Optimal butane pressure-flow diagram help

[jerfy]

Goodafternoon,

I am a young engineer new to this community so please tell me if I should post this in another forum topic. This seemed most applicable to pipe flow.

I am working on a butane pipeline that is blending into gasoline. To tighten the blend, I theorize that there is some optimal rate & pressure at which we maintain liquid butane flowing at some maximum possible rate. I am having trouble coming up with a method for this though. Should I be comparing pipe head vs. flow rate trends to determine the highest rate possible before vaporizing? Should I be inspecting the pressure head or the pressure induced by the increasing flow rate?

I think I am overthinking this. Thanks

 

[TD2K]

"I theorize that there is some optimal rate & pressure at which we maintain liquid butane flowing at some maximum possible rate"

More flow means higher pressure losses due to friction. I'm not sure what you are trying to do.

 

[jerfy]

Specifically, I am trying to maximize flow rate.

However, our system has been flashing butane as it is injected from our smaller butane pipe into a larger gasoline pipe. I want to maximize the rate but need the pressure drop minimized to prevent this flashing.

Additionally, I've realized a couple other factors come into play here, including the head on our supply tanks and the change in pipe volume I mentioned above. I came here to ask for suggestions for how to stop this vapor flashing. My initial idea is to refit our injection pipe with perhaps several gradually increasing pipe diameters, so as not to have such a drastic dP in the system causing flash.

 

[TD2K]

The butane will start to flash whenever the system pressure equals the vapor pressure of the butane. In the butane pipeline, the minimum pressure will be at the outlet since moving liquid through a pipe results in frictional losses. Now, if you have elevation changes, that adds another factor that you need to consider and which which can help or further hinder you in terms of the butane flashing.

What is the temperature of the butane and what is the pressure at the point where you mix the two fluids? Since you are blending the butane into gasoline, I assume the butane is essentially n-butane? i-butane has a higher vapor pressure and usually isn't used for gasoline blending for that reason (plus it's potentially more valuable if you have a process like an Alky unit).

The butane vapor pressure is strongly affected by temperature as you know. At 40F, the vapor pressure is about 18 psia. At 60F, 26 psia. At 100F, 52 psia. If it's not pure n-butane then the other components can increase or decrease the vapor pressure.

If you want to keep from flashing the butane, you need to keep the system pressure > the vapor pressure. If you also want to increase the flow rate through the butane pipeline, you need to increase the supply pressure into the pipeline or you install a larger pipeline or you install a parallel pipeline (twinning).

 

[MortenA]

There are very well know relations between, flow of a given fluid (with some specific charateristics) and liquid veocity (think Darcy Wisebach). Generally speaking the dP is proportional to the square of the velocity. So knowing this there is no "optimal" velocity - dP just increases. But you may have a maximum velocty based on your minimum pressure (to avoid flashing) and your pipeline dimensions. From there its "just" engineering.

Best regards, Morte

 

[jerfy]

Thanks TD2K and MortenA, I have discovered that the problem doesn't lie exactly where I thought it did so let me restate my problem and hopefully you can help.

We have a system which injects butane into a gasoline stream. We have a VFD that maintains our pump pressure as our supply tanks drain which in turn maintains our set flow rate. The issue is that as we increase our VFD output up to a certain point to maintain our set flow, we see a point where pump pressure starts to drop off as well causing our rate to slow.

Is this issue due to the pump cavitating and butane flashing at the pump? Can it be solved by setting a maximum VFD output for which not to exceed?

 

[TD2K]

< Is this issue due to the pump cavitating and butane flashing at the pump? >

It's quite possible. Since the butane is a saturated liquid in the storage tank (or almost certainly its a saturated liquid), the NPSHA for the pump is the difference in elevation between the centerline of the pump and the butane level in the tank minus the piping line losses.

What is the pump's NPSHR per its curve at the flow(s) you see? That is feet of liquid head you need to provide. Then calculate the NPSHA as above which will also need to be in feet of liquid. NPSHA should be several feet greater than NPSHR.

Also check for any suction restrictions like a startup strainer that was never removed and is plugged up now taking extra pressure drop? Has this always been a problem or is it recent? If it's recent, what changes to the operation has happened?

 

[jerfy]

Thanks for the continued help,

Let me back up a couple of steps. I have been doing some research on pumps and head due to this issue but I am no expert. We have a manufacturer provided pump curve for our pump (corrected for butane) but I think I should create one for our pump from empirical site data. Will this show me the same NPSHR as the manufacturer curve or would this be redundant?

From there, my next question involves the actual NPSHA calculation for our system. How can I calculate a friction value? Do I assume a dynamic coefficient? And to convert this friction (which I believe is PSI) to a feet value for head, do I just do height=Pressure/(density of butane*gravity) ?

 

[MortenA]

friction is the same as "head" and the relationship for fluids between pressure loss and velocity has been know for around 100 years. Google Darcy or Darcy Weisbach or Fanning.

The available Net Suction Pressure Head is equivalent to the hydrostatic head-frictional head-fluid vapour pressure and this value must be higher that the pump required NPSH. Use e.g. Churchill to calculate f af a function of Re, roughness and ID

 

[TD2K]

< How can I calculate a friction value? >

What's your background? This isn't really a forum to try and teach someone fluid flow if you've got little or no background in it (not to be nasty, just that it's just not something you can describe in a post). Is there a chemical or mechanical engineer where you work? Either of those should be able to help you out on fluid flow.

http://udel.edu/~inamdar/EGTE215/Pipeflow.pdf

is just a start on this subject.

< but I think I should create one for our pump from empirical site data > Why? Have you checked a couple of operating points and determined that your provided pump curve isn't accurate? On one of your other questions, someone provided some good reference links I saw.