2 replies to this topic

[mvancleave]

We have a once-through cooling system for a surface condenser. After leaving the pumps, the cooling water piping is routed underground to the condenser, which is located above ground (see attached sketch – not as good as one of Art’s, but it will have to suffice). Following the condenser, the piping is again routed underground to the outfall. Due to the elevations involved, the outlet of the condenser will be operating at a vacuum. Under the worst-case operating scenario, the pressure could be as low as 1.35 psia.

Under a transient condition, such as a pump trip, there is a significant potential for column separation and subsequent severe water hammer as the vapor bubble collapses. We have looked into a number of options to break vacuum under this situation to prevent water hammer and have decided to install a U-tube vent. The strategy is that, under all “normal” operating conditions, the water will remain in the U-tube and act as a loop seal. At or just after column separation, the pressure in the main circ water piping will be low enough to lift the water out of the loop seal and admit air into the main piping, breaking the vacuum.

I have been asked to verify the design. I am comfortable with the concept in general, but I am concerned about the U-tube depth. It needs to be deep enough to maintain a seal under normal operation, yet shallow enough to ensure that, when there is column separation, the seal will be broken.

The problem is in deciding what temperatures (vapor pressure) and densities to use and whether to measure to the top or bottom of the “U.” The cooling water should be less than 60 F all the time. The power house (ambient) could be over 100 F for a few hours a day on some of the hottest days of the year.

My thinking is as follows:

The top of the pipe at the U must be low enough to maintain a seal. For this condition, use the lightest water (density at 100 F is 61.966 lb/ft^3) to determine the column height that would be lifted considering ambient pressure (14.55 psia for our location in Toronto) on one side and 1.35 psia on the other. For this I get 366 inches. The distance from the centerline of the main pipe to the top of the U must be more than 366 inches.

The bottom of the U must not be so deep that the column above it cannot be lifted when the pressure drops below the trigger level in the main pipe. For this condition, I would use the heaviest water expected (density at 60 F should be ok, so use 62.364 lb/ft^3), but assume that temperature at the top of the column is 100 F (use vapor pressure of 0.95 psia for water). I get a maximum column depth of 377 inches (14.55 psia – 0.95 psia pressure difference and 62.364 lb/ft^3 density). The distance from the centerline of the main pipe to the bottom of the pipe cannot be any greater than 377 inches (making the current depth of 387 inchess too deep). The designing engineer asserts that we can use the top of the pipe at the U for this condition since “that’s where the air will come through first.”

Before I push back too hard on the depth of this thing, I wanted to see if anyone here has any comments or advice.

I hope this lengthy diatribe makes some sense.

Thanks for any help,

Mike

Attached Files

•  U_tube.xls   22.5KB   66 downloads

[Chem01]

Hi,
What is your cooling water supply source?
Instead of U tube why not use a simple vertical pipe at condenser outlet cooling water pipe? Its height can be evaluated and it will give the same result you are expecting from U tube.
Even in case where cooling water returns to cooling tower, often vertical vents are provided to accomodate similar situation.

[mvancleave]

Thanks for the response!

Source of the cooling water is Lake Ontario.

A standpipe would be a great solution if the pumps had enough head. The siphonic effect was taken into account when the pumps were sized and purchased. Venting the condenser outlet would essentially be adding 30 or so feet of head to the system, and we just don't have it. If this had been recoginized earlier, the pumps would have been sized differently and we would be in a better situation now.