PDA

View Full Version : R22 Flooded System - Trawler







chiefbecker
28-01-2008, 12:24 AM
Hello,

I am a Chief Engineer working on a 270' factory trawler in the Bering Sea, Alaska. I found this sight while searching for more information on superheat and the adjustment or monitoring of it. There is a lot of great information here. Makes a guy really think. We use a flooded R22 system to freeze around 130 metric tons of fish per day. If anyone has a good way of setting superheat on a flooded horizontal plate freezer, please let me know. I am running liquid into the freezer through a Danfoss REG 25 valve @45 psi and 4mm orifices into the liquid lines for each plate, suction pressure is the equivalent of -35^f and the suction temp is -32^f. We are getting good core temps on the blocks of around -26^f after 104 minutes. I would like to know if I am close to max efficiency for this type of setup.

Thanks for your help with this.......

Plank!
28-01-2008, 01:13 AM
Surely if its a flooded plate there would not be any superheat.

How is the suction set up after the plate? Is there a surge drum or reciever of some sort?
Is the liquid pump circulated?

If so you'll return wet - no superheat - and pull dry vapour from there to the compressor.
Wet return temperature will be the same as suction pressure near enough, dry return to the comp a few degrees higher.
So from the readings you've mentioned it all sounds about right.

taz24
28-01-2008, 01:16 AM
Hello,

I am a Chief Engineer working on a 270' factory trawler in the Bering Sea, Alaska. I found this sight while searching for more information on superheat and the adjustment or monitoring of it. There is a lot of great information here. Makes a guy really think. We use a flooded R22 system to freeze around 130 metric tons of fish per day. If anyone has a good way of setting superheat on a flooded horizontal plate freezer, please let me know. I am running liquid into the freezer through a Danfoss REG 25 valve @45 psi and 4mm orifices into the liquid lines for each plate, suction pressure is the equivalent of -35^f and the suction temp is -32^f. We are getting good core temps on the blocks of around -26^f after 104 minutes. I would like to know if I am close to max efficiency for this type of setup.

Thanks for your help with this.......

Wecome if you work in conditions that I've seen when I've watched the deadliest catch on TV, then my hat off to you sir.

Flooded evaps with direct expansion will have to be set just like any other direct expansion evap.
Measure the saturated pressure / temp of the liquid in the evap and the compair it to the actual measured temp of the suction line.

The most important thing I can think of for your application is not letting liquid back to the comp?

Do you have suction line accumilators before the comp or does it have a suction surge drum of any description?

Cheers taz.

NH3LVR
28-01-2008, 03:39 AM
Chiefbecker;
Welcome to the forum!
Always glad to see another fish-freezing guy come along.
I am sure you have a pump system here, probably with centrifugal pumps.
Setting the overfeed ratio is a bit tricky, but not too critical. If your suction line is 3 degees warmer than the suction temp you are awful close to being on the money.
Setting the feed by superheat is not the way to go.
US Iceman has a lot of knowledge regarding overfeed rates, much more that I. The problem comes when the feed ratio is too high, causing a condition called brining.
Effectively you are injecting subcooled liquid into a evap. If the feed is too high under some circumstances the subcooled liquid will fill the evap for some distance before starting to boil away.
But it sounds like you are in good shape to me.
I might try changing feed rates on one freezer a bit to see if it makes any difference. I think you will find you about correct.

chiefbecker
28-01-2008, 08:05 PM
I woke up this morning, its blowing about 65-70 mph the weather is generally crappy, but I recieved a
number of great responces to my first posting. The whole process is really cool. We use a direct satellite link from the boat to a server in Norway and I get responces from people on opposite sides of the earth.

The system that we have here has three screw compressors pulling on an LPR that holds about 4 tons of liquid. I run this at about 35% capacity. All of the suction lines from the plate freezers return to a 10" header that ties into this LPR. We pump the liquid at -40^f back up to the plates. I would normally just adjust the liquid feed rate until we had a good frost covering on the plates after about 25 minutes. If the feed rate is too high, we wouldn't get good core temps because of flooding in the plates. I just modified all of the piping and replaced all of the valveing in the plate freezer system during our last dry dock period. We installed thermometers and pressure gauges on the suction and liquid headers this time. Its the first time that I have actually had that info in one of these systems. The contractor that we used to supply the valves kept talking about setting the feed rate to a certain superheat but didn't really have an answer to what he felt the temp/pressure should be. I am trying to get the optimum feeding rate of the liquid so that we can reduce the freeze times. Shorter freeze times = more daily production = I get to home sooner. :)

Thanks for your help on this.......

US Iceman
29-01-2008, 02:39 AM
Hi Chief,

I'm glad to see that you mentioned the pressure and temperature connections on the evaporators. It would be helpful to know what the pressure difference is for the orifices (that the plate manufacturer designed their flow rate and inlet pressure for).

You can work backwards from the evaporating pressure. So, if you have a -40F evaporating temperature on R-22 (which is 0.5 psig), let's say the manufacturer designed the orifices to provide the required flow rate with a 20 psi pressure difference).

If we add the evaporating pressure (0.5 psig) + the design orifice pressure differential (20 psig for example), then the liquid pressure coming into the evaporator (after the liquid regulating valve) should be about 21 psig. What I'm recommending is to work backwards from know values to achieve the desired liquid feed pressure.

If the manufacturer has their design right, then by adjusting the liquid feed valve properly you should be pretty close the right overfeed rate.

I hope that helps. I started to write this post about fours ago, and was just able to get back to it.

If you have any questions, let us know.

chiefbecker
29-01-2008, 10:29 PM
Hi Iceman,

Thanks for your help with this. The info I have for the plate freezers really doesn't have any design or operation info in it. I have used a number of different freezers from various manufacturers and found that the installation will determine the orifice size in most cases. We played around with different orifice sizes in the freezers over the years even to the point of running smaller ones in the lower plates and larger in the upper. I found that around 4 mm works the best if you can keep the liquid feed pressure up around 45 psi and then regulate it through the REG valve on the liquid header and the suction pressure as close to 0 as possible. Any smaller than that and your hot gas time goes south. I am going to install a couple of pressure gauges in the liquid headers and see if I can regulate the pressure like you recommended. I will let you know how it turns out....

Sergei
31-01-2008, 05:19 PM
The problem with production plate freezers is that refrigeration load isn't constant. At the beginning of freezing cycle the load is maximum and minimum at the end of the cycle. I think that overfeed should be 1:1 and the beginning and gradually this rate will increase itself to 1:5 or 1:10 due to lowering of refrigeration load. Probably, the best way of optimization is to change liquid supply pressure and monitor the result of this change. Suction pressure should be constant. Initial temperature of the fish is important.

Tycho
31-01-2008, 09:18 PM
Hi Iceman,

Thanks for your help with this. The info I have for the plate freezers really doesn't have any design or operation info in it. I have used a number of different freezers from various manufacturers and found that the installation will determine the orifice size in most cases. We played around with different orifice sizes in the freezers over the years even to the point of running smaller ones in the lower plates and larger in the upper. I found that around 4 mm works the best if you can keep the liquid feed pressure up around 45 psi and then regulate it through the REG valve on the liquid header and the suction pressure as close to 0 as possible. Any smaller than that and your hot gas time goes south. I am going to install a couple of pressure gauges in the liquid headers and see if I can regulate the pressure like you recommended. I will let you know how it turns out....


What make of freezers is it?

amd that ship profile looks awfully familiar, you wouldnt happen to have a name for it?

chiefbecker
02-02-2008, 03:40 AM
We are using both DSI and Jackstone freezers. They are roughly the same size and the infeed orifices are 6mm @45 psi liquid pressure. The major problem is to balance the ratio of liquid entering the freezer. If you have a really good start, they flood to quickly as the refer load drops. I added a timer in the control side to shut down the liquid about 10 minutes before the end of the freeze cycle to draw off some of the extra liquid to improve the hot gas times. I will try reducing the liquid pressure and see if things improve.

The boat in the picture is the American Dynesty....

US Iceman
02-02-2008, 06:09 AM
I believe this is the sort of equipment Tycho works on all the time (when he is not chasing girls with a beer in his hand:D).

He will have some good input to this discussion.

I am not a big fan of orifices. They are designed to only work at one operating condition which is usually full capacity.

Look at it like this... If a 4mm orifice will pass "X" pounds (kgs) of refrigerant per minute at a specific pressure difference across the orifice, you can change the flow rate through the orifice by changing the inlet pressure. The outlet pressure of the orifice is the evaporating pressure so obviously it must stay relatively constant to produce the desired cold temperature.

And, they are good at passing liquid but during hot gas defrost they limit the amount of gas you can put into the evaporator (which really slows down the defrosting process).

I hope Tycho can post a flow diagram of the liquid feed and defrost piping. That would really help me to see what you are fighting Chief. I'll be the first to admit I have not worked on one of these systems, but I have worked on a lot of liquid overfeed systems and evaporator designs.

Sergei
02-02-2008, 05:20 PM
I believe this is the sort of equipment Tycho works on all the time (when he is not chasing girls with a beer in his hand:D).

He will have some good input to this discussion.

I am not a big fan of orifices. They are designed to only work at one operating condition which is usually full capacity.

Look at it like this... If a 4mm orifice will pass "X" pounds (kgs) of refrigerant per minute at a specific pressure difference across the orifice, you can change the flow rate through the orifice by changing the inlet pressure. The outlet pressure of the orifice is the evaporating pressure so obviously it must stay relatively constant to produce the desired cold temperature.

And, they are good at passing liquid but during hot gas defrost they limit the amount of gas you can put into the evaporator (which really slows down the defrosting process).

I hope Tycho can post a flow diagram of the liquid feed and defrost piping. That would really help me to see what you are fighting Chief. I'll be the first to admit I have not worked on one of these systems, but I have worked on a lot of liquid overfeed systems and evaporator designs.
Hi, Mike.
Certainly, these orifices aren't perfect and they are major restrictions for hot gas flow, but they are good restrictions.
1. Orifices create higher pressure in the coil(during defrosting) and head pressure can be lowered. When BPR sets to 70 psig, pressure in coil would 90 psig or higher.
2. Orifices restrict parasitic load from blow-by gas. Without orifices this load can be huge especially at higher condensing pressure.
I know that you prefer fast defrosting, I don't. I like gradual defrosting for two reasons:
1. Fast defrosting creates significant parasitic load from blow-by gas(even with orifices). This load initiate unnecessary starts and stops for compressors.
2. Safety. Properly designed plant has liquid drained for has gas line. In real life, this drainer can be plugged by scale, dirt and etc. Liquid will collect in hot gas line and fast defrosting can be disaster.
I think that 20-30 min defrosting is optimum for evaporator coils(not for plate freezers).
About mentioned plant. I would install hand expansion valves for each freezer. Liquid pump head should be 45 psig, but optimum pressures for different freezers can be different and can be adjusted by hand expansion valves.

US Iceman
02-02-2008, 07:26 PM
Hi Sergei,



About mentioned plant. I would install hand expansion valves for each freezer. Liquid pump head should be 45 psig, but optimum pressures for different freezers can be different and can be adjusted by hand expansion valves.


This would provide a better means of control and finer adjustment. No argument on that point.



2. Safety. Properly designed plant has liquid drained for has gas line. In real life, this drainer can be plugged by scale, dirt and etc. Liquid will collect in hot gas line and fast defrosting can be disaster.


Something can always happen. There are no 100% guarantees with anything. Hot gas defrost lines will almost always contain some condensed liquid, so whether a drainer is used or not, the problem remains. I simply prefer using drainers to provide some protection from vapor propelled liquid slugs.



1. Fast defrosting creates significant parasitic load from blow-by gas(even with orifices). This load initiate unnecessary starts and stops for compressors.


That's very true, if, you do not control the condensate. A defrost relief valve simply works on pressure, so when the valve opens it simply blows liquid and vapor down the pipe. A drainer on the other hand works on liquid accumulation only. But as I'm sure you know, when the drainer does open to allow liquid to flow past some flash gas also forms.

There are no ways to prevent this entirely, but the drainers come as close as anything I am aware of limiting the gas loading problem for the compressors during defrost.



2. Orifices restrict parasitic load from blow-by gas. Without orifices this load can be huge especially at higher condensing pressure.


Again, I agree with your comments. Orifices do cause the defrost process to slow down as the orifices can only pass so much refrigerant (condensate), which cause the orifices to choke.



I know that you prefer fast defrosting...


Yes, but I am also an advocate for controlling the defrost cycle in a better fashion than we currently do. Evaporators can be made to operate more effectively I feel. Orifices have been adopted as the means to regulate liquid into evaporators, and then all sorts of devices and methods have been implemented to offset the problems created by the use of the orifices.

Tycho once posted a video link that showed a quick defrost on a ship board system. It was very apparent from watching this that it is possible, but...the defrost still has to be controlled to prevent other reactions in the system.

I am not disagreeing with anything you have said Sergei. You are exactly correct on all points. In my little universe though, I think there are some ways we can do a much better job of controlling this process.

Very good discussion points all the way around though.:cool:

Sergei
03-02-2008, 05:04 PM
Every action should be reasonable. We have to evaluate pluses and minuses of this action.
Regarding to Tycho's plant. I don't see reason to test this plant for thermal and pressure shocks. Time for building up the defrost pressure should be increased at least to 1 min.

Tycho
04-02-2008, 06:04 PM
American Dynasty aye... Does it still have the original Kvaerner Kulde plant?

I don't know too much about the horizontal freezers from DSI and Jackstone.

but freezing in blocks in a horizontal freezer down to -32C in 104 minutes seems pretty impressive, normally I would assume that time for maybe -20 to -23 C, which is the temperature the blocks would be pulled at here in Norway.

Are you freezing cartons, fish packed in plastic or bare fish?

Most trawlers only defrost the freezers when the icebuildup is making a good contact between the plate and the block no longer possible. so I'm guessing at a defrost meybe every three to four cycles of freezing.



1. Fast defrosting creates significant parasitic load from blow-by gas(even with orifices). This load initiate unnecessary starts and stops for compressors.

I don't think this is a big problem in horizontal freezers, they are all fitted with risers inside the liquidline collector, thus no hotgas will get to the riser until the lower plate is defrosted.

I'dont know the dimensions of this make of freezers, but I know on the KBH 12J from kvaerner, we use 5 mm nozzles with 2-3 bar pump pressure (40-45 psi?).


From what I know, if you freeze cartons of fish, I think you can pull the blocks out of the freezer already at -23 - -24C and save some freezing time at that, ofcourse, you should check that the temperature rise as it makes it's final way through the factory doesnt go above -18C
(this is if you only defrost the freezer every few cycle, and defrost the freezer when it's empty to ensure the block temp doesnt go up.)

if you freeze unpacked fillet, you should not go below -30C in any case, as this dries out the fish, this also goes for the cargohold, if you have fish that is open to the hold (Not packed in plastic or in sealed bags) the fish will dry out.

That's all I have for now :)

chiefbecker
06-02-2008, 03:07 AM
Sorry, typo on the block temps. They are coming out at -23C not -32C. That would be pretty impressive.

We are freezing in both carboard pan liners for fillet block and plastic bag liners for surimi. We defrost on every cycle. One product we put up is mince and it really gunks up the plate when you squeeze the product. The other problem is keeping the blocks in the open freezer during rough weather when they have any ice on them. The last boat I was on we designed and built a flash economizer that sub cools the liquid from the HP to the LP and also installed a seperate hot gas return line from the plates to this economizer. This vessel holds about two tons of liquid when completely full. We run this at about 2.2 bar so that we have good flow between it and the LP receiver. We use the side port suctions of the Mycom screw compressors to keep the pressure down in that vessel and regulate it with a BPR. It really works well. We have a Mycom recip compressor that we use for an RFW chiller that we regulate the high side pressure to 12 bar for hot gassing the plates. This allows us to drop the high side pressure of the main system as low as we can get it. Usually around 7.5 bar depending on seawater temp. We use a Danfoss ICS valve set at 4.5 bar on the hot gas return lines at the plates to regulate back pressure in the plate during the hot gas cycle. We also change the hotgas infeed to the bottom of the liquid header and the hot gas return to the bottom of the suction header. This really helped to reduce the time it took to defrost the lower plates. By returning the hot gas back to the economizer we eliminated the jump in pressure in the LP everytime they hot gassed a freezer. It also reduced hot gas time to about 9 minutes per cycle.

US Iceman
07-02-2008, 05:39 PM
By returning the hot gas back to the economizer we eliminated the jump in pressure in the LP everytime they hot gassed a freezer.


But what does that do to the economizer pressure?

If this port pressure goes high enough you can raise the motor amps on the compressor and force the compressor to unload. That should be the typical response, unless your system has some different control configuration.

Tycho
07-02-2008, 10:44 PM
Sorry, typo on the block temps. They are coming out at -23C not -32C. That would be pretty impressive.

I scratched my head a little but took your word for it :)

A rule of thumb is that it should freeze about an inch per hour (down to -20 - -23)



We are freezing in both carboard pan liners for fillet block and plastic bag liners for surimi. We defrost on every cycle. One product we put up is mince and it really gunks up the plate when you squeeze the product. The other problem is keeping the blocks in the open freezer during rough weather when they have any ice on them.

I can understand the mince, I've seen it dripping of many a freezer if the crew have been sloppy with the cleaning.

Most freezers have the option to attach "roll bars" on every plate to prevent the blocks from sliding out of the freezer. when filling it up, you fold out the roll bar in the front and push the blocks in till they hit the rear bar, then fold up the front bar and they are locked in place (maybe 1 cm clearing on each side)

if there are spare holes on the side of the plates close to the front and rear, chances are you can add these.


We have a Mycom recip compressor that we use for an RFW chiller that we regulate the high side pressure to 12 bar for hot gassing the plates. This allows us to drop the high side pressure of the main system as low as we can get it. Usually around 7.5 bar depending on seawater temp.


I'm assuming this compressor has a pm valve on the highpressure and that it connects to the same condenser as the mycom screws, and that each screw compressor have a PM valve too.

So am I correct in assuming that the hotgas is connected between the piston compressor high side and the pm valve for that compressor?

With all those assumptions over with, I have to say that it would be better to fit an extra PM valve right before the condenser, and fit the hotgas on the main line between the compressors pm valves and the condencer inlet pm valve, and you could still lower the pressure a little bit, but the amount of hotgas available would go up conciderably.





We use a Danfoss ICS valve set at 4.5 bar on the hot gas return lines at the plates to regulate back pressure in the plate during the hot gas cycle. We also change the hotgas infeed to the bottom of the liquid header and the hot gas return to the bottom of the suction header. This really helped to reduce the time it took to defrost the lower plates.

Now the ICS is a good solution, expensive, but good, you get a bigger amount of liquid through the valve than you would with a OFV valve.

didn't the freezers have a riser in the liquid collector?

I have tried to get something similar implemented in the company I work for, but we have some blockheads with the argument of "what if some of the nozzles are plugged with dirt, how are you going to get it out if you can't reverse the direction?"

I can see where they are coming from, but in a new system, with startup filters and frequent filterchanges you shouldnt have this kind of problems, in an old plant ok. and if we make a plant like this and they have a few problems after 10 years of continuous operations, it is to be expected :)

Since they won't let me do it on our plants, I'll share it with you (I've seen it on two shipboard installations here in Norway, a looooong looong time ago)


The solution is EVRA solenoid valves, PM valves and a timer.


Liquid enters the freezer through the REG valve and the NRV valve (in that order), goes through the plates and exits through the suction valve.

the operator closes the suction valve and turns the switch for "defrost"

1. A EVRA valve bypassing the NRV and REG opens

2. A EVRA valve connected to the suction side opens for hotgas.

3. A timer set to X amount of time keeps the EVRA valve bypassing the liquid valves open until the bottom plate is nearly defrosted
(this keeps from flushing hotgas unrestricted into the main return line)

4. the EVRA bypass valve closes and the pressure inside the freezer builds up untill the OFV valve opens at 4.5/5 bar

5. Another timer shuts off the hotgas inlet valve

6. defrost finished.


The ideal would be to mirror this and have the hotgas inlet on the liquid side and the bypass and ofv on the suction side.

This could cut a normal defrost time of 4-6 minutes on a VERTICAL plate freezer by maybe 50%, and combined with robots for emptying the freezers this could save alot of time and increase production.

The robots are allready a fact, check it out at http://www.youtube.com/watch?v=sYMdxuiDZMw

They also have robots for filling and emptying horizontal platefreezers, with those, you HAVE to defrost every cycle unless you want to go chasing catering boards all over the factory :)

Tycho
07-02-2008, 11:11 PM
But what does that do to the economizer pressure?

If this port pressure goes high enough you can raise the motor amps on the compressor and force the compressor to unload. That should be the typical response, unless your system has some different control configuration.

I skipped this in my previous post, as with Iceman's reply I wanted to keep this separated from the other issues.

I'm gonna ask another question based on this statement:



By returning the hot gas back to the economizer we eliminated the jump in pressure in the LP everytime they hot gassed a freezer.

hmm, I did a little re reading before finally formulating this reply, and I noticed that in your post you say say "the last boat I was on", so now I'm a bit confused with my post above this, if I based it on the wrong facts or not, but I'll let it stay like it is.

Now, had this been a basic ovefeed system with OFV valves, you should see little to nothing rise in the LP when you defrosted a freezer, as the OFV would not only make the pressure in the freezer go up, but it would act as an expantion valve, causing only a tiny tiny rise in LP.

But with you having ICS valves acting as OFV valves, I think that is what is causing the rise in LP, since when they have their pressure difference, they are open, and first you flush through a whole lot of liquid and if the fisherman operating the system does not pay attention (Wich I know they are not), once the freezer is empty you have a short circuited system.

you didnt say what size ICS you have, but I'm guessing they are bigger than the normal DN 25 OFV's :)

If there is one thing I've learned, it is that there is fail safe systems, but to date there has not been invented a system that is fisherman safe :) (by fishermen, I mean the people working in the factory that operates factory equipment, including freezing equipment), seriously, To date there is not a single piece of machinery invented on this planet that a fisherman could not break if given the oportunity.



I don't mean to critisize in my above posts, I like the way this thread is heading, so let's keep it going :)