Multi-phase condensate/steam line sizing 1

[buzzie]

This is an issue for a plant expansion I am working.
Have 75 psig condensate (at its bubble point) flowing from a flash tank (without a pump) to a distant tank #2 (several hundred meters) where it will flash to 3 psig or less. Just before the tank #2 is a control valve to maintain level in tank #2.
I am trying to understand the dynamics of the flow between the two tanks to permit the line to be sized properly both upstream and downstream of the control valve.
First, what will occur upstream of the valve as the condensate flows from the first tank and line loss occurs? Currently plant issues will not permit the valve to be located near the first tank. Can a multiphase condition be accurately predicted?
Second, what is a reasonable criteria for sizing a multi-phase condensate/steam line downstream of the control valve. This is a design problem I have not run into previously. Thanks.

 

[rmw]

First off, you don't want the valve near the first tank, you want it as near the second tank as possible.

If you size your piping from the first tank to the second tank so that there is minimum pressure drop, then the temperature loss in the piping will help keep the condensate from flashing before it gets to the control valve.

If your first tank has any level to it at all, the height of the fluid above the drain will keep the condensate in the pipe above its saturation point. Remember that the fluid is saturated at its surface in the first tank, and as you go deeper in the tank the fluid is not saturated.

I would use the height of the level in the first tank above the drain point as the pressure drop limitation of the piping.

Optimally you will put the level control valve right at the second tank so that there is very little if any piping handling two phase flow.

I think I remember that

http://www.processassociates.com

has some sizing information for two phase flow in their "tools" section.

rmw

 

[katmar]

Excellent advice by rmw. It may sound expensive to design the pipe from the first tank to the valve for low pressure drop, but this is actually the economic way to do it (especially with non-pumped flow).

If you assume high pressure drop then you will have flashing and the pipe has to be larger diameter to cope with vapor so you are back to the same thing. Design it for low dP from the start and be sure you have liquid flow all the way to the valve.

Valve suppliers will be able to size and spec the valve for you for flashing flow - which you will get across the valve.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[25362]

I wouldn't like to sound pessimistic, but it seems to me that some degree of flashing should be expected in view of:

1. the line may be thermally insulated, if only for personnel protection, minimizing heat losses;
2. the line may contain vertical sections inducing a local lowering of pressure, which under certain conditions may induce hammering;
3. level control of the downstream receiver may result in unsteady flow rates with brief ebbs and surges.

Considering these possible scenarios the CV should be sized, and its trim, and downstream pipe, selected accordingly.

 

[djack77494]

Ideally, you would want to have a bit of static head with the source tank at a higher elevation than the destination tank. That would give you enough pressure to cover the hydraulic loss (which is hopefully very small). You really don't want to get into sizing the line AND control valve for two phases entering if there's any way to avoid it. Doug

 

[lastone]

Isn't there some use you can find for this energy rather than just throwing it away?

At today's gas prices, for every 10,000 LB/HR condensate that you are letting drop from 75 to 3 psig, you are sending $5/hr out into the atmosphere.

Enthalpy of 75psig sat liquid = 291 Btu/lb
Enthalpy of 3psig sat liquid = 190 Btu/lb

100 Btu/lb * $5/MMBTU * 10MLB/HR = $5/hr

If your condensate flowrate is 50 MLB/HR, you are giving away $220,000 a year!

Wouldn't this pay for a pump and an insulated line to return this heat to a boiler?

 

[LaSalle1940]

I'd suggest looking at

http://www.spiraxsarco.com

for on-line sizing tools; the situation described originally approximates flow of condensate to a trap (the valve at the inlet of the low-pressure tank is roughly analogous to the trap).

One thing with which I'd agree: you want to minimize the amount of in-line flashing. That phenomenon results in a larger line size, and sizing of the control valve at the inlet of the low-pressure tank becomes a bit of a challenge since now you're dealing with sizing a control valve to handle two phases.

If it's absolutely impossible to avoid dealing with in-line flashing flow, perhaps Leung's articles on the subject in Chemical Engineering Progress would help: granted, those are intended originally for flashing flow in relief devices and in tailpipes after relief devices, but the principles are still pretty much the same. Seems to me the articles in question were published in 1992 and 1996 (don't have my back issues of CEP handy so I'm doing this from memory).