PDA

View Full Version : Liquid feed



Volnei
25-02-2007, 02:57 PM
Hi Guys,

May everybody give more explanation why higher circulating rates are recommended when using liquid top feed for evaporators in an ammonia overfeed system?

ASHRAE Refrigeration Handbook:
Top feed: 6 to 7
Down feed: 2 to 4

Regards,

Volnei

US Iceman
25-02-2007, 11:40 PM
My belief is this is based on the need to provide higher mass flows to assure the coil is uniformly wetted with liquid refrigerant.

I also think (but have not had any expert mention this) that this is an old wives tale based on the use of orifices in the liquid header. Similar to how a bottom-feed coil is circuited, except the other way for top-feed.

I also think this header design for top -feed coil circuiting is why the manufacturers say you should not use hot-gas defrost on a top-feed coil (again no confirmation by a manufacturer, but something I believe).

If you use the top-feed coil circuiting with a refrigerant distributor, all of the above problems disappear.

NH3LVR
26-02-2007, 02:47 PM
I just found this link.
I have only scanned it briefly, but it looked interesting.

http://nhtres.com/technote01.pdf

US Iceman
26-02-2007, 05:25 PM
Hi NH3LVR,

That is an article based on one of my presentations for liquid overfeed systems from several years ago. Since this happened I try to be more careful with providing too many details.

As you see what happens..:rolleyes:

Dan
27-02-2007, 01:02 AM
Iceman, can you elucidate on figure 4 of that article? I was enjoying it up until figure 4, where we have a small amount of liquid and a large amount of vapor in a tube with subcooled liquid.

NH3LVR
27-02-2007, 02:07 AM
Hi NH3LVR,
That is an article based on one of my presentations for liquid overfeed systems from several years ago. As you see what happens..:rolleyes:
US Iceman;
As I said I thought it was a good article, although I had only scanned it briefly.
Should have suspected it came from you, being of such high quality.:)

US Iceman
27-02-2007, 02:59 AM
Hi Dan,



Iceman, can you elucidate on figure 4 of that article? I was enjoying it up until figure 4, where we have a small amount of liquid and a large amount of vapor in a tube with subcooled liquid.


That part of the article is discussing the effects of top-feeding a coil and what happens to the subcooled liquid. The liquid essentially just lays there.

On a top-feed coil using orifices in the liquid header you can have some liquid distribution problems caused by gravity. In other words, the gravity causes the liquid to separate and puddle on the bottom of the tubes. When this happens you have very little heat transfer surface exposed to liquid (let alone any boiling liquid, and completely discounting the subcooled liquid refrigerant).

This is similar to what is described in the Sporlan literature for when a refrigeran distributor is not used. As this is a two-fold problem; subcooled liquid in the evaporator and poor distribution.

Coil circuiting is an interesting topic and full of old wives tales and black magic.:rolleyes:

US Iceman
27-02-2007, 03:10 AM
Hi NH3LVR,

Thanks for the kind comments. Several years ago I spent quite a bit of time researching liquid overfeed systems for problems. I kept seeing the same thing happen over and over, so I thought... Why not spend some time on trying to figure these issues out?

Overfeed systems are relatively simple, but can be complex with all the things "we" try to do to them. As you can see from the article, a lot of interesting things start to happen once you install back-pressure regulators.

They affect the pumps and the coils.

Pump cavitation is another interesting topic. I don't know why it has to be so difficult to solve, but for some reason it seems to be.:confused:

This is one of those reasons why "industrial refrigeration is more art, than science". There is too much guess work done during the system design.:o

Dan
27-02-2007, 04:38 AM
That part of the article is discussing the effects of top-feeding a coil and what happens to the subcooled liquid. The liquid essentially just lays there.

Thanks, Iceman. But we are revisiting essentials that I have never quite gotten a grip on. If you have vapor and liquid, the liquid and vapor by definition, are saturated. The example of a pan of water is intriguing, but where is the flaw in the definition I speak of?

How is it possible to have a subcooled liquid with vapor existing? Especially in the amounts that the vapor volume exceeds the liquid volume as figure 4 indicates?

How can you have a fluid below it's saturated condition and a vapor of the fluid at the same time and place?

:)

NH3LVR
27-02-2007, 05:10 AM
Since we have gotten onto this topic I would like to see some commentary on Top Vs Bottom feed recirculated coils.
My experience with top feed has not been good.
I understand that they may be better for defrosting in some ways, however I had three top feed evaporators drive me nearly insane.
We had a pump cavitation problem develop (which was never fully explained). Whenever the pumps would cavitate the liquid from all three would return to the LPR, overfilling it and shutting down the compressors when the high level safety tripped.
We do not install many top feed coils and I have never figured any advantage to them.:confused:
Comments?

US Iceman
27-02-2007, 03:32 PM
How can you have a fluid below it's saturated condition and a vapor of the fluid at the same time and place?


This happens all the time in liquid receivers. The interface of the liquid surface is saturated. As you go deeper into the liquid (down), the static pressure increases. This is also subcooling (but we have been through this before).

In the case of the coil, the liquid pressure entering the coil has a saturation temperature. If the pressure is higher, the saturation temperature increases. However, the evaporator pressure is controlled by the suction pressure. This has a saturation temperature too.

What we have is; if the hand expansion valve is not properly adjusted, it does not reduce the liquid feed pressure down to the evaporator temperature. This means part of the lquid will not flash off until it is first warmed up.

The volumes are easy... By mass, the fraction of vapor has a much larger volume than the liquid. So in essence, the vapor occupies a much larger percentage of the pipe, than the liquid does.

Since the liquid is not surrounding the entire surface of the tube internal surface, part of the capacity is lost.

This pooling of liquid is also common in the suction lines we deal with, as all of the liquid does not evaporate. That's why we call then liquid overfeed.



How is it possible to have a subcooled liquid with vapor existing?


This is why the pan of water example was used. Water at 70F (21.1C) has 142 degrees of subcooling present. Water boils at atmospheric pressure at 212F (100C). So if we take... 212F - 70F = 142 degrees (100C - 21.1C = 78.9 degrees).

Water in a pan at 70F (21.1C) setting on the kitchen counter is subcooled with vapor on top.

This is slightly off the topic Volnei started, but I think it is related to the top-feed coil question.

US Iceman
27-02-2007, 03:49 PM
Whenever the pumps would cavitate the liquid from all three would return to the LPR, overfilling it and shutting down the compressors when the high level safety tripped.


Now that is interesting. How did the liquid return to the LPR? This would indicate the liquid lines were draining backwards to the LPR. For this to happen the check valves in the pump discharge lines would have to have been wide open and the pump impellers allowing flow backwards through them.

I would be more inclined to believe the coils were full of liquid, or the suction lines were trapped and caused the flooding during a suction pressure pull down.

When the suction pressure started to decrease, the liquid started to boil and pulled all of the liquid back to the LPR. If the suction pressure pulled down to fast this could have caused the pump cavitation. I've seen this before on blast cells after defrost.

I share your concerns about the top-feed coils, but for another reason. We always see warnings about not using top-feed coils for hot gas defrost, or not to use them below 32F (0C). Both of these are for the same reason I think. It's related to how the coil is circuited and the liquid feed header.

I think bottom-feed coils are used so frequently because the coil manufacturers know how to circuit the coil this way. And that the bottom-feed coil has some advantages in distributing the liquid up through the coil during operation. Remember, these coils use orifices in each coil circuit tube in the liquid header.

On a bottom-feed coil as the vapor begins to form it rises. the gas bubbles help to push the liquid through the coil. During this, the liquid is better distributed around the inside tube surfaces.

I have a friend who swears top-feed coils are better for defrost, operation, and performance if you do certain things.

taz24
27-02-2007, 04:34 PM
Water in a pan at 70F (21.1C) setting on the kitchen counter is subcooled with vapor on top.

This is slightly off the topic Volnei started, but I think it is related to the top-feed coil question.


I would agree with most of what you say but and there is always a but somewhere:)
A pan of boiling water at atmospheric pressure would be saturated would it not. Saturated liquid and the vapour would be saturated vapour.

Water at 99c would be subcooled
water vapour at 101c would be superheated.

taz.

Sergei
28-02-2007, 02:13 AM
Since we have gotten onto this topic I would like to see some commentary on Top Vs Bottom feed recirculated coils.
My experience with top feed has not been good.
I understand that they may be better for defrosting in some ways, however I had three top feed evaporators drive me nearly insane.
We had a pump cavitation problem develop (which was never fully explained). Whenever the pumps would cavitate the liquid from all three would return to the LPR, overfilling it and shutting down the compressors when the high level safety tripped.
We do not install many top feed coils and I have never figured any advantage to them.:confused:
Comments?
Regarding to cavitation. I agree with US Iceman that significant changes of pressure in this vessel most likely lead to cavitation. Sometimes partly closed discharge valve will help to stable pump operation.
Regarding to level. Overfeed system usually have a lot of liquid in suction line. When pump is running normally, we have a balance of pumping liquid and returnig liquid. When pump stop pumping(cavitation), all liquid from suction line just drain into the vessel. More coils were operating, more liquid in suction line, more liquid will drain

Sergei

NH3LVR
01-03-2007, 03:25 AM
Sorry I took so long to reply. I am playing Road Warrior this week, and living in Motels. The Evaps were free draining back to the LPR. When the pump started to cavitate the liquid all returned to the LPR.
The cause of the Cavitation was a mystery. Perhaps a good subject to discuss at a later date.

US Iceman
01-03-2007, 05:33 AM
Hi NH3LVR,



The Evaps were free draining back to the LPR. When the pump started to cavitate the liquid all returned to the LPR.


Now I am really curious about this. Perhaps when you get back from your road trip, we can start a separate thread on this.:confused:

Sergei
01-03-2007, 05:18 PM
I just found this link.
I have only scanned it briefly, but it looked interesting.

http://nhtres.com/technote01.pdf
I gave some thoughts to this article.
Assume that heat transfer coefficient of latent heat transfer is 20 times greater than sensible heat transfer. However, air side coefficient didn't change. U- value of the coil will change only 2 times(not 20 times) or 50 %. Temperature difference for sensible heat transfer will increase due to subcooling. Operating temperature difference is 13F. Subcooling is 17F. Average subcoolnig is 8.5F. Average temperature difference for warming up subcooled liquid is 21.5F. Temperature difference will increase 21.5/13 = 1.65 times. Total heat transfer will go down only 20%. If 50% of coil warming up subcooled liquid and 50% have boiling refrigerant, capacity of coil is going down just 10%.
Major reason of poor performance for top-fed evaporators is lost heat transfer surface for subcooled liquid. To eliminate this problem we have to increase liquid feed to fill up the coil pipes with subcooled liquid, but not reduce liquid feed.