Maximum Air Velocity Allowances in Pipe

[SimStil]

I have been tasked with designing the pipework for an in-house test rig of a compressor and the medium will always be air. I am struggling to find any guidance on the maximum allowable flow velocities within a pipe or what this would be based on. I understand that it's an act of balancing the size of the pipe against allowable pressure losses due to friction. In our case the pressure losses are not likely to be significant as the length of pipe will not be more than 10m, however we do require steady flow upstream of the compressor as that is where our flow measurement device will be positioned. The guidance we use for contract packages is a maximum of 25 m/s however since this particular case is for a test-rig, I don't want to blindly stick to that rule if it can be avoided.

In addition to this, can there be any issues with the flow velocity being too low? e.g. below 10 m/s.

I have only recently began my career as an engineer so forgive my ignorance in this area. We have guidelines in the department for certain things, however these have typically been developed decades ago and now very few people have a good understanding of what they're based on. The process conditions of the compressor to be tested are much more challenging than what we've experienced in the past so I want to make sure I have a good foundation of knowledge before I make any decisions - rather than blindly following what "has always been done".

Thank You

 

[LittleInch]

Sim stil.

Try doing a search on this site for maximum velocity in pipes and gems like this turn up

http://www.eng-tips.com/viewthread.cfm?qid=191447

A lot of research and empirical formulae for erosional velocities has been undertaken, but this usually requires some level of dust, dirt or liquid entrainment.

So your first issue is how clean is your air. And that means an actual filtration number or knowledge that your gas is superheated and there will be no liquid or oil or other contaminant in it.

for "clean" gas (methane) and short durations I've seen up to 40m/sec in some codes and standards, but ultimately you could go to the speed of sound / critical flow if you wanted to, but you would need to wear some pretty good ear defenders....

Low velocity so long as it doesn't get into laminar flow shouldn't be an issue so anything above about 0.5m/sec should be ok.

It's good to get your own understanding of why things are as they are, but be aware that in many occasions it is simply good practice, experience and not always codifed or even accepted between different people or companies.

Hope this helps



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

 

[hydtools]

So you will be measuring flow on the intake side of the compressor? Upstream?

Ted

 

[SimStil]

LittleInch - thanks for the information it's really helpful. Why is laminar flow undesirable in this case?

Ted - That's correct. The reasoning behind this is that the pressure and temperature upstream will be ambient and it means our measurement device will be more robust in comparison to downstream where the variation in pressure and temperature will be large.

 

[LittleInch]

Laminar flow might be Ok, but different equations start to come into play when it comes to pressure drop and flow measurements.

If inlet is doing 25m/sec+ then you're going to be operating the compressor below atmospheric pressure by quite a bit I suspect. Is that a realistic testing scenario?

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

 

[SimStil]

Sub atmospheric pressure at the inlet is not really an issue since it's what we're expecting considering we will be using a differential pressure measurement device. Ideally we want to carry out tests to replicate real world conditions as much as possible and the pressure is typically just below 1 bar anyway.

 

[georgeverghese]

Would suggest a max of 20m/sec. Given that this is the compressor inlet, also suggest keeping total losses (piping + intake filter) to less than 5% of atmospheric pressure, ie 5kpa at max flow coincident with max max intake temp. Also check what velocity range this flow meter element operates at. Turndown on velocity range for this meter must also match desired turndown for net compressor suction flow.

 

[bimr]

In addition to pressure drop considerations, the velocity of the air in the piping system is directly related to the noise produced by the airflow. If the velocity is high, the noise that is generated may be annoying. See the recommended velocities in the reference:

https://books.google.com/books?id=_...noise vs air velocity aeration piping&f=false

 

[LittleInch]

As George rightly says, sub atmospheric pressure is an issue as the density of the air being ingested into the compressor will decrease and hence your compressor won't be operating at full power compared to a much larger or shorter inlet line.

Having velocities that high on the inlet side sounds very strange to me and if this is a test rig, it doesn't mimic the real life situation the compressor will be used in. Therefore your tests won't match real life performance.

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

 

[SimStil]

Thank You all for your responses!