Flow of a Flashing liquid Through a Pipe

[Cotton]

I'm trying to predict the flow rate of liquid chlorine out of a 90 Ton tanker car in the event of a piping failure downstream of the tanker given the following information:
1. Pipe Size: 1", sch 80
2. Pipe Length: 90' with 12-90 Deg elbows and 3 Valves (1 angle valve at the tanker and 2 ball valves)
3. Tanker car vapor pressure 125 psi.

I have assumed a 1" hole (piping completely cut) to atmoshpere as the worst cast event.

The system we have uses cholorine gas generated by 2 hot water evaporators supplied by the tanker car. No pumps are used. We only rely on the tanker car pressure to force the liquid chlorine into the evaporators.

Any help will be appreciated.

Cotton

 

[GregLamberson]

Cotton

What I have used before is
B = (delta P x d^4 / (8.8166 x 10^-6 x K x p))^0.5, where:

B = flow rate in barrells per hr
K = resistence coeffecient (generally 1)
p = density of fluid (lbs/cu ft)
d = inside dia of the rupture (inches)
delta P = pressure differential (psi)


Greg Lamberson
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[dcasto]

If the line breaks at the tank, then you have an orifice of .985 ". The flowrate will be 107,000 lb/hr. If the line breaks about 1 foot from the tank, you have a two phase flow and there will be about 40,000 lb/hr and finally, if the line breaks near thend (about 100' equivalent lenght) the flow is 10,000 lb/hr.

 

[sailoday28]

dcasto (Chemical)
I'm just curious. Are you results based on thermo properties of chlorine?
Typically, it is a straight foward (but tedius) analysis to calculate homogeneous critical flow.
In the nuclear field, research showed that flows higher than that calculated for homogeneous blow down were observed. This goes back to around the early 1960's.
Dr. F. Moody of GE (Not the Moody for friction factors) came up with a steam water blow-down model based on the two phases flowing in an annular flow. This annular flow with slip, resulted in higher than that of homogeneous models but with flows within that actually measured. In fact back in the 80's, the USNRC recommended the the Moody blowdown model be used in safety analysis for line break.

Regards

 

[Montemayor]

Cotton:

Unless Milton Beychok posts his direct response to your query, I believe the specific solution to your problem is to to to Milton's website:

http://www.air-dispersion.com/usource.html

There, you will see the almost exact application described and the resolution given in a referenced equation. You will note that you are describing a fugitive emission of LIQUID chlorine. The pressurized liquid chlorine will adiabatically expand into a 2-phase mixture as soon as it exits the leakage point. The liquid chlorine emitted will evaporate to form an additional gaseous emission. Milton also gives you the simple equation to determine the amount of liquid and gaseous chlorine initially emitted.

I consider Milton a recognized expert and consultant in this type of emission problem and I would heed what he writes or says.

 

[dcasto]

Yeah, I ran my BWR physical data base for chlorine through my two phase flow pipeline model with Matinelli as the two phase flow equation.

 

[Cotton]

I appreciate all the discussion and information.

dcasto, how did you come up with a flow rate of 10,000 lbs/hr with an eqivalent pipe length of 100ft?

Cotton

 

[dcasto]

I didn't get the crane 410 book out but i estimated the eqvialent lenght of the 90's and valves as 10' and added to the 90'. Next I ran a two phase line flow with chlorine as a saturated liquid at 125 psig (its 87 F which seams right). I varied the flowrate until the outlet pressre was at about 15 psia. The rate was 10,000 lbs/hr. Oh, also ran the pipeline in air to add heat, a straight flashing would have yielded -28F exit temperature. I think the model predicted about 40 F. That is a variable that could be looked at in more detail, the amount of heat the line picks up.

 

[sailoday28]

dcasto (Chemical)Interesting. If you lowered outlet pressure to below 15 psia and flow still increased, then flow was not choked, but still a maximum.

Regards

 

[sailoday28]

Montemayor (Chemical) About the referenced website--
Quite some time ago, I emailed the author about errors, and received no response.
Listed below is the type of info still published on the referenced website.

mbeychok (Chemical) 8 Nov 04 13:00
One important point that is often mis-stated and/or mis-understood:

When a gas flow is at "choked" conditions (i.e., at sonic velocity), the LINEAR VELOCITY IS AT A MAXIMUM. By linear velocity, I mean ft/sec or m/sec for example.


I would like to know the reference for the above ---
since choked flow for any fluid exists when a further decrease in downstream pressure will not result in increased flow.
For a gas, under adiabatic conditions, this is at Mach=1
For a perfect gas under isothermal conditions choking occurs at Mach=1/sqrt(gamma). Clearly for T=constant, velocity is not sonic.

Realistically, conditions are between adiabatic and isothermal,

Regards