Expansion Joint - BEM type Heat exchanger

[kvbalu2011]

Dear all

I have to do a preliminary mech. design of a BEM(TEMA) type heatexchanger (Vertical - Fixed tube sheets). The exchanger shall be on Lugs on RCC framed structure. Would like to know the criteria for requirement of Expansion joints for the design conditions. Is an Expansion joint required on the shell. Design parameters are as below:

Shell MOC - ASTM A517 Gr 70
Tube MOC - SS316
Shell side
Design Temperature(deg.C) - 265 (In) & 194(Out)
Design Pressure (kg/sqcm) - 13.00

Tube Side
Design Temperature(deg.C) - 133 (In) & 149(Out)
Design Pressure (kg/sqcm) - 11.00

Maximum Axial Differential Thermal Expansion between Shell and Tubes-Tube has higher Thermal growth (Exp. Jt. Extension) : +0.179 mm

Kindly give your experienced opinion on whether expansion joints are required ?
Thanking you all in anticipation

 

[SnTMan]

kvbalu2011, difficult to say without running complete Part UHX calculations, but, based on the diff expansion alone, I'd say it is unlikely to need an expansion joint for the given conditions.

Still, it should be confirmed per Part UHX, or other applicable Code, for all operating cases to be encountered.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[SnTMan]

By "difficult" what I actually mean is impossible

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[marty007]

Just to throw in another detail, be certain that you have considered the worst-case situation (normally loss of shell-side flow). Unless there is some sort of control to eliminate this case, loss of shell-side flow can cause significant thermal differentials. These cases must now be clearly spelled out in ASME Form U-5 as of the 2013 edition of ASME VIII-1.

Cheers,

 

[georgeverghese]

These "design temps" are actually operating temps; so what is the mechanical design temp for this HX ? - note that mechanical design temp should always be the same for both shellside and tubeside.

Setting the design temps the same for both sides may not eliminate the need for this expansion joint, if you think this is required for the thermally induced differential stresses between shell and tubeside.

Diff thermal stress would be highest at startup if one side were to be warmed up to max op temp while the other side is still at lowest ambient temp.

 

[mechanical2014]

Hi Kv,

As a thumb rule, if you are getting abnormal tube sheet thickness then you have to go for expansion joints.

 

[marty007]

note that mechanical design temp should always be the same for both shellside and tubeside.

This is not true for most heat exchangers. The tubeside and shellside can have very different design temperatures. Common elements however should be designed in accordance with the highest temperature to which it is exposed.

So for example on a typical BEM exchanger you could have:
[ul]
[li]Tubeside design temperature: X°F[/li]
[li]Shellside design temperature: Y°F[/li]
[li]Design temperature for tubeside nozzles = X°F[/li]
[li]Design temperature for shellside nozzles = Y°F[/li]
[li]Design temperature for tubesheets/tubes = max(X°F, Y°F)[/li]
[/ul]

 

[SnTMan]

True, however calculations for fixed tubesheet exchangers, where differential expansion is to be considered, are generally carried out as follows:

Material strength properties are evaluated at the design temps, as per marty007, previous.

Differential thermal expansion is evaluated at operating temps, usually the mean metal temps (MMT) which may be a simple average of the inlet / outlet temps for each side, or perhaps something more "accurate" pulled from the rating data.

Strength properties may be evaluated at the MMT's at the designers' option.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[MrBTU]

kvbalu2011, you have not told us the tube length or the baffle spacing of this heat exchanger. The UHX calculation that SnTMan mentioned takes both of these into account to determine whether the heat exchanger can absorb the growth of the tubes, relative to the shell (+.179mm) without causing problems such as tubes pushing out of tube joint in tubesheet, or tubes buckling. Another piece of information required is tube to tubesheet joint detail; strength weld can handle more axial load than rolled only, for instance.

 

[kvbalu2011]

Many thanx for all your feedback,Members! I am trying to do a UHX design.

MrBTU,
Tube length = 4500mm, Single segmental baffles,26.13%cut, 450mm spacing, first baffle at 870 mm from inlet.
Tube-to-Tubesheet is expanded & strengthwelded.

Shell side medium - Steam
Tube side medium - Hydrocarbon (Kerosene)

Thanks

 

[georgeverghese]

In such an application with kero cold side exit temp at 149degC and steam exit at 194degC, the LMTD corr factor Ft would be = 1.0 even with 2 tubeside passes or multiples of 2. Hence dont see the reason for a TEMA M configuration here; it would be a lot more mechanically sound to go for TEMA AEU or BEU.

The process - thermal design engineer should be asked for reasoning for this current selection.

 

[kvbalu2011]

Hi georgeverghese

Thanx for your feedback. The process engineer would be definitely asked this question.