Pressure fed liquid tank
Pressure fed liquid tank
(OP)
Hello, I need some help.
Im working on a 100 gallon high pressure tank (1500psi MAWP) that needs to supply 125 lbs/sec @1000 psi. So far I’ve calculated the pressure drop from the supply interface through the valves/piping/etc and into the bottom of the tank.
Where I am getting tripped up is on the gn2 pressure feed.
The outlet piping is all 2” schedule 80, full bore valves and 2” weld neck 1500# flanges, the inlet piping is a little more complicated
There is a 2” section that snakes around, then is necked down to 1” tubing (0.065 wall) with AN (JIC) fittings and ultimately goes to a pressure regulator.
Ive been basically inherited this entire system, and the previous work was done by an engineer who was dismissed (he basically wrote the answer he wanted at the end of his calcs and then filled in above as he went along) they are wanting me to figure out what needs to go downstream of this regulator to get the flow we need.
Any ideas?
I had thought of using shacham and general flow equations, or maybe fanno flow.
Thanks for any tips. I think I have just been looking at this too long and over complicating it for myself.
Im working on a 100 gallon high pressure tank (1500psi MAWP) that needs to supply 125 lbs/sec @1000 psi. So far I’ve calculated the pressure drop from the supply interface through the valves/piping/etc and into the bottom of the tank.
Where I am getting tripped up is on the gn2 pressure feed.
The outlet piping is all 2” schedule 80, full bore valves and 2” weld neck 1500# flanges, the inlet piping is a little more complicated
There is a 2” section that snakes around, then is necked down to 1” tubing (0.065 wall) with AN (JIC) fittings and ultimately goes to a pressure regulator.
Ive been basically inherited this entire system, and the previous work was done by an engineer who was dismissed (he basically wrote the answer he wanted at the end of his calcs and then filled in above as he went along) they are wanting me to figure out what needs to go downstream of this regulator to get the flow we need.
Any ideas?
I had thought of using shacham and general flow equations, or maybe fanno flow.
Thanks for any tips. I think I have just been looking at this too long and over complicating it for myself.





RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
the reason I gave the information at the outlet, is because that is the requirements of the test program. Nothing else exists except the tank and the attached piping/valves/components.
I dont need for anybody to give me an answer, just some ideas on how to get unstuck.
Basically I am trying to work backwards (hence not giving the supply pressure) from the required outlet pressure/flow to see if this is even feasible with the existing system design.
RE: Pressure fed liquid tank
With the liquid density, you can determine the gn2 displacement/volumetric flow. With T and P in top of the tank, the gn2 density can be calculated. A mass flow rate can be calculated, if needed.
Characterize the pipe and fitting losses between the top of tank and outlet of regulator. Solve for the pressure there using an appropriate method - compressible, incompressible?
With the model number of the regulator and the outlet P, T, and flow you should be able to get vendor documentation to figure out what you have to supply at the inlet to regulator.
Characterize the pipe and fitting losses between the inlet of regulator to pressure source. Solve for the pressure needed at the source using an appropriate method - compressible, incompressible?
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
The liquid density is 89.93 lb/ft^3 at the test required temp/pressure
the 1000 psi needs to be at the connection point between the system and the test article. Ive calculated the pressure needed at the tank (its about a 140psi dP)
the liquid has temp requirements, so it has to be at 70 F (+/-15).
I have the volumetric flow rate, I have characterized the line from the reg to the tank inlet. But which equation do I use? The reynolds number is 5.49x10^6 so its turbulent
Remember I need the upstream pressure, all the equations (which do not give cfm, but scfm) do not have the P1 as a variable. Im using Crane 410
Thanks
RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
density = m/V = P.MW/(zRT)
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
Thanks, I will use that 1-11 and 3-20 equation as a sanity check to see if the other method (which is now working with the gn2 mass flow rate) I used checks out.
RE: Pressure fed liquid tank
Sounds like a good plan so here's my thoughts on your system.
When you been handed a crock of sh1t, you only have a limited amount of time before it becomes YOUR crock of sh1t. Therefore I would recomend you start at the basics and don't get too into the detail to start with.
What is the performence spec of your system? You quote 125lb/sec and 1000 psig - how accurate does this have to be and is there anything else controlling this. If all you have is a tank, some pipe and a gas supply feeding it you won't make either of those with any accuracy.
what is your supply amount - asusming all you have is your vessel and there's no other incoming liquid (you haven't said anything about this), based on a usable volume of 80%, you have 10.7 ft3 (0.3 m3). Your supply amount is 125 lb/second. At your stated density this is 1.4 ft3/sec (0.04m3/sec). Therefore you have 7.6 SECONDS worth of liquid in your vessel before you start blowing gas instead of liquid. As this seems to be some sort of propulsion / rocket fuel, maybe thats Ok.
what is most definetly NOT OK is the velocity through your 2" pipework. I get this at nearly 70 ft/sec (21m/sec). Therefore I'm not surprised your DP is 140 psi (not sure how long or what valves are in the way), but thats a big DP and a very high velocity. I don't know what sort of valves you're using to turn this on and off with, but if they are quick acting I don't like to think what your surge pressures / axial force is to stop that sort of flow instantaneously....
The excessive DP also brings another issue in that what pressure do you start with / contol to in your tank? 1000 psig so that what it is the millisecond you open your discharge valve at the far end or 1140 which means that at the start your end pressure is 14% higher than your stated delivery pressure??
The 1" tubing ( and that seems to immply 1"OD flexible which is even less ID and the inevitable fitting which reduce it even further have no hope of supplying you the right volume to make the supply rate you quote. You could be down to 1/2" ID on some of thos fittings and even allowing for the pressure compressibility, you're into hundreds of ft/second. I can't see your regualtor reacting fast enough or giving you the volume throughput you need from a 1" supply
So in my opinion you need to start from scratch and look critically at all the paramters and check them - clearly if you actually mean 12.5 lb/sec we are into a different ball game, but be careful you don't get sucked into just calcualting numbers which in reality are not achievable - step back and take a look at some of the output and ask is that within a reasonable range or not.
Think also how you intend to control the delivery or what actual thing needs to be controlled - pressure, flow or both? before you start worrying about your vastly undersized gas delivery pipework.
If you come to the conclusion it's all a bad design, write down your issues and send them to your supervisor / superior and tell them it will never work before it becomes YOUR design. At the same time give them a solution even if this is complete re-design. When you're doing tests, if you don't get the inputs right then a lot more time and effort is wasted than ripping out some 1" or 2" pipe and replacing it with 4" or 6".
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
RE: Pressure fed liquid tank
Yea the firing profile is only 4 seconds, and we will be refilling the tank via barrels. The 100 gallons I originally listed is the usable volume with 20% ullage space in the tank, so it’s a “true” 100 gallons.
The engines we fire usually hammer the system pretty hard, especially since we hard fill and have a spec to have pretty close to zero bubbles. The valves we are using are these huge ball valves, the actuator is almost as big as the valve, if remember right, they are not fast acting. But we will be testing the system with accelerometers, pressure transducers and water before actually putting the toxic stuff in there to see if we will have a problem. Usually what we do is fill to the test article, and let it do its thing, so its not like we are going to pressurize then slam open this ball valve, I know we are going to see some massive pressure spikes, but it is what it is.
We usually have pressure transducers on the tank and the run line and will usually bring up the tank pressure until we like the run line pressure then let the computer take over, we usually tune it pretty well so it can keep up normally (but Ive never been involved with something this extreme)
The minimum ID of the 1" fittings is 0.839, I too have no idea why they would build this giant system out of 2” line, then try to feed it with a 1” supply
Im 100% that its 125lb/sec, I actually read the requirement documents a couple times to make sure. What my plan was, is to see what kind of volume and pressure would be needed upstream and then see what our facility gas systems can deliver then just writing what is possible.
Thanks Biginch, I appreciate the input
RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
I'm really not sure about the phrase "I know we are going to see some massive pressure spikes, but it is what it is." I suppose with #1500 fittings you've got a lot of room to play with but if you just made it out of bigger pipe you won't have anywhere near as much.. If the firing is only 4 seconds long then the valves (maybe in the test item) will need to be pretty fast acting to me. either way you can do some calcuations / analysis on what it is likely to be beofre you break the pipework.
You really do need a lot bigger nitrogen fittings (and liquid line) but also need to check the capacity of the incoming nitrogen supply. In reality I think you really need a much bigger vessel which is 70% or more gas filled to get any sort of reasonably constant pressure and flow.
Good luck.
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
RE: Pressure fed liquid tank
The valves in the test article are usually fast acting electric valves. We usually do a modal analysis on the test stand/system to see if we are going to get any bad vibes.
I agree, I think we need to completely redesign these things, BUT the problem was the engineer who designed them and his project manager circumvented all the design requirements and started building WAAAY before they should have.
Here is a previous hot fire that I can show you guys
http://armadilloaerospace.com/n.x/armadillo/home/n...
and here is a video
http://media.armadilloaerospace.com/2009_04_17/Hot...
that was a wild test, the output was alot higher than we expected and it fried a ton of our equipment and wiring
Thanks again guys.
I think Im finally getting an answer Im comfortable with, almost 2,000 psi pressure and 15lb/sec of gn2 at the regulator inlet WOW
RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
Yeh, that sounds about right - sonic velocity through your 1" regulator probably....
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
RE: Pressure fed liquid tank
Thanks again guys
RE: Pressure fed liquid tank
Again, amazing!
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
Another tank / vessel like the one you already have filled with Nitrogen might be the easiest solution, but with an 8" cross connection.
I still can't get over the picture - but then I'm easily impressed.
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
RE: Pressure fed liquid tank
Good luck,
Latexman
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers
RE: Pressure fed liquid tank
Technically from our operating procedure the isolation valve wont close until the pressure comes down to the standby pressure, however the test articles valves will close and in the past we commonly used 300psi with much much lower flow rates and we would see 700+psi spikes, we actually had a filter fail.
we have mobile tube banks, that will probably help, but their outlet connections are 1-1/2 or at most 2" dont remember.
its definitely fun to sit on the test panel during a test, never gets old