Heat Exchanger Question

[LPorc]

I've run across a statement that doesn't quite make sense to me regarding heat exchanger efficiencies. The statement was that in terms of area counter-flow is more efficient then parallel flow, and that cross flow would be somewhere between the two.

Part of this information was based on the LMTD of the setups, and so far all the info I have been able to easily assimilate uses a correction factor for the geometry of cross flow, and vary wildly to say the least when it can be found at all - which is already a bit shaky since LMTD itself needs a correction factor.

It just seems like a cross flow, like a simple single pass radiator, ought to beat a counter flow of the same area since the temperature difference is greater over a larger portion of the area (even though it would seem to behave as a parallel flow in depth).

What's frustrating me is the correction factor seems to be more of a measuring actuals and then going back and adjusting the numbers to fit. That's one heck of a way to predict performance, after the fact!

Anyone got any links or explanations that will help me wrap my head around this?

[greenman100]

cross flow=singlepass rad, ok 90 degree angles, then?
parallel flow= what do you mean here, flow of liquid and gas in same direction?
Counter flow='Making sure this occurs in a rad requires that the liquid in the tubes moves through the coil in the opposite direction (counter flow) to how the gas, on the fin side, moves through the rad. This even applies to chambered cores but to a much smaller extent as cores are a much smaller system and the difference is very small ."so opposite direction is key here, got it, if I understand you correctly


going to google now

pieces are coming together


Well, it seems to me counter flow is best because it maintains the greatest deltaT

crossflow is second best because some of the rad has a high delta T, some has low

parallel means there is no deltaT by the end of the rad


dunno if I explained that very well, but crossflow does seem mose efficent, if I am understanding correctly

[BillA]

wheeezzzt
Occam's razor

no LPorc, the math is there; go to a text on Heat Exchangers

[Les]

Originally Posted by unregistered

wheeezzzt
Occam's razor

no LPorc, the math is there; go to a text on Heat Exchangers

William Ockham's Hamlet?
However some good heat-exchanger text - Wolverine

[LPorc]

Originally Posted by unregistered

wheeezzzt
Occam's razor

no LPorc, the math is there; go to a text on Heat Exchangers

Aside from my problems with integration, the math gives the answer at least as far as I have followed it However the math is a model used to predict behaviour. I can plug numbers into equations just as well as anyone else, I am looking for the concept behind the mathematical model. My forte is juggling concepts, not digits.

Actually, I think the integration is clouding my view. My math skills are quite rusty and I never did follow the subject very far post-Algebra. Instead of looking at it as ugly and painful math, I should look at it for what it represents.

Staring at the curves and thinking about it some more it is starting to make a little more sense. I hypothesize that a cross flow exchange is effectively a special case of parallel flow exchange. In our typical (or not so typical) single pass single thickness radiator with a relatively infinite supply of ambient air we've got an artificial case of cooling flow with a high specific heat, so we've just made clever use of what we have available. Each tube is essentially a parallel exchanger, and we have parallel flows in parallel, which still can't beat a counter flow in the same area of exchange.

Comments?