How do i determine the friction losses for a reduction in pipe diamete

[pumedog]

I am pumping wastewater from an aerobic tank to a membrane tank at a flow rate of 200gpm. The inlet piping to the pump is 4" SCH 80 PVC and the outlet is 3" SCH 80 PVC. The inlet to the membrane tank reduces down to 2" so I need to determine what losses i will see by going down to 2".
Any insight is appreciated.

 

[cvg]

use the hazen williams equation to estimate the headloss

 

[BigInch]

HzW only works for straight pipe with water!

For a Reducer,

Head loss, Hf = (1/Cc - 1)^2 * V2^2 / 2 / g
where,

Cc = 0.582 + 0.0418/(1.1-D2/D1)
D1 = upstream inside diameter
D2 = downstream inside diameter
V2 = downstream fluid velocity
g = acceleration due to gravity

For an Expander, divide the above Hf by 0.7

http://virtualpipeline.spaces.msn.com

 

[rcooper]

For a tapered reducer, the head loss is so low some people say they can ignored and treated as a pipe at the smaller diameter. The losses tend to arise where a positive pressure gradient in the direction of flow exists. For an expansion the pressure gradient rises because the velocity of the flow decreases, so for the total energy to stay the same, the pressure must increase. With a tapered reducer, the pressure is decreasing through the fitting, so significant losses do not tend to occur.

The losses in sharp reducers occur downsteam of the vena contracta where the flow expands and the positive pressure gradient exists.

 

[bimr]

Get yourself a copy of Cranes Technical Paper No. 410. There is a method in the manual to calculate the resistance of a contraction if you want to go to the trouble.

 

[BigInch]

Just curious,
Are there more answers to this problem than #2?

http://virtualpipeline.spaces.msn.com

 

[cvg]

assuming the following: Hazen Williams, C=100 and water, flow rate in 3" pipe is about 3 feet/sec at about 65 gpm - your head loss in the straight pipe assuming water is about 23 feet / 1000 feet of pipe. By switching to the 2" pipe and assuming the same flow rate - your head loss would increase to about 160 feet / 1000 feet of pipe. Unless your velocity is quite high, the head loss in the reducer will be low. Also, if you have other fittings such as elbows, these will also have miner losses that may need to be considered for an accurate estimate of the head loss.

 

[BigInch]

cvg what's the 1000 ft of pipe got to do with this?

If you solve the formula.... its pretty obvious the pressure loss across his reducer is about 1 psi, that's all there is to this.

http://virtualpipeline.spaces.msn.com

 

[cvg]

Hazen and Williams published their book of hydraulic tables in 1905. MY copy is the 3rd edition, 11th printing which is dated August 1965. That book includes tables of head loss for various sizes of pipes, values of C and much more useful data for hydraulic engineers. The tables indicate head loss per 1000 feet of pipe which makes it quite useful as a planning / design guide to select an appropriate pipe size. I would have to assume that the values were expressed in feet / 1000 feet so that the head loss values in the table could be expressed in decimal form rather than scientific notation. Anyway, it is quite easy to simply multiply the length of pipe times the head loss factor in the table and then divide by 1000 to get the actual head loss.

 

[BigInch]

How do you get 1 psi with that?

http://virtualpipeline.spaces.msn.com

 

[BigInch]

Just kidding. You want to sell it?

http://virtualpipeline.spaces.msn.com

 

[katmar]

BigInch, I would agree on a 1 psi pressure drop for a sudden reduction of this size, but only about 0.3 psi for a shaped reducer. What type of reducer did you base your calc on, and what is the source of your formula?



Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

Piping Handbook- McGraw Hill

This formula is for fast flow, which is usually better for me. They also give another group of table based coefficients for slow flows, but I don't like programming table lookup and interpolation functions and the previous is conservative for slow flows. And after all its only 1 psi. OK I grant you its a bigger number in kPa (6.89 kPa)

http://virtualpipeline.spaces.msn.com