Reboiler with hydraulic loop to condensate drum liquid hammer problem 2

[renier]

I have come accross a reboiler design that I am not familiar with on one of our plants where we have a severe water hammer on the condensate side of the reboiler. This reboiler is installed horizontally with the column bottoms circulating on the shell side and the steam going through the tubes (2 passes). There is not a steam trap on the reboiler's condensate outlet but rather a hydraulic loop to maintain the condensate seal. The condensate line is experiencing severe water hammer (Some supports have broken). The condensate drum is under pressure control with steam and condensate injected on the vapor space to control the pressure.

I would like to hear from someone familiar with this type of reboiler w.r.t the rationale behind the design and what the correct piping configuration on the condensate piping should be (which I suspect to be source of the problem in our case i.e. vapor pockets). I am still in the process of getting hold of the original designer and hope that he will be able to solve the problem for us, but I am hoping that someone has experienced the exact problem and has actually solved it satisfactorily. I would also like to hear if there are potential interim measures I can take to solve the problem to allow us to run to the scheduled shutdown.

 

[TBP]

Has the inlet steam pressure increased, or the condensate system pressure increased? Have the broken hangers changed the slope of the condensate line, making the problem increasingly worse? Was there an orifice installed in the condensate piping that was removed accidently?

 

[renier]

As I understand from colleagues this is pretty much a start up problem. Although we had some damage to pipe supports the piping does not seem to have shifted too much.

 

[TBP]

The loop WILL work, under the right conditions. But I'm guessing that it's based on operating at very low steam pressures. Does this unit have it's condensate run by gravity directly to a vented receiver? If it doesn't, it likely should. When dealing with low steam pressures, the condensate piping is critical - there's not much pressure differential available to make anything flow. And with a loop seal, too much steam pressure is bad as well - it'll just blow the condensate out of the seal. And it'll only have to spike high for a few seconds to do it.

 

[pmureiko]

This sounds like a low pressure steam application since a condensate leg is sufficient to prevent steam from blowing through. I imagine other steam users are also operating without traps. Somewhere in your system, live steam is entering the condensate system and causing water hammer. This would tend to be more of a problem during startup when trying to establish the liquid seal legs. Once the unit is up and running, the seal legs will fill and it will run normally. You may want to make provisions to fill the legs with water before start up to minimize the steam blowthrough.

Like all things, it will work under certain conditions and poorly under another set. I would prefer direct function and use a steam trap on your reboiler to get better operation.

 

[renier]

The steam supply is actually at 8 bar with the pressure downstream of the flow control valve at around 3.3 bar. The condensate drum pressure is being controlled at around 2.9 bar. I suspected that steam was blowing through and adjusted the setpoint on the condensate drum pressure controller but to no avail. The shock wave repeats itself every ten minutes or so, regardless of the pressure in the condensate drum.

 

[chela]

Dear Renier,
Have you checked the shell side is having any problem of cycling? Your problem typically sounds like an unsteady state on the column side pressure or column bottom level. This cycle can be self perpetuating with the cycling on the steam flow. put all these controls on "manual" and try to steady out the unit. Best of Luck

 

[renier]

For your information on this problem. We discovered that the distillation tower bottoms temperature was running below design and that our propylene losses to bottoms was also higher than design (No focus on optimization here because of low propylene demand). After upward adjustment of the setpoint the problem seems to have disappeared.

I suspect that with the lower tower temp setting the steam valve had to throttle back excessively to lower the steam temperature which caused the pressure differential between the reboiler and condensate drum to become to too low, thus causing the condensate flow to become unstable (Could not drop the pressure on the condensate drum too far, due to cavitation constraints on the condensate pump).

 

[nakagawa]

I think the boilling of condensate in the LOOP cause the problem.
Suppose there is no condensate flow at the end of cycling,
the steam pressure in the reboiler is lowest at this condition and cannot push the condensate to the condensate drum. Remember that condensate temperature is lowest also.
The condensate produced must be accumulated in the reboiler, flooding and killing effective surface area for heat transfer with increase in pressure and tempeature of the condensate being produced. The steam pressure eventually reaches to a point to overcome the hydraulic restriction of the system and start discharging the condensate to the condensate drum. Initially the condensate flowing through the upward portion of your LOOP is colder than the condensate inside the reboiler and condensate being produced, that is, the condensate initially in the LOOP is subcooled condition. The temperature of the condensate passing through LOOP become hotter and hotter until it start boiling. Once start boilling, density of V/L mixture in the LOOP goes lower dramatically decreasing the hydraulic restriction dramatically allowing big condensate to pass through until substantially all of the condensate in the system is discharged. LOOP will be filled with steam mostly. Steam press in reboiler and LOOP is lowest at this condition and flow of condensate stops and REPEAT the process.
The water hammer may be caused by the flushing vapor through the LOOP contacting cold condensate carried from elsewhere. I,ve observed that the temperature differrence between steam and water exceeds by 60 degree C, hammering takes places in a bath directly introducing live steam without any fine distribution tricks. but I wonder if your system possibly have such high temp difference. I strongly hope someone talk more and more about water mammer in the condensate system. thank you.