Operating temp/pressure Vs design temp/pressure for shell & tube HE & bellows requirement

[chemks2012]

Hello all,

Just a quick query please.
How do we decide the design temperature and design pressure for shell side and tube side of a heat exchanger condensers?

I have following conditions

Shell side: water vapour with nitrogen at 3barg and max at 180degC
Tube side: Cooling water at 4.5barg and at max at 30degC [and tubes will see steam temp of 100degC sometime]

At client’s existing site, they have metal bellows on some of the condenser shells [having similar design] which I believe to compensate thermal expansion due to high temperature of 190degC. Is there any guideline on when metal bellows are required please? I understand that there is separate CE marking time consuming procedure for certifying such metal bellows and we are trying to avoid having bellows but I can’t justify due to lack of information.

Any help would be appreciated.
Regards,
KS

 

[sshep]

Hey KC,

The stress in a fixed tubesheet design is caused by differential expansion between the shell & tubes at least 3 conditions: fabrication, normal operation (clean & fouled), and steam out. The supplier's design software can optimize trade-offs with thicker tubes and strength welded tubesheet connections, but at some point these aren't enough. An expansion joint may be recommended, or the owner can also consider an alternative bundle design (u-tube, floating head, fin-fan).

We did stress problems for a BEM type exchanger in school by hand (1982), I almost can't believe there isn't an easy tool online. Such a tool could be used for screening when licensed software isn't available. In any case the bellows recommendation (and design) will come from the exchanger supplier's designer and his software. It has to be taken seriously. If tube stresses are too high, tube failures will result. The supplier will specify and order the bellows so that the size and number of the convolutions will match the service. The material certifications for the bellows will be part of their deliverable, you should require it. Your company obviously allows a bellows, and I have never found it to be a big deal except that it can add to the cost.

How big of an exchanger is it?

best wishes,
sshep

 

[chemks2012]

Hi sshep,

Thanks for your input.
Yes, all make sense.

The physical dimensions are as below
Shell dia: 600mm (single pass)
Tube length: 1000mm (single pass)
Number of tubes: approx 130 in a triangular pitch

How would you decide design temperature and pressure based on operating parameters I stated?

Also I have another query on effect of noncondensable gases on overall heat transfer coefficient (OHTC). We have vapor flow rate of about 200kg/hr and noncondensable gas (Nitrogen) of 50kg/hr mixed with vapor.

The OHTC for pure steam and cooling watee system is about 1000 to 1500 W/m2 C. I read that even 0.5 to 2% by weight of noncondensable gases could reduce the OHTC by 50%.

As you can see, we got 20% of noncondensable gases but I am not sure how will this affect OHTC. Do you think it's reasonable to assume reduction in OHTC of upto 80%?

Thanks in advance.
Regards,
KS

 

[GHartmann]

First, I can't imagine a bellows would be required for an exchanger this small. But leave that up to the detail mechanical design.

Second: You would want to set the design pressure for the tubes above the shut-off pressure of your cooling water pumps. For the shell side you need to have some margin above your maximum operating pressure. For your exchanger I would suggest something like 6.9 barg (or a nominal 7.0 bar-g). For simplicity set both sides at the same design pressure. (This cannot always be the case, but works for your example)

For the design temperature a similar analysis must be made. For your exchanger something like a tubeside design of 150 deg C ; shellside design 200 deg C. However, if the shellside gas is essentially steam (100% water vapor) the design temperature should be above the saturation temperature of steam at the shellside design pressure (800 bar-a / 170 deg C).

The above practice is just an example of the thought process. Do you have a mechanical engineer in your work area you can consult?

Third: The heat transfer coefficient for condensation is much, much greater than the coefficient for gas cooling. Thus the high impact on the overall heat transfer coefficient when non-condensables are present. I cannot comment on the % decrease. I would recommend some type of heat exchanger simulation software to generate your answer.

Good luck.

G.Hartmann
BS ChE NC State University

 

[chemks2012]

GHartmann

Thanks very much for your input, much appreciated.
Very useful.

Regards,
KS