Air/Oil cooler, curve fit

[Jacc]

I can find it somewhat annoying when some fairly stressed and ignorant person steps into my field of work and asks for a "quick and dirty solution" or a "quick estimate" totally ignoring the complexity of the topic and doesn't even try himself.

Well, that being said...
Ok, I'm not stressed. I tried to search a little on this forum but I was unable to find something on a reasonable narrow search.

Basically I'm looking for cooling power of an Air/Oil cooler as a function of air flow and oil flow. Primarily as a function of air flow, actually. Vendor software only give a couple of data points (3 or 4, maybe 5). Ofcourse I put those points in Excel and added a trend line, but I don't think any of the curve fit options that Excel offers are very suitable.

I would think that a function like:
Pcool = Oilmassflow*oilcp(1-e^(-airflow))
could be made to fit pretty well (multiplying by the usual constants here and there ofcourse). Oh wait! Maybe not since the oil can only reject heat so fast.

Anyway, I hope you understand my problem. Is there a certain type of function that you would recommend? Or does anyone know of a air/oil cooler vendor that supplies this data? We are talking air/oil coolers with a cooling power of around 2,5kW/°C.

 

[MintJulep]

The equations governing heat exchanger performance can be found in any heat transfer text.

If you don't have one on your bookshelf a perfectly fine one may be downloaded (legally and free) here:

http://web.mit.edu/lienhard/

http://www/ahtt.html

 

[djack77494]

Dr. John Lienhard's generously "donated" and excellent heat transfer textbook will have all you want and more. I encourage you to download it and then read it.

In the meantime, please be aware that you are seeking to RATE an existing Air Cooler or (presumably) Shell & Tube Heat Exchanger. An exchanger's performance is governed by the standard heat exchange equation:

Q = U * A * Ft * dTlm

This equation must be solved in conjunction (or simultaneously) with the energy conservation equations governing both fluids. The latter equations are generally expressed as:

Q = m * Cp * (T2 - T1)

for each fluid (assuming no latent heat effects). The energy lost by the cooled fluid (oil)equals the energy gained by the air or water (neglecting losses). This is best set up using a spreadsheet, and again, I would encourage you to do this exercise.

In the meantime, you can "google" to find an on-line heat exchanger calculator or a suitable spreadsheet to download. At this point, I feel I've given you enough of the "easy" answers, so I'm not going to google the

http://www for

you. Good luck and don't neglect my encouragements.
Doug

 

[Jacc]

YOU!? Again!?

Just kidding, MintJulep! Thanks for the tip, I downloaded the book. But it's not like the answer jumps out of the book and smack you in the face, it's still some heavy reading that needs to be done. I'll give it a try but I would certainly appreciate some more input from people with heat exchanger experience.

 

[Jacc]

There is always someone that manages to post a reply while I'm typing mine. I haven't even read your reply, djack77494, just wanted to point out that my last post was supposed to be before yours. What can I say? I'm a slow thinker.

 

[Jacc]

Ok, now I've read some of Dr. John Lienhard's book. Very generous of him to donate it indeed, I did mail him and thanked him for it. I'm right now reading around p. 119 and I've learned about LMTD and the LMTD correction factor F.

>djack77494
In your equation
Q = U * A * Ft * dTlm
is Ft the correction factor and dTlm LMTD? Or haven't I read enough? Well I haven't read enough, I've only started but now it's time for sleep on this part of the planet.
Thanks for your inputs! I'll be back in a day or so.

 

[Jacc]

LMTD=Logarithmic Mean Temperature Difference

 

[Jacc]

I assume there is supposed to be a dot above the m in equation
Q = m * Cp * (T2 - T1) ?

 

[speco]

Jacc,

Right. It's mass flow rate times specific heat time temperature difference.

Speco

 

[Jacc]

I forgot my notebook where I have the solved equations at work but if anyone here speaks Excellish please see if this looks familiar:
Q=(Tin_oil-Tin_air-(EXP(-UAF*(1/(m_oil*cp_oil)+1/(m_air*cp_air)))))/(1/(m_oil*cp_oil)+1/(m_air*cp_air))/1000
I'll try to submit an easier to read variant soon.
The m's are supposed to be m dots ofcourse.

 

[vpl]

Jacc

I get the impression you're trying to combine the two equations, but I don't believe you have done it correctly.

Realize that you have 6 factors in the first equation:
Oil flow rate (m_oil)
Oil incoming temperature (T_{oil, in})
Oil exit temperature (T_{oil, out})
Water flow rate (m_water)
Water incoming temperature (T_{water, in})
Water exit temperature (T_{water, out})

The second equation is solved separately on each side, but the two heat transfer rates are equal. In other words:
Q_oil = Q_water or
m_oil*C_{p, oil}*[T_{oil, in} - T_{oil, out}] = m_water*C_{p, water}*[T_{water, out} - T_{water, in})].

You can generally solve the heat transfer equations if you know 5 of the 6 variables. If you don't know that many, you can make some intial assumptions and iterate until you arrive at final values that are acceptable to you.

(and all the "m"s should be "m-dots".)

Patricia Lougheed

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[djack77494]

Jacc,
Sorry for not responding quicker - I've been away from this forum for a few days now. Anyway, it looks as if that's been no problem whatsoever. Others, esp. Patricia, appear to have fully resolved any open issues here.
Good luck,
Doug

 

[Jacc]

Thanks for your thorough response, Patricia! Amazing how helpful people are here! I'm sorry about my poor follow up, I've been busy at work with boring work. This is fun work. Well, I guess I did not solve that equation right. Putting my calculated Tout values back in Q = U * A * Ft * dTlm didn't give the same answer but since I liked my solution I choosed to ignore that (it could be me calculating dTlm wrong when checking for instance).
Anyhow I've been given another task that is important and fun (this one is just fun) so I have to prioritize that. And because of that I will post a post in the Mathcad forum soon asking about the slow scroll speed.
But I will get back to this topic soon, I'm good at getting back to things again.
By the way, how do you format the text like that? I've looked around but I can't see anything obvious. But I haven't read the "Read this before you post" text, maybe the answer is in there.

 

[vpl]

Try clicking on the "Process TGML" link. Also [ṁ] will be there soon. :-D

Patricia Lougheed

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.