

(50 Hz X 120)/2 pole motor = 3000 RPM

(60 Hz X 120)/2 pole motor = 3600 RPM

The actual full load RPM will be slightly less than this due to slippage of the rotor in the stator. I thought this might help you understand the relationship with motor speed, Hertz, and motor construciton.

corineramer

31-08-2006, 06:18 AM

hi,

thanks for that info... us iceman where can i get a clearer drawing of the evapco horizontal recirculator system?? we need it for our lay out, we're gonna use cad in drawing it... right now im selecting the equipments and following the procedure in the brochures but i dont know if im doin it right.. hehehe.. tnx again for the big help

US Iceman

31-08-2006, 03:03 PM

Hi corineramer,

Try this as an example to start with. This example is for an open drive refrigerant pump, not a hermetic pump.

corineramer

31-08-2006, 03:43 PM

hi us iceman,

thank you so much sir... u have contributed a lot to our work... just a little push and where nearly finished..

US Iceman

31-08-2006, 10:36 PM

I'm glad to help you guys.

Push, push, push.... Did that help you get the project done?:D

corineramer

01-09-2006, 05:30 PM

hi us iceman,

yup!! nearly, we dont know bout the piping because there are a lot of things that we dont know yet, like the inclination of hte pipes the oil traps that must be "u" shape.. i saw it in the brochure.. we're not famaliar with that but will solve it, hehehe:D also do u have the price of surge drum, cooling towet and evaporative condenser of the EVAPCO??

US Iceman

01-09-2006, 09:53 PM

we dont know bout the piping because there are a lot of things that we dont know yet, like the inclination of hte pipes the oil traps that must be "u" shape...

What oil traps are you talking about?

also do u have the price of surge drum, cooling tower and evaporative condenser of the EVAPCO??

Sorry, that is one area I can't help you with.

But, it is nice to hear you are making progress. Keep up the good work.:D

corineramer

03-09-2006, 09:59 AM

hi us iceman,

yup, we also managed to get a hold of an ASHRAE copy for the sizing of the pipes, hehehe... we'll have to read it...

do u know where can we find a tunnel freezer for ice cream and filling machine?? i didnt like the one that we got because im not sure about the sanitary and its second hand.. its for the quality.. as for the filling machine we need to fill them in our container... our ice cream does not come on sticks and wrapper.. its in bulk cans..

thank you so much! rock on!!!\m/

US Iceman

03-09-2006, 09:30 PM

do u know where can we find a tunnel freezer for ice cream and filling machine??

Sorry, I won't be much help there either. Josip or Andy may have some ideas, but since I have a very limited exposure to ice cream making I just don't know who the manufacturers are.

corineramer

04-09-2006, 02:08 PM

hi us iceman,

ok tnx.. i'll just pm them...

corineramer

07-09-2006, 02:20 PM

hi us iceman,

do u know something that can help us in piping our ammonia system??

US Iceman

07-09-2006, 02:47 PM

What is your question? I will try to answer them...

corineramer

09-09-2006, 12:26 PM

hi us iceman,

for the suction diameter of the compressor and outlet diameter of the low pressure receiver are different... we dont know how to size them? our equipments have different diameters.. how can we pipe them? by using the larger diameter or the smaller or you can calculate for it??

US Iceman

09-09-2006, 04:40 PM

Always calculate the pipe size you need. The connections put on the equipment is the manufacturers attempt at balancing cost versus requirements.

When you are designing the system you have to know the operating conditions for each piece of equipment.

Let's look at an evaporator...

If the evaporator is selected for a specific evaporating temperature, then you will have some pressure loss (due to flow) to the next piece of equipment. The difference in pressure between these two components is determined by the mass flow and pipe size.

If the pipe is too small, you will have a higher pressure loss.

If the pipe is larger (for the same mass flow), the pressure loss will be reduced.

The pressure at the end of this pipe will affect the performance of that device.

I never recommend using the connection size as the required pipe size.

You do have to connect the pipe to the two components, so what you want to find is the required pipe size based on allowable pressure loss (this is what you use for your system design and equipment selections).

After you know the required pipe size, you will probably find that you will need some pipe reducer fittings to make the terminating connections.