Air cooled heat exchanger sizing and pressure drop correlations

[Barry675]

Hello,

I'm trying to size an 18 kW air cooled heat exchanger (HX) that needs to be mounted on a truck. The HX obviously has very tight space requirements and mass requirements (1 x 1m face area, 0.2m depth and max 60-70kg). In addition to the physical requirements, I also have a very tough power budget in terms of parasitics (max 0.6 kW for blower fans).

Blower fan power is the most powerful factor driving the design optimization. I obviously want to keep size and mass to a minimum but I must have a sufficiently large face area to keep the pressure drop/fan power down. I am trying to find a good correlation for the air side pressure drop as I want to double check the values that I've already calculated.

I have found several different friction factor correlations online and in books for my configuration but they can differ by a factor of 10 depending on which one I used and so the overall fan power requirement is extremely wide depending on what I use. This is the key factor driving the overall design so I have to be sure that I am within say 30% of a believable value.

You can find more details below in my spreadsheet:

https://www.dropbox.com/s/dp8y7dqmt0avqi5/Radiator Sizing Notes (Full Scale System.xlsx?dl=0

I would appreciate any help that people can offer or if you could take a quick glance at the spreadsheet to see if the values are believable. One note, I have not properly dealt with the fin efficiency yet as I'm struggling to understand the calculation method (dividing the finned area up into virtual hexagonal areas?).
My background is in physics and I am now doing a mechanical engineering PhD so I lack a lot of "gut feeling" for heat exchanger type problems. I don't know whether my blower power is reasonable or 10 times what it could/should be.

Kind regards,

Barry

 

[EdStainless]

Some of the radiator makers publish their own design guides based on their actual production values.
I would start hunting for those. Trying this form theory could have you off by 50%.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[georgeverghese]

You'll probably find everything you need to know in DQ Kern " Process Heat Transfer " in chapter 16 - chapter on Extended Surfaces , which covers both fin eff and pressure drop in crossflow banks for transverse fins. See page 555 for the graph on pressure drop.
Beware that the value for the heat transfer coeff at the fin surface may not be the same as the external surface htc for the bare tube (due to skin effects as air flows through the gaps between closely spaced fins), so a correction factor for this may be prudent.

 

[Barry675]

Some of the radiator makers publish their own design guides based on their actual production values.
I would start hunting for those. Trying this form theory could have you off by 50%.

Would you be able to suggest some manufacturers that I could try contact? Is it standard practice to contact a manufacturer with your overall design requirements and they can suggest a heat exchanger for you?

You'll probably find everything you need to know in DQ Kern " Process Heat Transfer " in chapter 16 - chapter on Extended Surfaces , which covers both fin eff and pressure drop in crossflow banks for transverse fins. See page 555 for the graph on pressure drop.
Beware that the value for the heat transfer coeff at the fin surface may not be the same as the external surface htc for the bare tube (due to skin effects as air flows through the gaps between closely spaced fins), so a correction factor for this may be prudent.

I had a look at that book just now but it appears the example given is for circularly finned tubes. The correlation for that case and the fin efficiency is much more straightforward than that for continuously finned tubes. Any other thoughts?

Regards,

Barry

 

[georgeverghese]

Not familiar with these "continously finned" tubes? Can you post us a picture or sketch?

 

[EmmanuelTop]

For ACHE sizing, check GPSA Chapter 10, in particular figures 10-14 and 10-15, as well as the worked out example.
Hudson Company also offers free software for preliminary sizing of Air coolers:

http://www.hudsonproducts.com/Default.htm

Dejan IVANOVIC
Process Engineer, MSChE