liquid overfeed

[nh3simman]

Can anyone give a clear indication of the effect of liquid overfeed on the inside refrigerant film coefficient?

[US Iceman]

Can anyone give a clear indication of the effect of liquid overfeed on the inside refrigerant film coefficient?


That is a really good question. I have read some old articles that say the recirculation rates are different than top feed coils or bottom feed coils. However, I also think is biased by the type of distribution method used in the liquid headers.

Personally, I think this needs to be based on the actual coil circuiting used and the quality of the mixture as it passes through the circuit tubes.

If we considered the case of too much liquid flow into the coil, the refrigerant might not evaporate. So the film coefficient would decrease.

On the other hand, if the recirculation rate was lowered the inside of the tube might approach an annular flow regime which would offer the best film coefficient for forced convection.

John Thome has written some interesting articles on some of this related operation with two-phase flow and heat transfer, but I'm not sure if his articles have any ammonia data.

If you have some information you can share, I would love to read a copy.

[Sergei]

Heat-transfer coefficient is going up 25% for overfed coils. W.Stoecker book(new one) p.305

[US Iceman]

Heat-transfer coefficient is going up 25% for overfed coils.


The heat transfer coefficient or the overall total heat transfer?

I would believe the overall total heat transfer increases 15-20% if you compared a DX evaporator to the same coil circuited for liquid overfeed.

In a DX coil a portion of the surface area is used for superheat. In the same coil circuited for liquid overfeed this same surface area is used for additional heat transfer (boiling not superheating).

[nh3simman]

Hi US Iceman, thanks for the comments. I am surprised that the ASHRAE Refrigeration Handbook, which has a whole chapter on liquid overfeed, does not even mention the effect on the film coefficient.

Sergei, you say 25%. I doubt this. If the look at the order of magnitude of a DX coil.
1/Uo = 1/ho + B/Ui
ho=30W/m2degC
Ui=2000W/m2degC
B=15
gives
Uo = 24W/m2degC
Now double Ui to 4000W/m2degC
this only increses Uo to 27W/m2degC
only 12.5% increase

But, I have yet to see if the recirculation could double Ui. After all, boiling is already a pretty vicious activity on the surface.

[nh3simman]

Heat-transfer coefficient is going up 25% for overfed coils. W.Stoecker book(new one) p.305

I assume you are referring to Stoeckers Industrial Refrigeration book. I will check it out today.

[US Iceman]

Hi nh3simman,

Most of the manufacturers rating tables (here in the US) give a capacity increase of something like 12.5-15% when comparing DX ratings to liquid overfeed/flooded ratings. The usual circulation rate is 3:1 or 4:1, nothing else is normally listed.

Looking at the chart Sergie mentioned shows a similar effect. As you increase the amount of liquid in the evaporator, the capacity rises. In my opinion this is strictly due to the coil surface being fully wetted, instead of a portion (of the coil surface) being used for the addition of superheat (for the TXV control).

I'm not sure how they saw a 25% gain (perhaps the coil was wet with no frost). That's the problem with graphs like this. Not enough information to clearly state the assumptions used.

For all I know this graph could be based on a DX frosted coil and then compared to a wet overfeed coil. That would make a big difference.;)

I'm pretty comfortable with the 12.5-15% increase which is in the same realm as your calc's stated.

Of interest though, the chart in Stoecker's book started to flat line after the circulation rate reached about 2.5 to 3. That tells me this is the practical limit for excess liquid.

The flip side to this argument would be too much liquid being pumped could start to reduce heat transfer. What the old-timers used to call brining (all sensible heat transfer, no phase change).

[nh3simman]

Hi nh3simman,

... As you increase the amount of liquid in the evaporator, the capacity rises. In my opinion this is strictly due to the coil surface being fully wetted, instead of a portion (of the coil surface) being used for the addition of superheat (for the TXV control)...





I think that the increase in wetted surface is the answer here.

But, consider this. A flooded coil (refirc ratio = 1) is already fully wetted. If I increase the circulation rate from 1x to 3x, I find it difficult to justify an increase in heat transfer.

[Andy P]

I think that the increase in wetted surface is the answer here.

But, consider this. A flooded coil (refirc ratio = 1) is already fully wetted. If I increase the circulation rate from 1x to 3x, I find it difficult to justify an increase in heat transfer.

Hmmm - that depends on what you mean by ratio = 1: there is usually a bit of uneven distribution between circuits so some will be overfeeding and some will be starved. It might even not be the same circuits being starved all the time as the liquid may by "hunting" between circuits. I guess you would need to get to at least 1.25 to be sure of wetting all the circuits all the time.

Check the Wolverine e-book (produced by John Thome) for good information

http://www.wlv.com/products/databook/db3/DataBookIII.pdf

cheers

Andy P

[nh3simman]

Thanks Andy,

The Wolverine pdf is excellent. I will have to study it in detail but I think that it resolves my question about the heat transfer coefficient with different feed ratios.

[US Iceman]

I guess you would need to get to at least 1.25 to be sure of wetting all the circuits all the time.


I agree with this. I believe the liquid distribution problems of the coil are overcome with a ratio of at least 1.25 to 1.5. If we could uniformaly distribute the liquid in the coil under all conditions then a 1:1 ratio would be all that is needed.

I suspect the poor distribution in the coils is why the higher pumping rates are used. It's an attempt to pump enough liquid to ensure the coil tubes are wetted.

On the other hand, I have converted some DX NH3 evaporators to liquid overfeed and they work just fine at the lower rates.



Check the Wolverine e-book (produced by John Thome) for good information


Yes! This version of the handbook has a lot more information on the mechanisms of heat transfer. Mr. Thome has also contributed to some very nice articles in the Journal of Heat Transfer (ASME) on two-phase heat transfer and flow regime maps you may want to review.

[Samarjit Sen]

Hi Mike,

I was trying to find out some details about the Liquid Overfeed System and this thread was nearest to what I want to know.

I would like to know as to how you evaluate the refrigerant flow rate aupplied to the evaporator and the flow rate of refrigerant vaporized. As per the Coil manufacturers they have mentioned n = 3.

Kindly guide me as you know I have no experience in Ammonia and this is a small project which we might get as our ***** Plants have been performing pretty well, and now the client wants to set up another plant with NH3.