Natural Gas Blowdown Data

[LeSabre]

Dear Colleagues:

I'm writing a computer program to characterize a long pipeline under a free flowing blowdown (I’m calculating pressure decay, blowdown time, temp profile, etc). In order to check my calculations, I need specific data from actual blowdown operations. I need the length of the pipeline, the diameter of the pipeline, the initial pressure, the final pressure, the size and length of the blowdown riser, the size and type of the blowdown valve and the recorded times it took to achieve intermediate pressures and the final pressure. Can anyone provide this kind of data? Thank you.

 

[pmover]

being curious . . .

what is purpose of writing the program? seasoned pipeline operators and/or technicians can reasonably determine blowdown times, even for new pipeline sections. is time really an issue? i doubt you will be successful in getting requested data, but good to ask.

btw, i do not have blowdown times, but time/pressure data to lower p/l pressure using a compressor i do have.

good luck!
-pmover

 

[vzeos]

pmover,

I think that blowdown time is becoming an issue as more and more transmission lines are introduced into class 4 locations. The duration of an emergency blowdown can be determinative of hazard level and frequently comes up at public hearings. I have used available programs and have found that they are ok for tanks or short pipes such as in a compressor station piping, but they fail with long lines, say 2 miles and longer. I think these programs do not account for the frictional resistance of flow along the run of pipe. I worked out an algorithm and need real data to test it...Thanks for your intrest.

 

[zdas04]

RGasEng,
I've had good success using sonic velocity down to 2 atm (abs) and recalculating velocity every 50 psi. This gives me blow down times to 2 atm within a few seconds. Then I triple the time already elapsed to get to zero (and I'm never very close, but below 2 atm the blowdown gets progressively quieter and I get fussed at less). This has worked very well on every line I've used it on (the longest was 9 miles of 8-inch).

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[Cockroach]

You can model a very generalized simulation of blowdown applications using a simple pressure vessel and full port valve. You will find that the average rate of change in volume equals that volume present. In other words, the average rate of change is constant which mathematically leads to exponental decay. There are variations of the model since it is a tad simplistic for accurate predictions. I guess you will need to define how accurate you would like to be.

I have used this approach for pipeline valves, particularly double-block-and-bleed applications. You can obtain even better results using thermodynamic tables if you know the internal pipeline fluid state and those of the ambient atmosphere. In this case, you would be pursuing an isentropic throttling scenario.

There is an excellent discussion of this phenonema in Rules of Thumb for Pipelines. The author's name escapes me now, but searchin the web for the title should ferrat good results.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[LeSabre]

I’ve been kicking this around with Tony for a couple of months now and I agree with what he wrote. Maybe I can explain it a little better.

The simple pressure vessel (PV) model does not take into account the geometry of the volume being vented. For illustration purposes, imagine a 10” pipe, 100,000 feet long. Now install a 3” blowdown stack on one end. The gas at the far end of the pipe has to flow 100,000 feet, under frictional resistance, to be vented out the stack. The PV model does not account for that pipe flow.

Now imagine a similar pipeline where the only difference is that it is now 8” dia. The PV model would predict that it would take less time to blowdown the 100,000 feet of 8” pipe because it has less volume than the 10” pipe. But if you take into account the flow in the pipe and note that the flow capacity of an 8” pipe is 55% of the capacity of a 10” pipe, then it would take more time to blowdown 100,000 feet of 8” pipe than 100,000 feet of the 10” pipe. This effect is not very noticeable in short lines where the PV model could be used. The tables below gives calculated times for the PV model and the new model. Note the new model has not been verified with field tests, which was the reason of the original thread.

 

[zdas04]

12345AM,
You didn't say what your starting pressure was for the blowdown table. If I assume it's 1000 psig, then blowing through a 3-inch at sonic velocity is 6.8 MMCF/d.

In 8-inch, 6.8 MMCF/d gives you a superficial velocity in the pipe of 3.3 ft/sec (which results in a Reynolds number if 1.3E6, pretty slow turbulent with an 8.1 psi friction drop over 100,000 ft). I get 8.75 minutes to get to 2 atm (absolute) from 1,000 in 100,000 ft of 8 inch (call the entire blowdown 35 minutes). Ten inch is 13.6 minutes for 100k ft to 2 atm or 54 minutes to zero.

I've blown 9 miles of 8-inch pipe (50k ft) from 1,000 psig to zero and the time was a lot closer to 20 minutes than 250 minutes.

In short, the actual blowdown through a small valve is controlled by the choked-flow period which is very easy to calculate. I've blown lines down through a pig receiver a few times and found that sonic flow through a 10-inch opening induces a high enough flow rate to give you some serious friction drops--but the end result added less than 3% to the blowdown time calculated by the choked-flow model.

I don't think you're on a track that matches what I've observed during hundreds of blowdowns over a couple of decades.

David