A Two Phase Steam/Condensate Question 1

[Bambie]

The continuous boiler blow-off system at our plant discharges a constant mass flow of 0.2 kg/s at 6 psig into a lake via 250' of buried 6"nps pipe with a 12:1 drainage slope.
At the lake surface the 6"nps pipe branches into three 3"nps pipes that drop 14' vertically below the lake surface where they are anchored.
Direct contact condensation (water cannon) plagued this design and damaged anchors until a vent was installed in the 6"nps line which depressurized the 3"nps pipes and maintained the steam/water interface at lake surface elevation.
MIC degradation of the buried pipe has resulted in leaks, so rather than replace it, there is a proposal to re-route the 250' of 6"nps pipe 30' above ground and then drop it down into the existing 3"nps distribution lines.
A vent is proposed at the 30' elevation.
My concern is that a 0.2 kg/s steam/condensate mixture descending 30' under gravity and atmospheric pressure could pressurize the 3"nps distribution lines and 'load the water cannons' so to speak.
My question is whether there is any way to predict the static pressure in this downcomer and whether the 3"nps lines could be effectively vented to prevent water cannoning.
Please see the isometric sketch attached.

 

[chicopee]

For other steam qualities than the 1/2% the traverse time will lessen. For example x=1%,t=8.76min; x=5% t= 1.87min. All calc. based on .44lbm/sec(=.2kg/sec); the only change is the specific volume at different quality values.

 

[chicopee]

I looked at your worksheet and you are treating the condensate as regular water, under atmospheric pressure along the travel path, flowing down a pipe and I am not quiet sure that is the way to analyze condensate drainage for 250' of pipe. I used the continuity equation along the entire pipe under different steam quality value.

 

[Bambie]

Chicopee and LittleInch,

I think mass and energy conservation laws are helpfu but ignor two phase dynamics.

Nature abhors a vacuum - steam most of all.

I think the following two phase flow regime occurs inside the non-vented, dead-ended 6"nps pipe with a static steam pressure at 6 psig.

- In horizontal sections, 0.2 kg/s condensate at a depth of 0.3" flows down the 8% gradient at 2 fps (based on the Manning equations).

- Overhead, high quality steam roars along at 50 fps in response to the vacuum created by the imploding steam layer against the cool pipe wall, however only 0.3 psid is required to create this flow (which is confirmed by single phase analysis and applying the Martinelli-Nelson Friction Multiplier).

Static pressure reduction from steam friction losses would require a very large vent, which would generate unacceptable steam release during intermittent blow-off (50 kg/s for 15 minutes every 12 hours and very often during warmup), so its size will be restricted.

Assuming some positive pressure at the vent, the 2 fps condensate stream will accelerate down the 30' drop and arrive at the distribution header doing 44 fps, throughly mixed with high quality steam and (I think) with some static pressure.

Perhaps a 45 degree inclined descent would help mitigate mixing and pressurizing?

 

[Bambie]

Here are the Martinelli-Nelson Friction Multiplier Curves for those without two-phase flow capability.

 

[chicopee]

Thanks for the link. Interestingly enough, I have not found any mentioned of the Chezy-Manning equation for flows of condensate and the reason is primarily due to determining pipe sizes under full flow. While your calculated time of travel is pretty close to actual events, you may still want to do a heat transfer evaluation so that you have an idea on the temperature at the vent.

 

[Bambie]

chicopee,

I tried a 1/2"dia vent size and detected no significant reduction in saturation temperature/pressure at the vent for the reasons stated above.

I think 3 additional 1/2"dia vents installed directly above the 3"nps pipes and an inclined 30' drop would help relieve pressure, reduce mixing and reduce the amplitude of water cannon loads.

I am severely limited in what I can do because of the dual blow-off function of this line (0.2 kg/s Continuous and 50 kg/s Intermittent).

 

[chicopee]

You should evaluate a pitch decrease of the line to allow more time for the condensate to cool off below 212dF at the end of the 250' length. I suspect the 50kg/s which is now a new revelation is the cause of those "canon effects" and not the 0.2kg/sec. Also reconsider lilliput1's suggestion by convincing your manager of the problems. "Blow-off heat recovery...was deemed un-economic when the plant was built" perhaps now is the time to reconsider.

 

[Bambie]

Chicopee,

The flow of 0.2 kg/s maintains a steady, saturated pressure of 6 psig at 230 degF. The condensate temperature is the same at the beginning as the end of the 250' line because the steam pressure drop is not significant at this flow.

Reducing the pipe gradient would increase condensate depth and invite steam void collapse transients.

Insulation would need to be removed in order to reduce condensate temperature, however the line needs to be insulated and heat traced for freeze protection.

The 50 kg/s intermittent blow-off is quite unremarkable except for the low howling noise and has never caused support damage in the distribution headers.

The direct contact condensation mode map (attached to my previous post) provides more information on this issue, if you are interested in it.

 

[BronYrAur]

This was mentioned earlier in the post but wouldn't it solve all of these problems if you put a blowdown separator in the plant and vent it through the roof? All of your flash steam would Escape there and you would have nothing but hot condensate water running down to the lake

 

[chicopee]

Make a plea to allocate more money and follow BronYrAur suggestion. Heat recovery in the blowdown will save some additional money and the upgraded system should pay for itself. Keep us abreast of the strategy that you'll be using.

 

[georgeverghese]

From what I read so far, the previous incident of the blowdown line anchor damage could be due to a high velocity slug of liquid water displaced by a restart plug of hot steam vapor from the blowdown valve. This long plug of water probably got into this blowdown line by reverse flow from the lake at the time when the blowdown line was not operating ( maybe you were running the blowdown intermittently prior to this incident) and the steam in this line condensed, resulting in a near full vacuum condition.

Which is why you've got this "pressure equalisation line", which would be better described as a vacuum break line.

For the new arrangment, would suggest one minor upgrade: To prevent steam from running out through this vacuum break line, add a check valve on the vacuum break line air intake, and also include a bug screen to prevent critters from getting in. Use a dual plate check valve with low torque springs and place this check valve on a horizontal section. Else you could also use a tilting disk type check valve.

 

[Bambie]

chicopee and BronYrAur,

Environmental restrictions on morpholine and hydrazine release to the air limit vent size.

 

[BronYrAur]

With A continuous blowdown of 1600 lb per hour at 6 PSIG, your flash steam is less than 2%. So we're talking 30 pounds per hour of flash steam that would escape from the vent. And that should be a maximum because you should have some cooling before you hit the vent and even less Flash. Are you saying that 30 pounds per hour steam release is unacceptable? I would have to think you have that much being released from your boiler feed tank, Etc.

If that is unacceptable, how about incorporating some form of radiator or fan coil to reject some heat before the blow down heads into the discharge pipe. Or better yet Heat your boiler feed water with this rejected Heat.

 

[Bambie]

BronYrAur,

I am saying that the vent size is restricted by the steam that is ejected during the 50 kg/s (396,832 pounds per hour) intermittent blow-off, which generates approximately 100 psig backpressure at the vent location.

 

[Bambie]

georgeverghese,

I wish you were correct, and I apologize for not showing the vacuum breakers installed on each 3"nps distribution line just upstream of each elbow.

These vacuum breakers work very well following intermittent blow-off (when continuous blow-off is not in service), when condensation attempts to pull the lake up into the boilers.

This water hammer scenario is not a problem with the buried pipe configuration and should not be with the elevated pipe configuration.

I actually want steam to 'run out' of this vent to prevent pressurizing the header and loading the 14' water cannons.

 

[georgeverghese]

Well, if you want to maintain a max backpressure of 6psig at the tail end of this 250ft long line in order to maintain a liquid seal at the 3inch lines and thus prevent steam and a high velocity water jet from being ejected out of these lines into the lake ( presume this is what you mean by water cannon? ), simple thinking tells me to add a PSV or PIC-PCV set at 5psig on this pressure vent line - you could lead the vapor discharge from this PSV / PCV line back into your boiler at the de aerator perhaps. Calculations should tell you what the max steam rate this PSV / PCV should handle.

Alternatively, if you estimate the backpressure at the end of the 250ft line is 100psig during a 50kg/sec intermittent blowdown, a better solution would be to (a) increase the line size on these 3inch tails to decrease the backpressure and (b) immerse these discharge tails deeper down into the lake to prevent a breach of the liquid seal.

100psig backpressure does seem too much for a liquid seal - So, for example, if you have 6 new tails each 4inch dia would bring the max backpressure down to 50psig, then ask for these tails to dip down 3.5x33 = 115ft below the water surface. Am presuming the bulk of this backpressure is due to the hydraulic limitation of these tails.

 

[Bambie]

georgeverghese,

I will assume that you are being serious, (although I wonder about the 115 foot extension) so if you open the direct contact condensation mode map that I attached on my Aug 8 entry, you will see the conditions under which water cannons operate.

Briefly, it is when the rate of steam flow down a submerged vertical pipe slows until it is less than the rate of condensation at the interface with the much cooler water inside the pipe.

The vacuum created allows atmospheric pressure to push this captive column of water up the pipe at an unbelievable rate and deliver a significant momentum force to the first elbow.

 

[georgeverghese]

Havent paid much attention to these 2phase flow maps - am not much of a believer in these flow maps as they are all approximate.

From first principles, the total surface area presented at the current water - steam interface in these 3inch tails at the lake isnt much; there is some local cooling effect, and we could account for this approximately in the compressible flow calcs that you'd be using to compute max built up backpressure. The cooling effect from this pipe being exposed to the atmosphere and the condensation it causes should also be taken into account in the compressible flow calcs. Suggest using isothermal compressible flow behaviour for each of say 10segments in the 250ft pipe and another 2-3segments in the tails at the lake. Use average conditions for each segment with trial values for segment temperature. But I suspect the max built up backpressure will be in summer at zero cross windspeed and little or no cooling effect.

This 50psig example described earlier is only to illustrate how you can set up these liquid seals. If your lake depth isnt much, you can adjust the no of tails and the dia of each tail to derive a new backpressure to suit the liquid seal depth you may be limited to at this lake. You can then scrap this open vent line you've currently got on this sketch. If it all seems a little too much, reduce the intermittent flow of 50kg/sec to derive a new backpressure that is manegeable.

 

[BronYrAur]

Okay I guess I misunderstood the original post with all of my comments. I was under the understanding that the blowdown is 1600 lb per hour being discharged at 6 PSIG. Apparently there's considerably more blowdown at a much higher pressure involved here.

That being said I really think you need a blowdown separator somewhere in the building. Do you have any low pressure steam requirements where you could Harvest flash steam and send it into a low pressure system?

 

[Bambie]

georgeverghese and BronYrAur,

Changes to continuous and intermittent blow-off flow rates, pipe diameters and process are not on the table.

The elevated re-route is proposed to avoid digging and replacing buried pipe because it will interrupt production less during installation.

I am convinced that the elevated re-route will present an elevated threat to production following installation due to distribution header pressurization, which will increase the incidence of water canon degradation.

Unless someone out there has analytical or real world experience with two phase flow dynamics in vertical lines, I will snip this thread.