Loud Sporadic Noise in Steam Condensate Discharge Line 1

[egscot]

I found a Swagelok article in the URL below to be very informative (see attachment). We are having loud sporadic noises generating in our steam condensate pump discharge lines. I think it is probably steam bubbles imploding in the line, but I can't understand how the bubbles initially develop, if that is what is happenign. Can anyone explain to me the dynamics of what may be going on in our 2” condensate return line and why the article’s insistence of having the line tie in at the top of the condensate return header to prevent the water hammer problem vs tieing in on the end, side, or bottom of the return header? How does tieing in on the top of the header change the dynamics of what is going on inside the line?

http://www.swagelok.com/downloads/webcatalogs/EN/Best-Practices-14.pdf

 

[TBP]

Go to

http://www.kirsner.org

for some excellent articles on water hammer in steam systems. Look at "Water Hammer In Boiler Plants" in particular. Spirax Sarco also has some very informative publications.

I only skimmed the Swagelok article, but one thing struck as out and out incorrect. They write: "...the threaded connections do not have the ability to withstand the expansion and contraction of the steam/condensate
system, therefore leaks will occur." This is simply not accurate. I have personally worked on steam & condensate systems that are WWII vintage and older that are filled with screwed connections, and they still work just fine.

 

[JoeySoap]

I think the point is, there are threaded connections and then there's threaded connections. They're not all born equal.

The take off from the top of the condensate line will be to ensure that incondensibles are removed from the system. A side or bottom take off will allow incondensibles to build up, they have to be vented.

 

[TBP]

Steam traps are designed to discharge the air in the system on start up. Air is more dense than steam, and goes down. However, egscot's line is a pump discharge. Any non-condensibles discharged by the traps will have gone out the vent from the condensate tank. Without seeing the system in question, the first thing I'd look at would be the check valve after the pump.

With systems like steam, condensate, compressed air, etc - connections off the bottoms of mains tend to act as collection points for any debris in the line. That's why dirt pockets are required for drip legs on steam lines.

With respect to threaded connections, the two main problems I encounter are improper threading in the field (cut far too deep), and using the wrong pipe dope for the application. The instructions on the container provide the info regarding max pipe size, temperature limitations, etc.

 

[katmar]

The most likely cause of noises like that is a steam trap discharging into the condensate line after the pump. Any flash steam from the trap will rapidly collapse when it hits the cold pumped condensate.

If there isn't any trap discharging into the line after the pump, the only way you could generate vapours would be to pump the condensate to a high point where its pressure is below the saturation point. If the pipeline then drops to a lower level where the pressure increases the vapour would then collapse. Of course, if the condensate tank from which you are pumping is at atmospheric pressure then this cannot happen.

It is generally good practice for connections into a non-flooded line to be at the top of the main to prevent backflow. Also, if there is a risk of the entering stream flashing it is better to enter the top of the pipe. But if all your condensate is subcooled then it is no different from pumping water and general water piping layout considerations would apply.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[JoeySoap]

A side issue, but not all types of steam traps pass incondensibles well at all. I have seen applications where I have had to install air vents in addition to the (float type) steam traps.

 

[TBP]

JoeySoap - I've never encountered a situation like you describe, but I know they exist. Have you tried Hoffman 'H' style F&T's? They've got 4 connections - 2 inlet and 2 outlet with one each, plugged. Instead of leaving one inlet plugged, put the extra air vent there.

 

[JoeySoap]

Thanks TPB. No, I haven't tried those but I will take a look next time I need such a device, thank you.