Calculating Gas Velocity 1

[amarks]

Hello,

Is there a quick version or rule of thumb version of calculating gas velocities in a gas well flow back line? There are so many variable, that if changed slightly can really change all of the calculations.

Just curious if someone has a quick and dirty calculation to get approximate values, making some standard assumptions, quickly without going down the rabbit hole of calculations to get very accurate results.

Would like to see what the velocity in a 3" 1502 pipe would be at various flow rates.

For example assuming:
flow rates of
45 MMSCFD
30 MMSCFD
15 MMSCFD

@ Various pressures
1500 psi
1000 psi
750 psi
500 psi

@ 80 Deg F gas temp

@ 0.7 SG of gas

@ 2.625 ID pipe

@ 80 feet long straight pipe

 

[pierreick]

Hi,
Should not be too difficult to key in data and check the results!

https://checalc.com/fluid_flow_compressible.html

Pierre

 

[Snickster]

Use the ideal gas equation P*(144)*V = m*R/MW*T*z for non-flow and convert to flow by dividing both sides by time in seconds:

P*(144)*Q = m/sec*R/MW*T*z

and then

P*(144)*Vel*A = m/sec*R/MW*T*z

Where

P is pressure in psia converted to psfa by multiplying by 144
V is volume in cubic feet
Q is flowrate in cubic feet per second
Vel is velocity in feet per second
A is pipe flow area in square feet
m is mass in pounds (weight mass W not W/g mass)
m/sec is mass flow in pounds per second
R is 1545 ft-lb/lb mole-deg Rankin (gas constant)
MW is the molecular weight of gas which is S.G. times 29 (MW for air)
T is temperature deg Rankine
Z is compressibility factor

Then solve the equation for "Vel"

 

[LittleInch]

If you want really easy stuff do it in metric.

Work out the velocity at atmospheric conditions for your flow then divide by the pressure in bar. probably within about 15%

Forget temperature - the difference is negligible
Forget density - its volume which matters here.
forget the length - 80 ft is nothing. there will be no real change in velocity from one end to the other unless you're doing hundreds of feet a second. 80 miles yes, 80 ft, no

What's a "flow back line"? Gas lift or gas injection?

that's a pretty thick 3" pipe if your OD is actually 3.5" - sch 160?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[amarks]

All - Great replies, thank you kindly for all of the input, helps greatly.

pierreick - I like the online calculator, although I'm not sure how its getting velocity without a flowrate?

Snickster - great information, thank you. I'll rearrange this equation to solve for Vel.

LittleInch - Are you talking about using the equation Snickster posted, only solving for Vel in metric? I'm not so much worried about the change in Velocity, just worried about max velocity and erosion.

Flowback line is a smaller 1502 style line usually 2, 3, 4 inch line. 1502 iron is rated for 15,000 psi. The ID of the 3" 1502 line is 2.625". This is gas flowing back from a well that is being boosted to a collection site.

 

[amarks]

Snickster - how do you get the variable m/sec. Also what is the difference between "m" and "MW"? Is it safe to just use Z=1 for NG?

 

[Latexman]

m/sec is not a variable. It’s the units of mass flow rate (m).

Snickster gave you the definitions. Please read them:

[ul]
[li]m is mass in pounds (weight mass W not W/g mass)[/li]
[li]m/sec is mass flow in pounds per second[/li]

[li]MW is the molecular weight of gas which is S.G. times 29 (MW for air)[/li]
[/ul]

Z = 1 for NG depends on how close it is to being a liquid, but it should be okay in your case.

Good Luck,
Latexman

 

[amarks]

Latex - how do you get mass flow rate?

 

[Latexman]

Rearrange the equation for m/sec = . . . and then solve.


Good Luck,
Latexman

 

[LittleInch]

Amarks,

Don't think so.

Let's break this down then you can make a spread sheet.

1 convert mmsfd to m3/sec. Multiply by 0.33

2 Find [Edit] square area of your pipe ID
(Id/2)^2 x pi. ID in metres

3 velocity at 0barg = (1) / (2)

4 To find velocity at a pressure in bar divide (3) by the pressure.

Answer in m/sec. For ft/ sac multiply by 3.3.

So 45mmscd at 500 psi (worst case) is
4242m/sec divided by pressure.

So 500 psi is approx 35 bar, so a velocity of 121m/sec. Pretty fast.

Accuracy of this method prob about 10% due to using perfect gas law.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[amarks]

Latexman - that would leave me with too many unknowns, wouldn't it?

LittleInch - That is so helpful, thank you so much for this. One question, if when finding the volume of length of pipe, instead of using 1m length I used 24.384 m (80ft) would that give me the velocity in the full length of the pipe?

Thank you again for all your help.

 

[Latexman]

that would leave me with too many unknowns, wouldn't it?

No. Can you be specific?

Good Luck,
Latexman

 

[LittleInch]

No.

Sorry, its actually pipe area you're calculating so the calculation (actually just the square area) gives you velocity in m/sec. see EDIT This will be virtually the same at the start of 8ft as it is at the end.

If you mean how long does it take then that's something different

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[amarks]

LittleInch - Interesting, so the length of the pipe doesn't change the velocity whether its 1m or 20m?

Here is what I came up with on various pressures and flowrates for velocities. Does this look correct to you? Seems a little high to me.

 

[1503-44]

1 to 20m is short. It's so short that we assume (rightly, or wrongly) that the pressure and temperature do not change very much. When pressure and temperature and the pipe diameter do not change, velocity remains the same. If there is a big pressure drop, or great change in temperature, then velocity does change. If the pipe diameter is small, then pressure change can be great and velocity will change with the drop in pressure. When in doubt, check the pressure drop and recalculate the velocity at the lower pressure. It will be higher.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[amarks]

1503-44 - Fair point, appreciate your input.

 

[LittleInch]

Looks about right.

A 3" pipe is pretty small for the highroad flowrates, even at higher pressure.

Below 10 bar the simple method doesn't work as you should be using bara, but that just complicated the formula. You need to round down to the beard 5 or 10 ft/ sec to avoid spurious accuracy. 10% is about as good as it gets here.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LittleInch]

Buy yes, Mr 44 makes the point better than me.

Unless this is a vent line doing sonic flow, velocity over 20m is going to be pretty close one end to the other. 20km is very different.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Snickster]

I will do one example for you from the values you previously gave with compressibility factor = 1.0, for lower pressures and temperatures this comp. factor will be close to 1. For higher pressure and temperatures need to calculate - check internet for calcs.

15 mmscfd
500 psig
80 deg F
0.7 SG
2.625 ID

First find mass flow rate in pounds per second using ideal gas equation at 14.7 psia and 60 deg F

15 mmscfd = 173.6 scfs

P(144)(Q)= m/sec(R/MW)(T)

14.7(144)(173.6) = m/sec(1545/0.7(29))(520)

m/sec = 9.29 lbs/sec

Now use ideal gas equation again at actual conditions and this mass flow rate:

P(144)(Vel)(A)= m/sec (R/MW)(T)

514.7(144)(Vel)((3.14/4)(2.625/12)[sup]2[/sup]) = (9.29)(1545/0.7(29))(540)

Solve for Vel = 137 ft/sec

 

[pierreick]

Hi amarks,
To reply to your question, the driving force is the pressure difference. This is a requirement to calculate the mass flowrate thus the volumetric flow rate and the velocity. All the relations are given in the calculator I shared with you.
Additional information about NG is given in the attached document.
Avoid shortcut.
My view
Pierre