l have a pump rated at 4m3/h and 35m head. lt pumps edible oil through an economizer and the to dearator with some nozzles. From the dearator there is another pump feeding a vessel (rated 9m3/h and 40m head) and there is a flowmeter reading 1800L/h (1.8m3/h). Some how the dearator gets empty or starved with oil. My question is how so as the dearator pump is just pumping 1.8 m3/h from the dearator, while the one feeding it is feeding at a rate of 4m3/h. see attached schematics
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations JAE on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Flow inconsistence
- Thread starter Thuba
- Start date
HTURKAK
Structural
- Jul 22, 2017
- 3,333
IMO ,the answer is simple . The capacity of pump 101 could be 4 m3/hr but still the flow is less than 1.8 m3/hr or the flow is more than 1.8 m3/hr.
I will suggest you two alternatives,
-provide frequency converter for pump 102 modulated with the oil level at the deaerator.
- check if possible to lift the deaerator above the vessel and delete pump 102
You may also consider to provide more details to get better responds
Last edited:
EdStainless
Materials
I take that all of your flows and heads have been corrected for SG and viscosity.
I agree with HT, you are either not getting 1.8 out of 101 or the measured flow is wrong.
I agree with HT, you are either not getting 1.8 out of 101 or the measured flow is wrong.
LittleInch
Petroleum
This is why you don't double post. There are some similar replies in the chemical forum.
This was mine
What type of pumps are these?
PD type pumps will generally pump at a fixed flow, +/- 10%.
Centrifugal pumps generally provide a fixed differential head +/- 15% from your duty point.
T
The duty point flow is only one point on a curve. What flow you actually get is dependant on your downstream frictional and head losses.
This was mine
What type of pumps are these?
PD type pumps will generally pump at a fixed flow, +/- 10%.
Centrifugal pumps generally provide a fixed differential head +/- 15% from your duty point.
T
The duty point flow is only one point on a curve. What flow you actually get is dependant on your downstream frictional and head losses.
- Thread starter
- #5
yes we use a flowmeter dedicated to our product, soybean oil corrected for S.G. and viscosity at given temperature
- Thread starter
- #6
Assuming l pump a volume of 2m3/h from a pump, l expect that to be constant in system through continuity equation. Only pressure should drop due to frictional losses.
LittleInch
Petroleum
Please answer all the questions if were going to get anywhere. pump volumetric flow being nearly constant only applies to PD type pumps.
- Thread starter
- #8
its a centrifugal pump, and the discharge valve is fully open
LittleInch
Petroleum
The pump 102 is only capable of pumping what is coming into the deaerator on a steady state basis.
So pump 101 is only pumping at 1.8 m3/hr or less.
This means that your pump is connected to a system with a higher friction losses than the pump can supply at 4m3/hr.
"rated flow" for a centrifugal pump is just one point on the curve chosen to be as close possible to BEP for the duty requested.
But this is like saying my cars "rated speed" on the motorway is 90 MPH. But if the motorway goes up hill with a wind against you and is full of people and luggage then it will only actually travel at 55 MPH. Duty point for a centrifugal pump is not a fixed point.
So pump 101 is only pumping at 1.8 m3/hr or less.
This means that your pump is connected to a system with a higher friction losses than the pump can supply at 4m3/hr.
"rated flow" for a centrifugal pump is just one point on the curve chosen to be as close possible to BEP for the duty requested.
But this is like saying my cars "rated speed" on the motorway is 90 MPH. But if the motorway goes up hill with a wind against you and is full of people and luggage then it will only actually travel at 55 MPH. Duty point for a centrifugal pump is not a fixed point.
- Thread starter
- #10
Lets assume l am pumping water, that comes to a pump at 1000kg/h and the pump adds a pressure of 5bars, hence leaving a pump at 6bars and 1000kg/h. l then added a valve on the simulation to simulate system pressure drop on COCO simulation. On stream 3, only pressure decreased, but mass flowrate remained constant.
Attachments
My understanding from your numerous previous posts that you are taking existing pumps from somewhere and placing them in systems they were not originally designed for. I think that is why there is such a mis-match in stated design flow rates of the two pumps. Correct? These are some of these same pumps you previously posted about.
My guess is that the system friction loss for the first pump causes the first pump to operate at a very high head and low flow - high on curve. I mentioned on other post that this may be due to under sizing of the spray nozzles or other reason that there is a very high pressure loss in the first pump system. Therefore, the first pump is pumping at 1.8 m3/hr and the second pump is just pumping what is being supplied by the first pump.
My guess is that the system friction loss for the first pump causes the first pump to operate at a very high head and low flow - high on curve. I mentioned on other post that this may be due to under sizing of the spray nozzles or other reason that there is a very high pressure loss in the first pump system. Therefore, the first pump is pumping at 1.8 m3/hr and the second pump is just pumping what is being supplied by the first pump.
- Thread starter
- #12
thats correct, the pumps are being repurposed. So by the fact that the pump operating at very high head and low flow, it means we would need a pump with a higher head?
First figure out what flow you really want/need. What flow is required by your process? How much flow are you trying to deaerate in your system? 1,8 m3/hr? 4 m3/hr? Once you determine the required flowrate of your system, check the required head output of each pump in meters at that flow.
For the first pump you need to determine at the required flowrate what pressure (equivalent head in meters) you need at the spray nozzle inlets to get the total flow required summed across as many nozzles you have. Then add pressure drop in piping and economizer back to the pump discharge. This is the required pressure at discharge (head in meters), subtract the suction head at the inlet of the pump to get the total differential required by the pump no. 1. Compare this to the actual performance of the pump of 35m at 4 m3/hr. Is the pump performance close to what you calculate as required? What is off? The flowrate of the pump or the head output of the pump?
If the flow of 4m3/hr is about what you need but at a higher calculated head then possibly you need to change out the spray nozzles to ones that give more flow at a lower available pressure. This will reduce the required head of the pump. I don't see anything in the first pump system that could be causing a high pressure requirement other than the spray nozzles. 35 meters head is basically 50 psi. Assuming you have about 10 psi drop in your piping and equipment back to the pump then you have about 40 psi available at the nozzles which should be about typical required for spray nozzles. So it appears you have enough pressure at nozzles but not enough flow indicating the nozzles are undersized and need to be replaced.
For the second pump I would check head requirements based on a maximum flowrate that can be supplied by the first pump as this will be the maximum flow possible by the second pump. So you have a total head available of 40 m at 9 m3/hr. So given the actual flow to be supplied by the first pump you calculate the required head of the 2nd pump and if less than 40 m then, which I believe it will be, then your second pump is OK. In fact you will likely need a throttle valve as you show in your calculation diagram on the discharge of your 2nd pump to induce head loss to get the flow of the 2nd pump to match the first.
For the first pump you need to determine at the required flowrate what pressure (equivalent head in meters) you need at the spray nozzle inlets to get the total flow required summed across as many nozzles you have. Then add pressure drop in piping and economizer back to the pump discharge. This is the required pressure at discharge (head in meters), subtract the suction head at the inlet of the pump to get the total differential required by the pump no. 1. Compare this to the actual performance of the pump of 35m at 4 m3/hr. Is the pump performance close to what you calculate as required? What is off? The flowrate of the pump or the head output of the pump?
If the flow of 4m3/hr is about what you need but at a higher calculated head then possibly you need to change out the spray nozzles to ones that give more flow at a lower available pressure. This will reduce the required head of the pump. I don't see anything in the first pump system that could be causing a high pressure requirement other than the spray nozzles. 35 meters head is basically 50 psi. Assuming you have about 10 psi drop in your piping and equipment back to the pump then you have about 40 psi available at the nozzles which should be about typical required for spray nozzles. So it appears you have enough pressure at nozzles but not enough flow indicating the nozzles are undersized and need to be replaced.
For the second pump I would check head requirements based on a maximum flowrate that can be supplied by the first pump as this will be the maximum flow possible by the second pump. So you have a total head available of 40 m at 9 m3/hr. So given the actual flow to be supplied by the first pump you calculate the required head of the 2nd pump and if less than 40 m then, which I believe it will be, then your second pump is OK. In fact you will likely need a throttle valve as you show in your calculation diagram on the discharge of your 2nd pump to induce head loss to get the flow of the 2nd pump to match the first.
Last edited:
Similar threads
- Locked
- Question
