Analysis of Heat Exchanger

[sonckerr]

Hello gents,

I'm an intern and I've been tasked to run an analysis on our heat exchangers. We have HXs that are 1 pass and 2 pass exchangers. Water-to-Water and water-to-seawater. How can I calculate their efficiency/performance based on original design specs?

I have inlet/outlet temps, flows, and pressures. Possibly more points.

I appreciate any feedback!

 

[CastMetal]

You could start here.

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

http://www/ahtt.html

Comprehension is not understanding. Understanding is not wisdom. And it is wisdom that gives us the ability to apply what we know, to our real world situations

 

[sonckerr]

Do you know which equations to use and what points are needed?

 

[vpl]

Do you know which equations to use and what points are needed

yes. Generally you're going to need to know the heat rate being transferred, the tempertures (inlet and outlet) of both fluids and the flow rates for both fluids. If one piece of information is missing, you can generally calculate it.

Two equations are Q=[ṁ][Δ]T and Q=UALMTD (you can find the equations for the log mean temperature difference in the text book above).

Now how you put them together and get meaningful results -- that takes quite a bit of time, and usually at least an Excel worksheet (if not a computer program).

If you haven't taken heat transfer, go back to your boss and 'fess up. If you have, then go read the pertinent sections in your heat transfer text book (or use the free one given by CastMetal).

If you happen to have access to EPRI documents (ask your boss), there are a couple documents that outline methods to compare actual conditions with design ones. I can look up the titles, but it's no use providing them to you if you don't have access to them.

You might also search around the site (and this forum) -- there are free calculators out there and some forum members have posted solutions to other problems that you might be able to use.

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.

 

[IRstuff]

Firstly, you need to have clear definition of what the "efficiency" you are supposed to calculate actually means, i.e., are we talking about W/m^2 per gal-K/min, or what? So, before you go hunting for equations, you need to know what your equations are supposed to produce in the end.

In some gross sense, the net heat transfer coefficient of the whole exchanger in W/K or W/m^2-K might be considered to be an "efficiency" as well.

As an addendum tp vpl's comments, consider

http://www-unix.ecs.umass.edu/~rlaurenc/Courses/che333/Reference/exchanger.pdf,

which describes the design process for various heat exchangers.

Another question to be answered is what is the end goal of your assignment? Are they really expecting you to produce a high-quality result, or is this simply a learning exercise for you? In either case, you can start with the Wikipedia articles on heat transfer

http://en.wikipedia.org/wiki/Heat_transfer

and heat exchangers

http://en.wikipedia.org/wiki/Heat_exchanger

A final subject is your request for "equations." I think you know that you'll essentially have failed the assignment if you went back to your boss and asked for "equations." In many cases in both school and the real world, the "equations" don't necessarily already exist, and it's often the task of the engineer to come up with "equations" to describe and model what they're designing or analyzing. Moreover, internship is supposed to be a learning experience, and you will not get the full benefits of the experience if someone just gives you the "equations."

I would suggest that you read the Wikipedia articles and try to, based on the information in the articles, to come up with your own model for the exchangers, and only then compare your results with what's in the exchanger article in the first link.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[willard3]

I would recommend against using Wikipedia for anything but toilet tissue......it's unvetted crap.

 

[CastMetal]

willard,

You're showing your age dismissing wikipedia so quickly. While not always accurate many of the engineering related articles provide ample citation of the source material that most would consider well above the level of "toilet paper." As a starting point to quickly gather multiple "vetted" references you would be hard pressed to beat wiki.

Stealing from someone's signature on this site, "Trust but verify".

Comprehension is not understanding. Understanding is not wisdom. And it is wisdom that gives us the ability to apply what we know, to our real world situations

 

[IRstuff]

Wikipedia is neither completely unvetted, nor completely crap. There is some level of vetting, as indicated by the editorial comments on the backside of Wiki pages, and, in most cases, barring malicious edits, there is reasonably correct information, and certainly, sufficient information for the OP to get a simple, general, understanding of the topics in question.

By the same token, information posted on this site likewise contains some level of "crap." While there may, or may not, be postings that mitigate and dispute the posted "crap," there are no edits to remove alleged "crap" and the reader is always advised to double and triple check anything they read posted by unrenumerated and anonymous posters.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[sonckerr]

IF ANYONE ELSE COMES ACROSS THIS PROBLEM HERE IS MY SOLUTION:

I’ll tell you what I’d do. On the original spec sheet for the exchanger you should find the design heat transfer coefficient (U) and the surface area of the tubes. The heat transfer coefficient is the best way to tell how well the exchanger is performing versus design.



To calculate your current heat transfer coefficient you start by calculating the total heat transferred across the exchanger using the Q = m cp dT equation. You’ll have to decide whether to use the cold side or the hot side to do this calculation or do both and average the Q. If all the instruments (thermocouples and flowmeters) are correct, the Q should theoretically be the same regardless of cold or hot side. But as the great Yogi Berra once said “In theory there’s no difference between theory and practice, but in practice there is”.



Then you use the following equation Q = U A dTlm where Q is the heat transfer rate (found in the other equation above), U is the heat transfer coefficient, A is the heat transfer area from your spec sheet and dTlm is the log mean temperature difference. You calculate the log mean temperature difference as follows:



dTlm = (dT2 – dT1)/LN(dT2/dT1)

Where for parallel flow exchangers:

dT1 = T (hot in) – T (cold in)

dT2 = T (hot out) – T (cold out)

and for counter flow exchangers:

dT1 = T (hot in) – T (cold out)

dT2 = T (hot out) – T (cold in)



Then solve the Q = U A dTlm formula for U and compare that with your design heat transfer coefficient and you’ll know how well you are performing relative to design.



I should also just remind you to be sure you are dimensionally homogenous and use absolute temperatures (either Rankine or Kelvin) for all of these formulas.

 

[vpl]

Do you think you might need to take fouling into account?

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.

 

[sonckerr]

Can you explain how I should take this into account? The fouling resist. are low, about .0005.

I was considering having pressure guages installed on the 3/4" bleeder valves to give an idea of pressure drop.

We don't have any inlet/outlet points for flow either. Is it okay to use the original design specs for that or do they change noticibly over time? We may be able to rent ultrasonic clamps to get this but is the cost worth it? Possibly, I could use pump curves with the pump discharge pressure, but we have multiple pumps.

Thanks!

 

[vpl]

Can you explain how I should take this into account?

It's not simple. That's why I asked if you had access to EPRI (Electric Power Research Institute)publications. However, the difference between the design fouling and the actual fouling is usually why people go through the exercise of figuring out where they are now versus the design value, at least in my industry.

I'd be surprised if the fouling hadn't changed or that the flowrate was the same as design. Not that it doesn't happen, but I've seen a lot of the opposite -- Plants who routinely pump muddy water through their heat exchanger and then wonder why they're not getting the cooling they expect. And plant operators who close control valves during normal operation so that systems don't get overcooled or who add a new load that diverts some of the heat exchanger flow.

Unfortunately, I really don't have the time to walk you through how to do it (going back to "it's not simple.) Maybe someone else will.

But -- how do you know that you can trust what anyone here tells you if you don't understand what's being done? Not that anyone would lie, but I've seen often enough where someone posts a detailed solution, followed by ten different versions tweaking things.

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.

 

[rmw]

Your Hx's either have nameplates or TEMA design data sheets somewhere in the records. Find out the design data and determine the design duty. Very few Hx's run right on design point. Most design points have not only 'fouling factors' as part of the design sizing calculations, which makes them a little larger than actually requied, but they have (a) the engineer's fat, (b) his manager's fat, and (c) his manager's fat added to the design, so they are over designed (or ironically grossly underdesigned - how does that happen?)

Then take your actual conditions, flows and temperatures and calculate where the Hx's are with respect to their original design duty point. The formulae have been given above amply to get the job done. Tell your boss that you will have to be able to accurately measure the flow rates into each of these Hx's and that without that, you can't get the job done. (I am assuming that your temp measurements are reliable).

rmw

 

[davefitz]

I would suggest using the e- NTU method to predict or evaluate the HX performance, as discussed in the classic text "Compact Heat Exchangers". In fact , nearly all the form factors or F factors used in heat exchanger texts that do use the LMTD method will in fact calculate the F factor using the e-NTU method.