Refrigerant mass vs flow

[chilly1]

If for instance you use ethlyene you will need twice the flow to equil the same capacity of R14.
Started up my ethlyene rig today Low side temp is -93C with no insulation on the pipes only the highstage. Interstage is -35C
Ethlyene charged with unit off to 275 psi. this is my static at 20 C
This means there is only gas in system. Total system charge is 8 oz. All is well however, the gas flow rates are higher and the suction runs at 5psi to get these temperatures. If I go any lower than this I get a rise in temperature,,, Un insulated this is the bottom... load was about 100W... These temperatures are load independant, as I ran the load up to 150 and the temp of the block stayed the same. There was a rise in the discharge pressure under the increased load. Discharge went to 210 psi when it was 180 with 100 W. . My high side pressures with the load or without stay about the same. 180 psi high and 18 suction... I am runnning about a -5 suction at the compressor...

I am running two sc12mlx compressors... and a 10kbtu lytron condenser...

[DaBit]

R1150: Enthalpy of evaporation @ -100C is 94 kJ/kg according to CoolPack

R14: Enthalpy of evaporation @ -100C is 247 kJ/kg according to CoolPack.

Thus, based on this information (and to be honest I don't thrust the R14 numbers; Coolpack often glitches with such low evap temps), you need 2.6 times as much mass to cool a load X with ethylene.

How does this translate to volume?

Density of liquid R14 @ -100C = 1470 kg/m3
Density of liquid R1150 @ -100C = 562 kg/m3.

Thus, to cool the same load X with ethylene, we need a volume flow of 247/94 * 1470/562 = 6.87 times the R14 volume flow.


Hmm. I don't believe the enthalpy numbers.....

[DaBit]

Now, this is more accurate. Data taken from Air Liquide datasheets. No numbers for the same evap temp, but the difference isn;t huge

R14:
- density @ -128C: 1603 kg/m3
- latent heat of vaporization: 135.95 kj/kg @ -128C

R1150:
- density @ -104C: 568kg/m3
- latent heat of vaporization: 482 kj/kg @ -104C

Same math: for the same load, volume flow of ethylene will be 0.8 times volume flow of R14.

[Gary Lloyd]

Hmmmmm... interesting stuff. I knew there was a difference, but had no idea it was that great. How do the more common refrigerants compare?

[DaBit]

Which ones would you like to see compared?

By the way, this isn't everything. Refrigerant heat conduction is also an important parameter since higher heat conductivity lowers SST to metal temperature difference. Vapour density is also important since the stroke of the compressor is fixed.

[Gary Lloyd]

R134A
R22
R290
R404A
R507
R410A

Sooner or later, a cap tube chart will evolve in here, based on heat load instead of capacity. That would be very handy. :D

[chilly1]

Molecular mass R14 is 88 , R1150 is 28.8 this does mean that R14 is the denser material. I noticed that my pev needed to be open less for the R508 than the R1150/
Good idea Gary all we need to do is figure the vapor pressure in the evap times the displacment over time and multiply this times the latent heat and and this will give us our capacity. Now all we need to do is figure out the flow rate using different size cap tubes. Anybody up on turbulance in complex systems...

[DaBit]

Originally posted by chilly1
[B]Molecular mass R14 is 88 , R1150 is 28.8 this does mean that R14 is the denser material.

Sure, R14 density is also a lot higher. But that doesn't say much about the latent heat of evaporation.


I noticed that my pev needed to be open less for the R508 than the R1150/

R508b is another beast than R14. The only advantage of R14 is it's low boiling point. For the rest it is quite a §§§§ty refrigerant. Heat conduction and capacity is low, which means that it will probably cause discharge temperature problems too.


Good idea Gary all we need to do is figure the vapor pressure in the evap times the displacment over time and multiply this times the latent heat and and this will give us our capacity.

It's not that simple. Subcooling also plays a big role, and latent heat varies with temperature and pressure, as does heat capacity etc.

That's why we have a lot of different equations of state to calculate various properties of refrigerants. These are not simple equations. Proof:

http://www.icecoldcomputing.com/generic/imagehandler.php?hval=c34c8107c660c92875c5bbae8fdb c2b8&quality=high


Now all we need to do is figure out the flow rate using different size cap tubes. Anybody up on turbulance in complex systems...

A research carried out by a guy named Wolff in 1995 resulted in an application of the Buckingham pi theorem with the right coefficients to calculate massflow through a captube from a given set of input parameters.

I have tried to convert this into a useable program, but I don;t have the full article, only two lousy bad quality copies with the formulas on it. Which is not enough to implement.

The software is at the point where I can convert load in Watt at a given evap temp with given subcool and superheat into mass flow, and the massflow matches Coolpacks output. It also implements the equation of state for various pure refrigerants. Zeotropic mixtures like R404a is still a problem. But I am missing something since massflow output is wrong by a factor 0.1-100.

These are the required parameters to calculate the mass flow:

http://www.icecoldcomputing.com/generic/imagehandler.php?hval=5f58f08977b5a15339659721104b 6204&quality=high

If anyone can obtain that article, I would be very glad to receive a copy. If anyone owns the ASHRAE refrigeration handbook, I would be glad if you gave me the correct literature number or whatever I need to get the article. I am sure I do many people a huge favour by writing such a program.

[chilly1]

I will be returning to the college for the spring semester in a couple of weeks, The book store will open at that time, I will have them issue me an ashra book and get the formula for you.. The textbooks that I use in my classroms are geared toward the mechanic and not the engineer but I do have access to engineering and other data I will see what I have. I lost all my texts fron the university when my home burnt, and I never replaced them.. I think one of my students has a couple of engineering texts. I will also look at the university's library for you...

[Russell_hq]

DaBit, do you have anymore information on this article? where it was published, i.e. the journal title? the title of the article? I have access to a large database of journals and i may be able to find it, i tried using Wolff and 1995 but this return alot of results. Can you give any more information so i can narrow down the search.

[DaBit]

No, unfortunately not. I have a few scanned pages of the ASHRAE Refrigeration manual, and in teh captube section it mentions this model and describes it as work done by 'Wolff (1995) et al.'

The remark that calculated results are less than 5% off from measured values is promising...

With the original article I should be able to finish the software.

To gather the refrigerant data, the REFPROP program from the NIST would be nice too since I lack the equations of state for many refrigerants. Probably even REFPROP61.DLL alone is enough.

But spending $200 on it just to do you guys a favour is out of the question.