System curve and centrifugal pump suitability

[LEO31]

I’m trying to calculate the system curve of an already existing plant, the complexity lies in having 2 different lines, each with its respective pump

1- A substance contained in a closed tank with a liquid level of 40 ft connected to pump 1.
2- a closed tank of pure water connected to pump 2

The two lines will merge together inside a static mixer (resulting into a single current with different physical properties than the 2 starting lines) and will need to travel across a heat exchanger changing its temperature and pressure and therefore physical properties, again.

I’ve only done this type of calculation by hand with a basic line with only one fluid that needs to go from point A to point B with a couple of valves in between, nothing this major and I don’t really know where to start nor if some software is perhaps necessary.

 

[erobz]

I’m trying to calculate the system curve of an already existing plant, the complexity lies in having 2 different lines, each with its respective pump

1- Pure sulfuric acid contained in a closed tank with a liquid level of 40 ft connected to pump 1.
2- a closed tank of pure water connected to pump 2

The two lines will merge together inside a static mixer (resulting into a single current at 3 bar with different physical properties than the 2 starting lines) and will need to travel across a heat exchanger changing its temperature and pressure and therefore physical properties, again.

My goal is to see if the 2 already available pumps are suitable to handle a different flow rate than the current one, and in doing so I am trying to find the system curve as a function of the mass flow, to then be able to find the operating point given by the intersection with the 2 available pump curves

I’ve only done this type of calculation by hand with a basic line with only one fluid that needs to go from point A to point B with a couple of valves in between, nothing this major and I don’t really know where to start nor if some software is perhaps necessary.

I made this system in my first engineering job. I didn't have to worry about the effects of viscosity with the sulfuric, because we used positive displacement pumps. Ours was a precise mixture, mixed slowly (over the course of a day) to reduce heat rate from the exothermic reaction; even still we had to have a simple homemade heat exchanger. I assume with your setup you are trying to make a bunch of it in a hurry?

Anyhow, at the very least this one is going to require linearization of a system of non-linear equations, and iteration to solve. If you get all the fluid properties and get dimensions, and pump curves as a function (Head as a function of discharge ) I can try to work through it with you.

 

[LEO31]

I made this system in my first engineering job. I didn't have to worry about the effects of viscosity with the sulfuric, because we used positive displacement pumps. Ours was a precise mixture, mixed slowly (over the course of a day) to reduce heat rate from the exothermic reaction; even still we had to have a simple homemade heat exchanger. I assume with your setup you are trying to make a bunch of it in a hurry?

Anyhow, at the very least this one is going to require linearization of a system of non-linear equations, and iteration to solve. If you get all the fluid properties and get dimensions, and pump curves as a function (Head as a function of discharge ) I can try to work through it with you.

Thank you for your answer ! You got that exactly right, it's a fast batch production. I already got all the properties and I guess I could easily derive an expression for the pump curve from its graph.

 

[berkeman]

My goal is to see if the 2 already available pumps are suitable to handle a different flow rate than the current one

Can you say a few things about the safety considerations in such a design change? Thanks.

 

[erobz]

Can you say a few things about the safety considerations in such a design change? Thanks.

Yeah, I'm not responsible for checking the safety of the design changes. My only involvement will be to assist the OP in developing their theoretical computation with non-corrosive fluid of high viscosity.

 

[erobz]

Thank you for your answer ! You got that exactly right, it's a fast batch production. I already got all the properties and I guess I could easily derive an expression for the pump curve from its graph.

What is your application for the solution, if you don't mind?

 

[erobz]

I've been asked to understand if the production can be pushed by using a higher mass flow, but need to check if I can use the already existing pumps

I was asking what end proceed is it used in, or is making the diluted solution the process itself?

 

[LEO31]

I was asking what end proceed is it used in, or is making the diluted solution the process itself?

Sorry for the confusion, correct: making the diluted solution is the process itself

 

[erobz]

Who is?

The OP ( engineer ) is I believe. At least in a perfect world. I’ll hold off if you think it is potential liability. I’m not an engineer, I’m just a stay at home dad!

 

[LEO31]

I'm not much more of a student myself, I'm just trying to learn the job and it's my first time having to deal with some real systems rather than the basic textbook ones ! It's all very theoretical really, this won't be actually implemented in real life by me anytime soon !

 

[erobz]

I'm not much more of a student myself, I'm just trying to learn the job and it's my first time having to deal with some real systems rather than the basic textbook ones ! It's all very theoretical really, this won't be actually implemented in real life by me anytime soon !

Yeah, its all very different in practice...real engineering doesn't tell you know a,b,c,d, etc...and we are working in chapter x! It takes time, and much effort.

 

[berkeman]

It's all very theoretical really, this won't be actually implemented in real life by me anytime soon !

That doesn't quite answer my question (to my satisfaction) yet...

Can you say a few things about the safety considerations in such a design change? Thanks.

 

[LEO31]

I've only been trying to understand how to find the system curve for a complex system, it's a theoretical project and I haven't been asked to think about safety consideration as of now. Am I misunderstanding something here ?

 

[berkeman]

Your post was reported to the Mentors as being potentially dangerous. That's why we are asking about what safety considerations you have been taking into account in your analysis. Maybe you need to start including that in your "theoretical" analysis...

 

[erobz]

I've only been trying to understand how to find the system curve for a complex system, it's a theoretical project and I haven't been asked to think about safety consideration as of now. Am I misunderstanding something here ?

If you get help from here, and your system eventually cracks a pump casing, blows a seal ( or something ) and someone gets hurt...Lawyers can/will be involved. I don't think it's right, but...Lawyers... Shouldn't need to say anything more about that.

 

[LEO31]

This is not actual work, as I said I'm not much more of a student, I only want to understand how to find a complex system curve, nobody has ever taught me this and I've only came here for some guidance to learn ! it doesn't even have to be with this specific system, anything will do. I had no idea it could have been interpreted as a dangerous question, if this is not the appropriate space I apologize and I'll keep on looking for learning material elsewhere. Again, I'm sorry for the misunderstanding.

 

[erobz]

This is not actual work, as I said I'm not much more of a student, I only want to understand how to find a complex system curve, nobody has ever taught me this and I've only came here for some guidance to learn ! it doesn't even have to be with this specific system, anything will do. I had no idea it could have been interpreted as a dangerous question, if this is not the appropriate space I apologize and I'll keep on looking for learning material elsewhere. Again, I'm sorry for the misunderstanding.

I want to help you solve a fluid mechanics problem too. But it seems they don't want us to continue unless you are going to show you understand what components could fail, how you might handle (control) such an event, or better yet...just avoid it altogether. Safety is important. But its a bit of a catch 22, for this website because you can't really know what safety issues might arise until you have some understanding of the Physics...

There would be no problem with you saying there is fluid with density $\rho$, viscosity $\mu$, they are mixed by this piping system, and they go to this other tank, etc... or just say it's a school problem. Conceal, evade, and conceal some more...this is the world we live in. If you decide to go elsewhere for help on this, keep it as general as possible so you don't run into this problem again. You aren't the first, and you won't be the last that has divulged a little too much info for the moderators.


EDIT: Look up Energy Equation in Fluid Mechanics, and Hardy-Cross Method (how to solve a flow network). Also research the effect of fluid viscosity on pump curves.

If that fails just do what you will undoubtedly find 80% of every engineer you meet does. Be wildly overconfident, iterate, and pray no one is hurt in the meantime.

Best of Luck, Sorry I wasn't allowed to help you through the procedures.