Mitered Elbow Pressure Drop 3

[scMechE]

Hello everybody!

At my job we are designing a ducting system that will convey flue gases. I've searched in several books and technical documents but i haven't found a lot of information about mitered (or gored) elbows pressure drop. For the design we have chosen to use elbows with the following characteristics: 90 degree, 5 gore, and r/D=1.5.

One of the documents that i reviewed was the "Duct Design" chapter in ASHRAE's Fundamentals Handbook. In this document i found some fitting loss coefficients for mitered ells, but i have some doubts about this coefficients:

1) In the document it says that these coefficients are "Local Loss Coefficients", represented with the letter "C". I don't understand the difference between these local loss coefficients and the more common "Resistance Coefficients", usually represented with the letter "K".

2) Does the local loss coefficient consider the material that the elbow is made of, i.e., does it consider the surface's roughness?

3) Does anybody know if these local loss coefficients can be used in the case of high temperature gases (like in this case) or they are valid just for low temperature gases, e.g. air at ambient temperature?

Or perhaps, someone has found the pressure drop coefficients for mitered ells somewhere else? A friend told me that i could find this kind of informations in API RP 533 standard.

Thanks in advance for your help, i really appreciate it!

 

[TD2K]

Crane's "Flow of fluids" has resistance coefficients for mitred elbows.

I'm not familar with your 'duct design' reference to comment on the resistance coefficients provided there.

 

[katmar]

The ASHRAE data seems a bit low compared with other sources. As TD2K has pointed out, the Crane TP410 manual gives data for mitre bends but these are for individual mitres. If you take the Crane values and multiply by the number of mitres or welds (4 for your 5 gore bend) you will get values almost double the ASHRAE values.

The Hooper 2-K and Darby 3-K methods only list data as far as 3 welds (4 gore) and their values agree very closely with Crane - most likely because they are based on the same source data (or perhaps on Crane's data directly).

Referring to your numbered questions:

1) A great deal of pipe and duct fitting experimental work was done in the early 20th century in Germany, where the symbol C is used in exactly the same way as English authors use K. I am confident the ASHRAE C values correspond with the usually used K in English texts.

2) The small amount of experimental work available (to me) shows that the duct surface does affect the pressure drop. But the experimental work is inconsistent and contradictory. It seems to me that the accepted values as given by Crane etc are based on roughnesses slightly worse than commercial pipe, but are used as being on the conservative side. Idelchik breaks the K values down into 2 parts, i.e. the part due to the change in direction and the part due to the length of the flow path. The part of the K value due to the length of the flow path is affected by the roughness of the duct surface. However, the equivalent length of your 5 gore bend will be around 23 diameters and the flow path is probably 3 or 4 diameters so the roughness affects only a small part of the overall pressure drop. The diameter of the duct has a far larger impact on the K value than does the roughness.

3) The way the K values are defined they can be used for any fluid - just make sure you multiply the K value by the velocity head determined for the actual fluid to get the pressure drop.

Katmar Software - Uconeer 3.0

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[stanier]

Idelchik covers such bends

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[ione]

It's always a pleasure to read Katmar’s posts (clear, precise definitely deserves a star).

 

[katmar]

Thanks ione. At the risk of being accused of starting a mutual admiration society I have to say that I always value and enjoy your posts too (and not only when you compliment me!).

Katmar Software - Uconeer 3.0

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[scMechE]

Thanks everybody for your help!

I looked for Darby's, Hooper's and Idelchik's methods and now i have plenty of data to continue with the design and choose a method that suits the best to the characteristics of the elbows we are designing.

Best Regards!