Maximum gas velocity in steel pipe. 2

[NewfieEng]

I am designing new piping header for a Natural gas piping. The pressures are around 50 psig and the flow is around 20 MMSCF/D of 0.6 SG NG. I know that 60fps is the rule of thumb for maximum velocity in a steel pipe. But my customer is questioning using smaller pipe (20 in pipe) which is already in place instead of the ideal designed pipe size (~24in). I know one of the major draw backs to high velocities is corrossion.

Aside from not being a recommended practice and the corrossion issue what are some other drawbacks to using smaller pipe and having gas velocities higher than 60 fps? and what would be an absolute maximum velocity in steel pipe to avoid excessive corrossion and other problems? 80fps? 100 fps? 120 fps....

Thanks.

 

[NewfieEng]

I should add that the 20" piping run is short (less than 50ft). so pressure loss is not that big of a deal.

 

[zdas04]

Your question is very specific to particular company policies, not physical phenomena. I have clients that use 100 ft/s as the maximum design velocity for natural gas and they don't start worrying about tearing the passivation layer from the pipe until over 150 ft/s.

Other clients look at keeping dP/mile under 5 psi/mile in 20-inch (84 ft/sec at 50 psig) to minimize hp required to recover friction drop.

When I look at 20 MMCF/d in a 20-inch pipe with an exit pressure of 50 psig, I get a velocity of 25 ft/s using the AGA equation. That sounds perfect to me.


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 harder I work, the luckier I seem

 

[SJones]

Corrosion would only be an issue if there is liquid water. Erosion might become an issue too with both water and solids. I am used to working with the numbers you quote in metres per second (not feet per second!) for carbon steel and higher for stainless steel under the right conditions. The only restriction becomes noise and vibration. Check out your fluid composition and piping configuration with a materials and corrosion engineer.

Steve Jones
Materials & Corrosion Engineer

http://www.pdo.co.om/pdo/

 

[katmar]

Tweedie, you are probably using the density of the NG at standard conditions instead of at the 50 PSIG. I agree with David's calculated velocity. Over a length of 50 ft you will get a pressure drop of around 0.01 PSI, so I would regard the use of the existing 20" pipe as very safe. This pressure drop is about 1 PSI/mile, to put it in context with David's rule-of-thumb.

regards
Katmar


Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[athomas236]

In the absence of any pressure drop criterion we would use 40 m/s

Regards,

athomas236

 

[zdas04]

Katmar's point is far too important to skip over--it is totally meaningless to calculate bulk velocity directly from a volume flow rate at standard conditions unless your gas is actually flowing at 14.73 psia (or 14.696, or 15.025, whatever "standard" means to you) and 60F (if that is your "standard" temperature).

This mistake is so ubiquitous that I made it in my masters theses and turned an important study into a pointless acedemic exercise--and no one on my theses committee caught it. Three years after I defended it, I got a call from someone who had gotten the document out of the library and said that at his conditions the equations didn't track reality. Once I got over the "denial" stage of conflict I checked the math and found my mistake. There was no way to salvage the relationships that had seemed valid in my data and I've been embarassed by it ever since.

While "volume flow rate at standard conditions" is a wonderful commercial concept and a pretty good surrogate for mass flow rate in many calculations, it has zero direct physical significance at virtually all flowing conditions. You have to convert it to "volume flow rate at actual conditions" before you can calculate velocity. That is true if you are calculating average (or bulk) velocity, if you are calculating a velocity profile, or if you need a maximum velocity--none of these concepts can reasonably be calculated using standard conditions.

In the example you gave, if the downstream pressure is exactly 14.73 psia at 60F then the bulk velocity is 109 ft/s, if the downstream pressure is 50 psig your bulk velocity is 25 ft/sec, at 100 psig it is 14 ft/sec, and at 1,000 psig it is 1.5 ft/sec (all these numbers are based on SG=0.6 and elevation=5535 so Atmospheric pressure is 12.0 psia, other elevations would result in slightly different velocities). So, depending on pressure your velocity is somewhere between way too slow and maybe too fast.

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 harder I work, the luckier I seem

 

[bulkhandling]

When the velocity of gas in a pipe is close to sonic velocity or sound velocity at the condition, noise and vibration are not avoidable. My rule of thumb is max. gas velocity at the condition is not over (or too much over) half of the calculated sonic velocity. Of coarse, pressure drop is another judge.

 

[NewfieEng]

Thanks for the input David and others. I made a mistake in my numbers the actual flow is 50MMSCF/D not 20 MMSCF/D.

I do understand the example you described with the velocity difference at Standard conditions and at Actual Conditions. And I do convert to actual conditions to get and accurate velocity. But that is very good point to make.

Knowing that some customers use 100 ft/s and 40 m/s (131ft/s) as cut offs for maximum velocity makes me feel more comfortable about using over 60 ft/s if the customer requests it. After all it is their policy and if they want to exceed it, it is their decision.

 

[JStephen]

A question of curiosity here- with gasoline lines coming into tanks, they'll limit the velocity to avoid static buildup. Can that happen with a gas?

 

[zdas04]

It definately can, and while it is less of an explosive hazard than a vented gasoline tank (since you typically don't have an oxygen source to go with the fuel and ignition source) it causes all sorts of other problems. I've seen mist extractors in separators turn into electrostaic precipitators and the solids pulled from the flow simply dance on the surface.

David

 

[TangoCleveland]

We transport different pressures of air through long lines, and have a velocity limitation of 0.3 Mach. This keeps noise levels down, primarily. We'll also look at pressure drops so that the end of the line user has sufficient pressure at his required flowrate.

Larry