Glycol Effect on Chilled Water Coils

[BronYrAur]

What am I missing here? I know glycol has a derating effect on heat transfer, but the numbers I am seeing here don't make sense. I know that based on density and specific heat changes, a solution of 30% propylene glycol at 40F should have less than a 10% derate in heat transfer capability. Correct?

I ran a chilled water coil selection using 40 degree water. I then kept the airflow and inlet temperature the same, and I kept the entering fluid temp and GPM the same. All I did was switch from water to 30% propylene glycol, but the total heat transfer was significantly less (34% less). See attached selections.

What am I missing to explain this difference over what the density and specific heat changes would suggest? Thanks for your help

 

[itdepends]

A 10% change in heat capacity does not = 10% change in heat transfer.

By keeping the glycol flow the same as the water flow you've had a disproportionate impact on the LMTD which is impacting on your heat transfer. Sorry I can't comment further without looking up the program and the acronyms used for the inlet/outlet fluid temperatures.

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[TD2K]

The inside heat transfer depends on the thermal conductivity of the solution, heat capacity and viscosity as well as the flow and ID which in this case is constant. You have air on the other side which is a lousy heat transfer medium (but it's free) which is why you have finned tubes on the air side. Since the air side temperature is nearly constant, the change in dT on the liquid side is due to mainly changes in the inside film coefficient and the change to a glycol solution isn't helping you.

Water is a great heat transfer medium, low viscosity, high specific heat, high thermal conductivity. It's hard to beat. If only it wasn't susceptible to freezing and corrosion.

I have some reference material which gives you an indication of how the inside heat transfer coefficient changes with fluid properties but it's at work, will have to look it up tomorrow.

 

[TD2K]

So, the reference I had is in the GPSA data book in the shell and tube exchanger section.

Effect of a change in viscosity on the tube side inside heat transfer coefficient is (u2/u1)^0.47, turbulent flow.

For thermal conductivity, it's (k1/k2)^0.67

For specific heat ratio, it's (Cp1/Cp2)^0.33

For mass flow rate, it's (G1/G2)^08 (you have a constant volumetric flow rate which is not a constant mass flow rate).

I didn't dig into the inside heat transfer film coefficient equation to confirm the above order but I suspect that is why viscosity is u2/u1 which the others are the original value divided by the new value.

This is only 1 part of the overall heat transfer coefficient. You have the air side resistance and any fouling coefficients that could be built into this program, these tend to off-set the effect of a change in the inside heat transfer coefficient. Given the change you are seeing, the air side isn't as controlling as much as I thought it might be.

 

[BronYrAur]

I see my error about GPM versus mass flow, but that still doesn't get close on the 2 coils. So if I'm understanding correctly, the physical properties of the coil are causing the difference. If so, I have never realized that before. The literature always talks about a reduction in capacity when glycol is added, but the correction is always listed as simply increasing flow by some percentage. The coils I attached are just for my own understanding, but I have run into situations before where someone wants to add glycol to a system that used to be water. Until now, I had only thought that the pump may be an issue because it is more difficult to move glycol (especially cold) than water. And I also knew about the reduction in capacity that I thought could be overcome by just increasing flow.

But if I am understanding correctly, every piece of heat transfer equipment may no longer deliver the heat it did before. Even if flow is increased by the "correction factor" listed in the glycol literature. Correct?

 

[IRstuff]

What you seem to keep missing is that the thermal conductivity of 30% glycol is significantly lower than that of pure water on the order of almost 25%

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[BronYrAur]

Yes I am missing that ..... and it would seem that most of the glycol literature is missing that too. The majority of what I see discusses reduced specific heat and increased specific gravity. So there is concern about having to pump more of a heavier fluid. But the literature leads me to believe that simply increasing flow makes the heat transfer issues go away. And I am learning here that it doesn't. Here is a typical example that makes it seem as though an increase in flow makes up for the reduced capacity of glycol.

http://www.deppmann.com/wbcntntprd/wp-content/uploads/2010/09/printer_friendly_2010_09_13.pdf

I would be delighted for an article that actually discusses all of this. All I can find are either generic ones that just list "correction factors" or ones that list tables of physical properties without explaining how they affect the heat transfer.

 

[IRstuff]

Any decent heat transfer text would cover the subject. The whole concept of thermal diffusivity is tied directly to thermal conductivity

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[BronYrAur]

So to close the loop of this, the part I was missing - and the part that almost all of the glycol literature misses - was the heat transfer. Being able to contain heat (specific heat) is one thing. Being able to transfer it is something else (thermal conductivity). Selecting a coil knowing about the glycol is fine, but attempting to put glycol in coils that were selected for straight water may lead to serious trouble. I typed up the following to send to some others around my office as a reminder. Do you think I accurately explained things?


Just a reminder about adding glycol to existing systems that don’t have it now. There are 3 main detriments to adding glycol to a system:

1. Glycol has less heat capacity than water.
2. Glycol is more difficult to pump than water.
3. Glycol has less thermal conductivity than water.


Most of the glycol literature does a good job of explaining the first 2 detriments; however, the literature almost always neglects to mention the third. Having less heat capacity means that it takes more glycol to deliver the same amount of heat as straight water. So it takes more of a heavier fluid to convey the same heat energy as water. It takes more pump head to move the glycol around, and you get less heat out of it. The glycol literature does a good job of reporting this and lists “multipliers” or “correction factors” to account for the difference. Most of the literature boils it down to saying that a certain percent increase in flow will make all of the problems go away. They also point out that a pump change may be required due to the increased head, but otherwise, they lead you to believe that more GPM is the answer. THIS IS ONLY PART OF THE STORY!

It must be understood that this only accounts for the heat capacity of the glycol and NOT its ability to TRANSFER that heat. Glycol also has less thermal conductivity, so it will not give up the heat as easily as water. Almost all of the literature neglects this part, which can come back to bite you. If you are installing a new system with glycol, all of the equipment, heat exchangers, coils, etc. will be selected with glycol, so there should be no issues. If, on the other hand, you are putting glycol in an existing water system and re-using coils, you may now have grossly undersized heat transfer devices.

The magnitude of the problem changes with fluid temperature, and it is much more prominent for cooling applications. For heating with a glycol temperature of 180 degrees, the heat transfer from a coil will be on the order of 10% less than it otherwise would be with straight water. You may be able to get away without changing the coil. A 40 degree cooling coil with glycol, on the other hand, may only deliver on the order of 50% of the heat transfer that it did with straight water.

So bottom line, if you are adding glycol to a chilled water system, be extremely careful that your equipment and coils are going to function correctly. Simply increasing the GPM won’t cut it!

 

[gruntguru]

Surely there is a fourth detriment. The higher viscosity of glycol affects its ability to convect (ie remove heat by physically moving hot glycol molecules away from the source). This, combined with the strong temperature dependence of glycol's viscosity (much stronger than water) accounts for the increasingly poor heat transfer performance as temperature is decreased.

At 200*F Ethylene glycl has 4x the dynamic viscosity of water.
At 40*F it is 30x.

je suis charlie

 

[georgeverghese]

Agreed, viscosity effects on the heat exchanger, especially when using PPEG solutions, can also reduce heat transfer. This applies to the HX which rejects heat and to the one which absorbs heat also. The other thing to watch out for is the startup condition when the PPEG solution may be cold and much more viscous - what minumum circulation rate should the pumping circuit enable?

 

[BronYrAur]

Thank you for the further clarification on viscosity. That's where I was going with my comment about the problem being worse for chilled water applications as opposed to hot. Thanks all for you comments and assistance on this.