Head loss in a circuit of pump and pipes

[physea]

Hello! In an experiment (I don't have the details) the predicted head loss in a circuit of pump and pipes was higher as the flow rate increased, than the actual head loss (pressure differential). Can you tell me please what could result in that? What systematically acting factor created an increase in the difference of head loss between predicted and measured?

 

[russ_watters]

You made an overly conservative assumption about the pipe friction and fitting loss, probably...or an error somewhere.

 

[physea]

You made an overly conservative assumption about the pipe friction and fitting loss, probably...or an error somewhere.

OK but this won't explain the increasing trend of the experimental/theoretical as flow rate increases? It would be a constant difference!

 

[sophiecentaur]

OK but this won't explain the increasing trend of the experimental/theoretical as flow rate increases? It would be a constant difference!

I would have expected the pressure drop to be proportional to the flow rate. After all, with no flow, there would be no pressure difference. A (correctly) calculated head loss would show this too so the error would also grow as flow increases. The electrical analogue to this would be a wrongly marked resistor and a consequential change in measured V/I slope.

 

[physea]

I would have expected the pressure drop to be proportional to the flow rate. After all, with no flow, there would be no pressure difference. A (correctly) calculated head loss would show this too so the error would also grow as flow increases. The electrical analogue to this would be a wrongly marked resistor and a consequential change in measured V/I slope.

Why it would grow as flow increases?

 

[sophiecentaur]

Why it would grow as flow increases?

As far as I can see from what you wrote, you have a measured resistance to flow and you are comparing this with a calculated resistance. If that's what is happening then there will be a steady ratio between pressure drop and flow. You have to acknowledge that, with no flow, there will be no pressure drop (calculated or measured) so how can there be a constant difference?
(Unless I am reading your OP wrongly.)

 

[russ_watters]

OK but this won't explain the increasing trend of the experimental/theoretical as flow rate increases? It would be a constant difference!

It should neither be constant nor even linear: it should be a square function.

Why it would grow as flow increases?

I don't understand why you would think this. Does your car need the same power at any speed?

 

[russ_watters]

I would have expected the pressure drop to be proportional to the flow rate.

Pressure drop is proportional to the square of the flow rate.

 

[sophiecentaur]

Pressure drop is proportional to the square of the flow rate.

Oh. Is that right for low flow speeds, without turbulence? Is there not a linear region?

 

[Chestermiller]

Hello! In an experiment (I don't have the details) the predicted head loss in a circuit of pump and pipes was higher as the flow rate increased, than the actual head loss (pressure differential). Can you tell me please what could result in that? What systematically acting factor created an increase in the difference of head loss between predicted and measured?

Let's see the actual calculation and schematic of the system.

 

[russ_watters]

Oh. Is that right for low flow speeds, without turbulence? Is there not a linear region?

Laminar is linear because there is slipping along the wall, like pushing a block across the floor. Look up the Darcy-Weisbach equation.

 

[sophiecentaur]

Laminar is linear because there is slipping along the wall, like pushing a block across the floor. Look up the Darcy-Weisbach equation.

OK. Not Totally Wrong but only right for low flow volume. It's a square law for the OP then, which gives even more departure from constant difference.

 

[physea]

I was thinking that there is a problem with the Uavg used to calculate Darcy losses.

A rotameter is used to measure flow rate at the beginning of the circuit. The liquid runs through the circuit that has bends, constrictions and other fittings. Would the U obtained from the rotameter be the Uavg? I expect it to be an overestimate.

1) Why would the velocity drop across the circuit? due to energy losses?
2) Why would that create an increasing trend of Darcy vs Experimental losses? Do higher rotameter values create higher Uavg?

 

[sophiecentaur]

I was thinking that there is a problem with the Uavg used to calculate Darcy losses.

A rotameter is used to measure flow rate at the beginning of the circuit. The liquid runs through the circuit that has bends, constrictions and other fittings. Would the U obtained from the rotameter be the Uavg? I expect it to be an overestimate.

1) Why would the velocity drop across the circuit? due to energy losses?
2) Why would that create an increasing trend of Darcy vs Experimental losses? Do higher rotameter values create higher Uavg?

1) Does the velocity actually drop? Have you some values or a graph?
2) Have you taken on board my Electrical Equivalent? Most people have done some EE at School and it's more of a common language than fluid dynamics.

 

[Chestermiller]

Please see my post #10. If you don't provide the information I requested, then it is my determination that you are just wasting all of our time. In that case, I will have to close this thread. You have a few hours to respond.