Hazen-Williams Equation Accuracy? 8

[sjohns4]

Since this is an empirical equation, is there a "range" (i.e. velocity) at which the equation is the most accurate?

The reason I ask, recently I have had some hydraulic problems come across my desk where the velocity in pipes has been close to 10 fps down to 0.9 fps.

Thanks,

Mike Johnson, EI

 

[BigInch]

Have a talk with the company's lawyer. Perhaps you should submit a design change notice with the referenced article attached. If they reject it, your potential liabilities are potentially negated.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sjohns4]

My area too mandates the HW equation & C factors. If that's what the authorities require, so be it, but that's not to say some of the more challenging projects in the future won't get checked by another method to verify something wasn't missed.

You're right about the fittings: minor loss co-efficients were used in the same equation no matter what friction loss equation is used. In my example, comparing HW & DW proved that (assuming DW is near flawless) HW was accurate under these conditions.

And they weren't using HW for gas - I think it was mentioned before that the HW equation, as any equation, should be used within its limitations - in the case of HW, for water only.

 

[katmar]

A lot of opinions have been expressed here, and I think it is worth writing a bit of a summary. The original question asked for the range in which the Hazen-Williams equation was most accurate. The one non-negotiable factor is that the fluid must be cool water. Various statements have been made about the velocity, pipe size and Reynolds number, but none of these factors is sufficient by itself to define the zone in which H-W is accurate. The zone in which H-W is accurate is what is called (in Darcy Weisbach terminology) the transition zone where the friction factor is dependent on both the relative roughness and the Reynolds number.

In general terms you can say that for H-W to apply, the relative roughness (i.e. absolute roughness divided by the pipe ID) must be low. Also, as the Reynolds number gets larger, the lower the relative roughness must be in order for H-W to apply.

In practice this basically means that H-W works best for smooth pipes, which is not surprising since, as cvg pointed out, the original research by H-W was done with iron pipe. The C factor determined for smooth iron pipe at a Reynolds number of 40 thousand will still be reasonably applicable if the flow rate is increased to give a Reynolds number of 4 million. On the other hand the C factor determined for a rough concrete pipe at Re = 40 thousand will be way out if it is applied at Re = 4 million.

Mike (sjohns4) noted that at Re = 10[sup]5[/sup], H-W and D-W gave very similar results. This makes sense because at this Reynolds number all but the very roughest pipe would be in the transition zone. The differences in the conventions used by the different branches of engineering are highlighted by Mike's comment that it is easy to find H-W C factors, but difficult to find roughness factors for D-W. For me as a chemical engineer it is exactly the opposite. I have files full of D-W roughness factors for different pipe materials, but without Google I would not know where to find a H-W C factor.

Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[rconner]

Wow (saw all the stars so thought I’d come back in)-- throw out terms like “problem”, “litigation”, “serious…implications”, “lawsuit”, “lawyer” (or even more piled on, “lawyers”?), “liabilities”, and “bad piping”, along with a few colloquialisms/analogies etc. and it appears “the stars really come out”!
With regard to the specific paper with the titillating abstract excerpted by one poster, I do not know anything other than what I see here/there about this specific case study. Also, while I would not even attempt to defend any of the parties involved (perhaps with the exception of to at least some extent good ol’ Williams and Hazen, who I suspect are probably not here to defend themselves and I feel have likely provided some good service in general at least in their day to the engineering community e.g. see also thread

http://www.eng-tips.com/viewthread.cfm?qid=159749

!) I feel however some statements made in this thread perhaps should be clarified. The poster stated, “ASCE has a paper on the use of this equation…” (referring it appears to the Hazen-Williams equation). I think he means that a paper was presented by these authors in an ASCE forum, and I suspect also published in a Journal or Proceedings etc. of that event. As far as I know this should perhaps not be construed as an official position of ASCE with regard to these matters, and additionally while at least normally some peer review is provided to such papers, I believe ASCE even typically has a specific disclaimer in this regard (I don’t have this specific book that contained this paper but I believe it might read something to the basic effect, “Any statements expressed in these materials are those of the individual authors and do not necessarily represent the views of ASCE, which takes no responsibility for any statement made herein…)
I also rapidly read the entire “eight page paper” at the web address provided, and (while I admit I got a little glassy eyed at times, with my some limited basic knowledge of the subjects) this is what I took from it. This is obviously some very large concrete water pipe, with a problem at least perceived by someone, that apparently resulted in a great deal of time and effort (and I suspect the expenditure/consumption of some $?) on the part of it appears at least three different technical consultants in an attempt to prove some sort of points. I do not know who is right or wrong, or for that matter when all is said and done if there really is any right or wrong in this case (and as far as I know don’t really have a “dog in the hunt”, as I don’t mind practitioners using any or all formulas, fairly applied, that work for them!). I did notice however that in “Table 2” the authors have included the results of some sort of back-calculations from loss measurements of “the second engineering firm” (with approaches I guess they allege are more/better applicable than Hazen-Williams) they have done saying the epsilon or “equivalent roughness”, it appears they define the same symbol elsewhere in their paper as “equivalent wall roughness” (maybe the same as others have referred to as “specific roughness”?), is from 0.004 to 0.009 METERS. From my USA vantage point I see that they say this equivalent wall roughness is roughly (so to speak!) 3/16 to 3/8 of an inch!
Honestly, I’m not sure that (at least from the head loss results standpoint) even this sort of super-sized (compared to the bulk of mileage of water pipelines out there) case study is necessarily a fair indictment of Hazen-Williams if that is what someone is trying to do. I wonder if you were to ask Mssrs. Williams or Hazen way back then (or maybe even any reasonably knowledgeable practitioner using H-W since?), “Hey, the pipeline I’m building will result in flow losses exactly like would a pipe (such as in testing of Nikuradse et al long ago but ) that has 3/16” to 3/8” sized angular aggregate grains in close proximity glued to a smooth wall. What Hazen-Williams “C” value should I use for this wall condition in the Hazen-Williams formula?”, I wonder how many would come back with as high an answer, “Use a C of 120”??
To ask another question or put it another way to some of the (It would appear quite experienced) hydraulic expert-types viewing this thread, what would you say if the Owner were to come to YOU with the green field (without knowledge of this magnitude equivalent wall roughness) question, “In 4,000-5,000 meter reaches of commercially available 2.286 meter I.D. water pipe flowing at 1.655 m/s flow velocity, what do your fancy dancy equations and computers (e.g. based on say “Churchill”?) spit out for my head losses?”
As far as colloquialisms and forensic engineering in general, I understand a quite crusty ol' former football coach from a major university in the state of Alabama (who is still sort of kicking, and is probably more well-known for his successful football coaching as opposed to knowledge/ propriety of ophthalmological expression) once said, "Hindsight is 50/50." With regard to analogies, Nothing stays the same forever, but if I have a nail gun and my neighbor has a hammer, can I say he was wrong or negligent if that is what he used to build his house? (and is it even possible I might have some chance of inadvertently driving a nail at least a little further into a leg or head with the more complicated gun??)
Everyone have a good weekend.