Authoritative reference for "pipe/flow sizing nomogram" 5

[ginsoakedboy]

I often need to size piping/tubing to allow highest flow-rate for heat transfer purposes, and I am not very familiar with Piping or Pipelines-related industry standards.

So, I have to turn to web-based sources for information. Many of the supplier websites recommend the use of the "pipe/flow sizing nomogram". I have attached an example.

Can anyone point me to the original reference publication(s) or standard(s) that may be referenced in technical reports rather than these websites that be unreliable?

Thank you in advance.

 

[zdas04]

Wouldn't eng-tips.com count as one of those "unreliable internet sources"?

David

 

[ione]

David,

This is the best of the week! A star for you

 

[ginsoakedboy]

You got me there!

I just need to know which standard refers to the nomogram so I can verify and use the standard as the reference.

 

[Latexman]

Ask IPS Flow Systems.

Good luck,
Latexman

 

[ginsoakedboy]

Wow, 3 responses to my question and all of them smartass comments. I get it, OK. All of you are indeed very smart and very very funny. This bit of info may be trivial all you MVPs, but please resist the urge to make fun of a naive question.

All I am asking for is the name of the standard or code.

 

[Latexman]

What do you expect for such a lame question? If you were a real Engineer you'd know how to do fluid flow calculations and wouldn't need a stinking nomograph. So old school!

Good luck,
Latexman

 

[ginsoakedboy]

Latexman,

I use the nomograph to determine the "recommended or allowed maximum flow velocity through a pipe". Feel free to make fun of me if you know a "fluid flow calculation" that will allow you to compute this.

The related question has been posed many time on this forum, but I have not seen any code or standard being referred to. (Neither has any question been made fun of.)

Even the MVP zdas04 claims to use some numbers (in thread378-30500 and thread378-147153), but doesn't refer to his sources. Others provide links to company websites etc. or refer to API 14E for velocity limitations based on erosion considerations.

And, then there are numerous websites with the nomographs (

http://www.tavernpc1763.com/consultant/coefficients.htm,

http://people.clarkson.edu/~wwilcox/Design/econdia.pdf,

http://rpitt.eng.ua.edu/Class/Water Resources Engineering/M3c Hazen Williams.pdf),

but don't mention the original source for these.

To reiterate, where did these nomographs originate from?

Bonus question: What is the limiting factor being considered in these nomographs (if not erosion)?

 

[zdas04]

The reason you are getting snarky comments is that you are not asking how to solve a problem, you are asking us to document a solution for you. Many of us will not do that. I only provide references to paying clients. For free you get opinions. Apparently I'm not the only one who feels that a request for documentation is not a reasonable question.

I do regularly respond to questions about minimum and maximum velocities. In gas, the answer is opinion, company policy, wild guesses. There is no authoritative source of that information. In the threads you've referenced I've tried to provide the result of a fair bit of experience. There is not a source. In liquid there are a couple of sources in NACE documents (and maybe in API), but when I've used them the dP and required pumping hp was too high and I scaled way back (i.e., bigger pipe).

I don't use nomographs, so I can't tell you where they come from. If I couldn't trace back the calculations underlying them I wouldn't use them personally. Since I don't use them I've never tried to trace back to the underlying arithmetic.

David

 

[ginsoakedboy]

zdas04, thank you for the patient explanation.

Your middle paragraph indicates that the limiting factor that determines the maximum velocity is unique to each application in a very broad sense. Some of the factors that I have seen quoted or used are cost (material, power requirements), erosion, frictional losses, pressure drops, vibration, maintaining sub-sonic flow ... and so on.

Where does one start when you need to design a system to remove heat? No way, you are going to start with a detailed analysis trying to optimize pipe flow using the above criteria. So, the nomograms are quite handy to start with a crude estimate of the pipe-size, which is then verified with numerical brute force (read CFD/FEA). Somehow nomographs seem to fit that approach.

Or, is there a better way to do this?

 

[zdas04]

In gas, the first physical upper limit is that your fluids change properties dramatically around 0.6 Mach, but that really isn't a significant sizing problem very often--the pressure drop at that velocity kills your project economics. Below that people use a range of numbers that are based more on rumor, fear, and superstition than on science. I like to keep the max low enough that compression to make up the dP doesn't kill the project economics. That usually works out to around 15 psi/mile which tends to be 30-50 ft/sec. Erosion concerns are just nonsense for clean fluids. If you are carrying sand-blasting sand or the like then you can erode the passivisation layer and increase corrosion rates, but if your solids are minimal you have to get pretty fast to touch the passivisation layer with fluffy gas.

For liquids, erosion is not a myth. The simplistic equation the people use doesn't do much to prevent it. Actual erosion characteristics seem to be more a function of Reynolds Number than of fluid density alone, but there isn't any research that I've found to support that.

If I'm designing a heat exchanger, I start with the thermodynamics and fluid mechanics of the fluids. Sorry, but that includes a whole bunch of flow-in-pipe stuff. Any shortcut is going to be jock-o-block full of assumptions that you will probably violate if you don't know about them. I'd have to know a bunch more about your process before I would recommend a flow equation to use.

David

 

[hacksaw]

nomographs had been used for years followed by simple slide-rules, but in this "modern age" computers rule the day.

Cranes flow manual is a good starting point for a modern basis, but you may have to contact the originator of the chart if you want to track down the specific reference

 

[Latexman]

Good reference hacksaw. Crane TP 410 is full of nomographs too. Chapter 3, I think, because I've never used them.

Good luck,
Latexman

 

[waterpipe]

ginsoakedboy,

There is no myth behind these nomograms. They are all graphical representative of the pressure loss formulas.
There was no need for a "numerical brute force" in old ages to solve the pressure loss formulas, but even doing some simple iterations were out of the computing capacity (and obviously out of the patience of that time engineers). So the engineers came to 2 solutions: either use a simplified equation that didn't require iterations (like hazen williams equation) or preparing a nomogram.

Considering nowadays computing capacity, none of these "adaptions" is necessary. Since your nomogram is based for water flow in a pipe and I'm the "Waterpipe", you should take my advise (just to keep the fun started at early replies) and use Darcy–Weisbach equation. You can search for an Excel sheet to solve it or simply use an online calculator such as this one:

http://www.lmnoeng.com/DarcyWeisbach.htm

Be careful, it's for steady state, incompressible, single phase flow (what did you expect from a waterpipe!).

Now, you don't need any reference or standard to backup your calculations. You can even draw your very own nomogram and name it after yourself and have the reference of it in hand!

Final notes:
- If you deal with a 2-phase flow, then look for the pressure-loss formulas applicable to 2-phase flow. Being a waterpipe, I am not of much help in this devision.
- regarding the acceptable range of velocities, you might find it helpful to take a look to this post thread378-293477 for water flow. BUT I believe this issues is completely dependant on the sector of application. Considerations in dealing with fluid flow in a heat transfer application are not the same like those in water distribution networks or gas transmission pipelines. I am sure you've dug the related handbooks to your sector.

Hope this would help.

 

[cvg]

recommend you buy and read Handbook of Hydraulics, Brater and King, at least for steady state, incompressible, single phase flow - it is invaluable.

 

[zdas04]

Waterpipe,
Good post. The only thing you left out was ASSUMPTIONS. To get a really ugly equation that doesn't have a closed-form solution to fit on a nomograph you have to make a large number of assumptions. Many of them were perfectly reasonable, some weren't. I came across one for a downhole oil pump that assumed flowing bottomhole temperature was always 60F. Basically bottom hole temperature is never 60F, and the equation that went into the nomograph was very temperature dependent. At 3,000 ft the nomograph was 25% off the calculation it purported to represent, but 14,000 ft it wasn't in the same universe.

It is rare for someone reproducing a nomograph to list the underlying assumptions. That is why I don't use them (that and those computer programs you mentioned).

David

 

[waterpipe]

Thank you for the example showing the role of assumptions.
I don't use nomograms and online calculators either. I prefer the handy Epanet for hydraulic calculations (quite a nice program for water). As you said, using nomograms without knowing their underlaying assumptions could lead to fancy results.

 

[cvg]

most equations have underlying assumptions. use of software also require a full understanding of the underlying algorithms and assumptions. Use of any equation, nomograph or other method requires the engineer to know what he is doing which means a full understanding of the assumptions, simplifications and limitations of his method.

 

[zdas04]

I couldn't agree more cvg. My contention is that the farther you get from the field data, the more assumptions have been hidden. A nomograph or a for-purpose slide rule make it very difficult to determine what is hidden.

A computer program is no better. Before I'll use a program I feed it a set of data that I know the answer to (either through manually solving it or from field data, preferably both). If the program is close enough then I get another set of data and try again. If the program is close enough 3 times in a row then I'll run it on data that I don't know the answer to. That is the only way I'll trust a program someone else wrote (I'm even tougher on the ones I write).

The underlying equations are not a panacea either. I'm currently working on building a class where I talk about the main gas flow equations that people use. I picked a real world case and solved it for the isothermal gas flow equation (iterating until the calculated Reynolds Number was within 0.01% of the guess Reynolds Number) and got a result that was amazingly close to field data (2% low on flow rate). Then I ran it in AGA fully Tubulent, Weymouth, Panhandle A, and Oliphant. None of the closed-form equations were within 10% of the field data. One (Weymouth) was 40% low. [I've actually got a flow slide rule that is based on Weymouth.] I was so suprized by these results that I got another set of data and tried again. This set of data matched the design conditions for AGA Fully Turbulent and that equation was within 5% of field data (isothermal still did better, but not by as much). The others were way off. Then I tried it with a set of data that matched Weymouth's underlying assumptions and Weymouth was 9% off field data. etc. At the end of the exercise I reached the conclusion that it was more work determining which of the closed-form equations fit than it was just to iterate on the isothermal equation.

My point is that when I started 30 years ago we could all describe the equations we used and their limitations. Today that is a foreign concept. We are worse off for it.

David

 

[moltenmetal]

What most are forgetting here is that economic pipe sizing is an optimization exercise between the first cost of piping, installed (i.e. the cost of pipe, fittings, accessories and labour) against the pipe operating cost (principally this is compressor or pump energy, but it also includes maintenance/repair/replacement sometimes).

It should be obvious that that optimum varies greatly depending on application.

For non-slurry, non corrosive liquid and gas services, erosion at anything approximating economic piping velocities is, in any practical sense, a myth- unless perhaps you're talking about 50+ year lifetimes for your piping. Think about the velocities encountered 24/7 inside the trim of a control valve before you contradict that.