12 replies to this topic

[masino]

Dear all

 

I would like to ask you about relationship between temperature approach and MTD in heat exchanger design.

 

In energy saving project, one of heat exchanger have these specification below

----------------------------------------------------------------------------------------------------

Tube (Crude Styrene) 36.3-->77.0C                4 passes

Shell (Process Condensate) 82.0-->67.2C       2 passes

TEMA:BFM

 

From HTRI, MTD corrected is 11.2C

Function: Benzene/Toluene Feed Preheater

 

-----------------------------------------------------------------------------------------------------

 

simply temperature approach is cold outlet minus hot inlet (82.0-77.0)

in these temperature profile, there would be a crossing point.

 

I want to know how MTD will affect heat exchanger efficiency better?

 

If MTD is decreasing in terms of revamping or energy saving, heat exchanger is designed better?

 

MTD is closely related to temperature approach?

 

One more thing to ask!!

 

What is the advantage of BFM over BEM?

I usually work with BEM type. So it is not familiar

 

Thanks in advance

 

Yours

Edited by masino, 20 August 2013 - 01:00 AM.

[PingPong]

What is the advantage of BFM over BEM?

BEM has only one shellpass per shell. With two (or more) tubepasses per shell this means that the fluid in the tubes is partly in cocurrent flow with the fluid on the shell side, and only partly in countercurrent flow. Consequently an LMTD correction factor F is to be applied for calculating the corrected MTD.

 

 

BFM has a longitudinal baffle and therefor two shellpasses per shell. It is sometimes used to achieve fully countercurrent flow in each shell so that F = 1.0

At least in theory. Problem is however that those long baffles may leak at the seal with the shell, aspecially after being used for a while. In my 30 years of process design experience I have never applied them, and have never seen others use them either. Make sure that it is really the intention to use BFM for this service.

 

 

From HTRI, MTD corrected is 11.2C

That seems not correct for BFM, I would expect it to be 14.2 ^oC with F = 1.0

 

Even for BEM it seems too low, as I would expect 12.1 ^oC with F = 0.85

 

 

in these temperature profile, there would be a crossing point.

A temperature cross of this size is not a problem if two BEM shells in series are used. If a BFM type (with two tubepasses per shell) is used one can even operate with a temperature cross in one shell as due to the longitudinal baffle it is operating in full countercurrent mode.

Edited by PingPong, 20 August 2013 - 06:01 AM.

[masino]

To. Pingpong

 

I really appreciate it for your advice.

 

This is BFM type.

 

Dia is 680mm and length is 6096mm.

 

As F factor is increasing, heat exchanger performance is considered to be better?

 

That's why BFM has an advantage

 

MTD corrected from HTRI is 11.2C (I read it from HX data sheet)

 

Data sheet is based upon HTRI result.

 

How can I explain the lower MTD corrected than expected?

 

I do not think HTRI input has an error.

 

 

[PingPong]

Forget HTRI for a moment and take a simple calculator.

Tube (Crude Styrene) 36.3-->77.0C               

Shell (Process Condensate) 82.0-->67.2C     

Now calculate by hand for these temperatures what the LMTD is for countercurrent flow.

I get 14.2 ^oC, what do you get?

 

For countercurrent flow, as is the case in a BFM type, the LMTD correction factor F = 1.0 assuming the longitudinal baffle does not leak. Then CMTD = F * LMTD = 1.0 * 14.2 = 14.2 ^oC for a BFM type.

 

Maybe HTRI assumes a lot of baffle leakage for BFM type and therefor uses a very low F of only 0.79 to obtain a CMTD of only 11.2 ^oC.

In that case however I see no point in using BFM instead of BEM type.

 

You could run HTRI to design the exchanger as a BEM type and see whether that results in a bigger or smaller exchanger than BFM.

[srfish]

Since the exchanger has 4 tube passes it is not a counter-flow type. It needs a LMTD correction factor. If you do not have access to correction factor charts  let me know and I will refer you to some texts that have it.  You will either need a single BFM type or 2 BEM types in series.

[PingPong]

You could be right.

 

 

 

Tube (Crude Styrene) 36.3-->77.0C                4 passes

Shell (Process Condensate) 82.0-->67.2C       2 passes

 

I had interpreted this exchanger as having 2 shells with 2 tubepasses each, so 4 tubepasses total in 2 shells in series.

 

[masino]

To Pingpong & srfish

 

I appreciate for your comment and advices

 

 

I want to make it clear.

 

Why do I need to consider BEM in terms of leakage problem?

 

Yours

[PingPong]

Masimo, you do not need to consider BEM.

I was under the (probably wrong) impression that your BFM has two shells, in which case there would have been no advantage to use BFM instead of BEM. If however it is only 1 shell with 4 tubepasses then, as srfish correctly pointed out, the LMTD correction factor F will be much less than 1.0 , because then it would not be a fully countercurrent design.

 

Please clarify:

1) is this an existing exchanger with an existing tubebundle, or an existing shell with a new tubebundle, or an entirely new exchanger?

2) does the BFM exchanger that you described above have 1 or 2 shells?

[masino]

this is the totally new heat exchanger for energy saving project

 

Yours

[PingPong]

Then, in theory, you can handle this service with one BFM shell. Because of partly cocurrent flow (due to more than 2 tubepasses), plus thermal and physical leakage over the longitudinal baffle, you get that combined LMTD correction factor of 0.79 (= 11.2 / 14.2). This can be considered as the product of F = 0.85 and Delta = 0.93

Normally baffle leakage correction is called Delta factor (at least for BEM), and I expect you can find that back somewhere in the HTRI output for BFM.

 

Regarding efficiency of the exchanger: depends on how you want to define efficiency.

I would say it is: (77 - 36.3) / (82 - 36.3) = 0.89 = 89 %

because purpose of exchanger is to heat up crude styrene from 36.3 to .... ^oC (as high as feasible). Als the heating medium is 82 ^oC the crude styrene can obviously never be heated more than that.

 

All the above and previous posts are assuming that there is no vaporization of crude styrene taking place in the exchanger beyond a certain temperature. Otherwise it becomes more complicated to determine F, Delta, CMTD, and efficiency.

 

I do not know how extensive your energy saving project is. Unless it is only adding one new exchanger (this one), one would normally use Pinch Technology or Exergy Analysis to analyse the heat integration of the whole plant.

 

Note that your exchanger has a pinch (minimum temperature difference) of only 5 ^oC (being 82 - 77 ^oC) which is very low, indicating that this is probably a bit overdone from an economical point of view.

Edited by PingPong, 25 August 2013 - 04:55 AM.

[masino]

Dear Pingpong

 

In terms of energy saving projects, rearrangement of heating and cooling network have been performed with addition of

new heat exchangers (including this question's one).

 

I have a lesson that I should look closely into the result sections in HTRI.

I am afraid I lost some of crucial points (for sure)

 

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Note that your exchanger has a pinch (minimum temperature difference) of only 5 ^oC (being 82 - 77 ^oC) which is very low, indicating that this is probably a bit overdone from an economical point of view.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

 

I want to ask you about your reply above.

For the economic analysis of heat exchangers performance, do we need to consider only about a pince, which I have thought that it is the same meaning with temperature approach.

In HTRI result section, MTD corrected is shown to indicate whether the designed heat exchanger work well or not.

 

In order to enhance the performance, do I need to increase a pinch in this case?

 

Yours

[PingPong]

When a Pinch Analysis on plant's heat integration is performed, the outcome is usually an economical optimum minimum temperature difference between the Hot and Cold Composite Curves.

The lower the minimum temperature difference, the lower the hot and cold utility requirement, but the higher the total required exchanger areas to be installed. That is the economical trade-off, and at some minimum temperature difference is the economical optimum.

 

It would surprise me that the optimum minimum temperature difference in this plant would be as low as 5 ^oC, that is all I meant. But if energy is very expensive or scarse in your plant then it can be justified. I expect that decission has already been taken by others, so you have to proceed with this. I merely tried to increase your insight for future projects.

[masino]

I really appreicate for your clear explanation.