Is it possible to complete the whole preheat-evaporate-supheat process inside just one HEX? 1

[Carol3377]

Dear all,
In my design, water will be heated by hot gas from liquid state to gas state to get high temperature steam.
In order to meet the compact requirement, is there a possibility to complete the whole heating process inside just on shell and tube HEX?
I am using the Bell-Delaware method to design the HEX, and am now confusing about how to deal with the calculation.
The water side will experience three phases and could you please give me some advice about how to conduct the calculation, or show me some reference materials on this issue?

Thank you.

 

[itdepends]

You can go from cold water to steam in one HX, but you won't get superheat until you add additional heat to the steam in the absence of liquid water. i.e. downstream in a superheater.

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[Carol3377]

Thank you for your response. The heat is enough to heat the water from liquid state to dry gas state. But I am not sure about the design process of the HX.

 

[georgeverghese]

It should be possible to do this in a vertically oriented HX to get this water superheated. But entrained moisture content may or may not be acceptable due to the high velocity of the superheated steam. If not acceptable, an external KOD downstream of the HX could be used to remove unwanted moisture (water droplets).

Typical design process is to split up the HX into these 3 heating duties and evaluat ethe HX area to derive the surface area required for each - a few iterations of the entire thermal calc may be required to match estimated terminal temps in each zone to match the calculated - quite involved if you do this manually.

 

[moltenmetal]

Yes, it's possible to get superheated steam out of a water-fed boiler. And yes, it takes a hell of a lot more area to do so than you'd expect. What you're fighting here is not really a shortage of heat exchanger (metal) area, but the time required for heat transfer between droplets of entrained water and superheated steam exiting the exchanger. If you don't have enough time for that heat transfer to occur, the water droplets can persist. Depending on the intended use of that superheated steam, those droplets can be a problem, or not. A downstream separator will remove some of the entrained water and will evaporate some of it (dropping the degree of superheat), but there has to be somewhere for the rest of the water to go, i.e. up the pressure gradient, if you want to re-evaporate it.

 

[BronYrAur]

I don't claim to be an expert on this, but things that I think need to be addressed are as follows:

Your "feedwater" to this "boiler" will have to be regulated and at proper pressure when entering. Not sure what pressures are involved here, but you may need a high-head pump and valve, similar to the feedwater valve on a high-pressure boiler. Now, how to control the pump??? On/off, continuous with VFD or small bypass??

Will this water need to be pre-heated/deaerated and chemically treated?

Low water cutoffs, similar to a boiler?? What happens if the unit is dry with only the hot gas and then you introduce water? Will there be thermal shock or violent flash?

What regulates the hot gas into the heat exchanger? I'm sure it doesn't run wild.

As you mentioned, you are in the process of sizing the unit. Sounds like this hot gas will not condense during this process, so you are limited to a sensible temperature change on the hot gas side, which may require a lot more heat transfer area that you think.

I'm sure I am missing some other safety items (like relief valves and other controls) but this is all I can come up with on the spot.

 

[davefitz]

Such a device is called a "steam generator" and there are several vendors that supply such heat exchangers. Most are "once-through" devices without steam drums . If you have available a lot of pump pressure, then the simplest design has a common inlet header, a common outlet header, and inlet orifices or capillary tubes to provide good tube to tube flow distribution ( refer to ledinegg static stability and thermal hydraulic sensitivity) and avoids dynamic instability, both of which need to be explicitly examined in the design . Metallurgically one would need to avoid use of austenitic stainless steel, to avoid the chloride stress corrosion issue, so the inlet part may be ferritic and the outlet section can be either Inconel , wolfram-modified ferritic, or duplex SS tubing, depending on the expected max outlet temperature . The stability characteristic of the system of multiple tubes can be greatly improved if (a) constant upflow heating plus (b) increase the diameter of the tubes as the specific volume of the fluid increases.

If you do not have a lot of available pump pressure, then the device would likely need to be subdivided into at least 2 systems, with at least one intervening mix header. The likely location of the mix header would be at the saturated steam location, but designed to permit occasional 2-phase flow. One such header design is the water fed from the upstream circuit feeds the bottom centerline of the small intermediate header, and the header discharges to the second section through tubes along the side centerline.

"Nobody expects the Spanish Inquisition!"

 

[chicopee]

Davefitz mentioning "once-through " devices should give you an idea for your design. Another design that may give you so more ideas for your design is a type of superheater that fits under the belly of older versions type of firetube boilers exposing the superheater to the burner flame within the refractory chamber. Also study superheaters of watertube boilers and as mentioned in the above posts is that you need more heat to get superheated steam.