Flash steam in the condensate return header

[maceng11]

Hello all,

I will start my thread by describing the problem at hand:

An industrial plant is divided into 4 units. A typical steam condensate collection system is located in each unit is comprised of of a collection tank, heat exchanger to cool the condensate, redundant pumps and the I&C control system to turn on/off the pumps. The systems are located at the lowest elevation of the plant (elevation 0.0ft) and the condensate is lifted (pumped) at a pressure of 72psig @195F through 2.5in pipe (very long run) to a higher elevation from each unit to the 6.0in horizontal condensate return header which is located at 45.0ft elevation. The condensate reaches the header at 155F and unknown pressure. This header then collects the condensate from all units to be discharged to a sump channel which has a pressure slightly below atmospheric. The condensate return header is plagued with the problem of steam hammer which is often severe enough to break hangers an supports.

I have examined the problem and believe that the vapor pressure of condensate reaching the header is still sufficiently high to promote local steam flashing which then collapses as it condenses due to heat loss and we get the steam hammer effect along straight runs, bends and elbows (no need to detail the phenomenon as I think it is quite common). What I am proposing as an engineering solution is to install a pressure control station (e.g. pressure control valve PCV)to impose a slight back pressure on the header to make sure it flooded with condensate all the time and keep the its pressure sufficiently higher that the vapor pressure of the condensate. It goes without saying that this PCV must be installed as close as possible to the final discharge point; otherwise, the condensate will have enough time and distance to partially flash back into steam.

I would really appreciate your feedback, comments and any documented evidence of same or similar solutions.


Many thanks,

Maceng

 

[ione]

How can condensate in a sub-saturated state (195 F @ 72 barg) flash into steam?

 

[katmar]

The only way I can see that you might get flash steam in the 6" header is if the sump channel is at a significantly lower height than the header and the discharge from the header down to the channel is causing a syphon and reducing the pressure in the header to below the saturation pressure at 155[°]F i.e. below 4.2 psia.

This would be easy to check by installing a pressure gauge to see what is happening. If this is the cause it could be fixed by simply installing an open rising vent from the header so that air can be drawn in if the pressure goes below atmospheric.

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[maceng11]

It is true that the condensate is sub-saturated (195 F @ 72 barg) once it leaves the pump. But if you account for the static and dynamic head losses due to pumping it through a 380ft of 2.5in pipe which changes its elevation by 45ft to reach a 6.0in long header which essentially has no pressure control and is non-flooded, the header acts as a flash vessel causing some condensate to initially flash but also has enough to form vapor pockets as it travels through the long header (1200ft).

katmar, this is exactly what I suspect is happening. I am not at liberty to discuss the detailed design of the plant, but it is a fact that the elevation of the sump channel is 26ft lower than that of the header and is designed to have a constant steam of water flow which may create a suction (negative) pressure. However, considering the length of the header, I dont see how a vent line can help the situation especially when we know that the condensate can locally form equilibrium vapor anywhere inside the long header?!

Your help and discussion are highly appreciated.

Maceng

 

[katmar]

The condensate can only form equilibrium vapor in the header at a point where the temperature and pressure are in equilibrium. And the only place where you can generate the sub-atmospheric pressure necessary to be in equilibrium with 155[°]F is where the drain from the header down to the sump channel is connected to the header. If you place a vent at this point, or else increase the diameter of the drain line so that it can never run full and generate a syphon, then you can never have a sub-atmospheric pressure. Since your temperature is quite safely below the atmospheric boiling point of 212[°]F there is no other way to generate flash steam.

However, having said all this, I must say that I have never seen something like this happen. It is much more likely that you have a steam trap or control valve passing steam somewhere and injecting live steam into the header.

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[hacksaw]

afraid it is a common problem

while you are returning 155F condensate to a header, you make no mention of the other feed temps. As mentions in above posts you may have steam being dumped in the system, due to a malfunction trap.

a partially floded CR line will hammer well enough, as long as you form slugs in the system get a temperature reading on the return line

don't recommend a pcv in a return line that will create other problems, but you do have the maintain pressure in the return drum with a suitable vent and silencer

 

[maceng11]

I was recently informed that the temperature of the returned condensate is actually 200F. However, since the long return piping and header are uninsulated, the wall temperature drops to 155F in certain locations when the no condensate is flowing for some time.

Also, I am certain that there is no direct steam injection into condensate piping or header.

Hacksaw:
What sort of problems can be created by installing a PCV? please elaborate.

The only purpose of this pcv is to exert a small back pressure on the header to maintain it flooded with single phase liquid. This could also be accomplished by raising the elevation of a portion of the header just before the discharge point to maintain it in flooded condition?

 

[MJCronin]

My two cents...

Install a vented condensate pumping unit such as those mafe by Spirax Sarco or ARMSTRONG) halfway along the 380 ft header.

This would be require low pressure steam for motive power.

http://www.spiraxsarco.com/resource.../pumping-condensate-from-vented-receivers.asp

The final reciever should be a vented flash tank (not a "trough")

Penn Seperator makes products just for this purpose.

http://www.pennseparator.com/prod.htm

Ensure that the final pressure drop occurs at the face of the new vented flash tank.