design of a counter-flow tube-in-tube supercritical steam heat exchanger 3

[xlr8shun]

Hello, I am trying to design a tube-in-tube heat exchanger that is capable of cooling a 200 bar supercritical steam line from 500 deg C to 40 C, using building utility water as the cooling fluid in the outer tube. I would like to design the exchanger so that the cooling fluid outlet temperature does not exceed 40 C. The first concern I have is that with the hot fluid inlet temperature exceeding the boiling point of the building water, flashing/boiling could occur making the heat transfer performance of the heat exchanger difficult to predict.

I have come across closed-form heat exchanger analysis such as found in the link below, however those solutions rely on the assumption of a constant specific heat value, and from what I can tell would not take into account the heat exchanged due to phase change in bringing the hot fluid from 500 to 40 C.

https://web.mit.edu/16.unified/

http://www/FALL/thermodynamics/notes/node131.html

I am hoping someone can point me in the right direction in terms of a resource or approach that would be applicable to this specific scenario, either as a closed-form solution or suggestion for an iterative approach using changes in thermodynamics properties (e.g. enthalpy)

 

[EdStainless]

Is this constant flow and temperature?
Start by just doing an energy balance, how much water will this take?
If the steam flow is much it might turn out to an impossibly large amount.
There are other ways to do this also.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[xlr8shun]

Yes, the flow and inlet temperature is constant. I have done an energy balance based on the enthalpy differences between start and end states calculated from REFPROP. From an energy balance standpoint approximately 5 gpm of cooling water will balance the change in enthalpy of the hot stream from 500 to 40 C with a cooling water temperature rise of only 15 C, this is approximately 20 kW of heat absorbed by the cold stream. What I'm looking for now is guidance on the actual heat exchanger design in terms of tube sizes etc. to ensure sufficient cooling of the exchanger surfaces.

 

[EdStainless]

Why not go direct injection? More compact and easier to maintain.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[xlr8shun]

Direct injection could work but unfortunately I need to keep the hot and cold streams separate, plus I suspect that I might have a hard time finding an affordable direct injection controller solution in a reasonable time period. This is for an R&D application, ideally built with off the shelf components that can handle the temperature and pressure (i.e. small tubes and high pressure fittings)

 

[davefitz]

First of all, the use of building water without purification will lead to a severe buildup of deposits on the heat exchanger surface. Demineralized water should be used , and consideration should be to controlling its oxygen content ( I recall that 25 ppb O2 is recommended for the combined oxygenated treatment method).

Secondly, the boiling heat transfer coefficient can be obtained from the correlations published by Siemens KWU of Erlangen, Germany in the 1990's. One can either improve the heat transfer coeficient by using rifled tubing, or one can reduce the heat transfer coeeficient needed by deliberately applying a thermally resistive layer to the metallic interface between the 2 fluids.

Finally, the HX you are designing is similar to those already commericailly available for steam purity monitoring systems. It may be less costly to just buy one.



"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[EdStainless]

There are lots of DI units available and some have integral controls.
Spirax, Mansoneilan, PICK, Spence, Komax, and the list goes on.

The other option might be a small S&T HX for just the de-superheat, and then use tube and tube for the balance of the HT.
The real problem with Tube in tube is that they have poor heat transfer. There is no turbulance because there is no cross flow.
These get very long. I have built pasteurizers for food processing and you end up with a rack 25' long, 4 tubes deep and 8 tubes tall.
That is over 600' feet of tubing.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed