Centrifugal Pump Cosmos Floworks Problem 3

[019342806]

I am very new to Floworks and have created a centrifugal pump that consists of an impeller and a scroll casing. I need to assign an environmental pressure at inlet and outlet and let Floworks tell me the flow rates.
I am trying to simulate the fluid flow in the pump but I am having a few problems. I cannot select the impeller as a rotating region, even with it disabled. It was suggested that I should create a hidden part that covers the whole impeller, this worked and I assigned this as the rotating region, the airflow around the impeller now looked good but the results around the edge of the casing don't look right, there seems to be a separation of the two regions.
I have watched a Solidworks tutorial and on this you can see a hidden 'cavity part' and so I think I am on the right lines.
Please could anyone tell me if I’m doing this correctly?
Regards
Steve

 

[morgan349]

have you checked your mesh? or do you have any screenshots? I take it you have done the cosmos tutorial which covers this exact application?...regards

 

[019342806]

Hello,
Thanks for your reply.
Where can I find this tutorial? I have done a similar example which uses the global rotating option, this was taken from the PDF file in the Floworks help files.

Being as the impeller is offset in the casing, do I have to create a rotating region that is big enough to cover the whole of the fluid domain and not just the blade tips?

I have tried to attach a screen print but I’ve never done this before so I’m not sure I’ve done it correctly. This shows the rotating region part at the blade tips and you can clearly see that two separate airflows are being created. I need the rotating region to part of the whole cavity.

Regards

Steve

 

[019342806]

Sorry,
Here is the picture.

 

[morgan349]

The tutorial is in 'C:\Program Files\SolidWorks\COSMOS\FloWorks\lang\english\Docs' and is called 'Tutorial.pdf'

Look at chapter 12 'Rotating Impeller'.

Looking at your images, the bottom side viewdoesnt quite look right at the boundary between the impeller and the housing.

The tutorial will show you how to set up this problem.

Also pay real attention to your mesh. I cannot say this enough.

 

[019342806]

Hello,
Thanks for you reply, sorry about my late response.
I have looked at the tutorial and also had a look through the other files. The pump example uses the global rotating feature, I think that I need to use the local rotating feature to solve my problem.
When using the local rotating setting the manual states that "A component must be used to represent the volume of the fluid region". The picture attached shows an example, i.e. a fan in a tunnel and the whole fluid area covered.
Because the fan in my pump is offset to the casing does this mean that this part has to cover the whole casing (see picture)? Before I was only covering the impeller diameter.

Regards
Steve

 

[Minura]

Hello,
I am having this same problem, when i try to place a local rotating region. I have done the tutorial but it isnt really helpful as it is dealing with a global rotating region. I am unable to select my impeller, would appreciate all the help.

 

[019342806]

Hello,
When using global rotating everything is assumed to rotate so the impeller will be rotating but so will the casing, airflow and everything else you have in the model, you have to specify the items which are stationary by assigning them as real wall, stators in boundary conditions.
Also don't forget to tell floworks the axis of rotation.
Goodluck

 

[Minura]

Hi,
Thanks for the response. Got a few more questions on it though.
If I were to apply a global rotation, and select the axis of rotation could I make the rotation in the opposite direction by simply putting a negative in front of the rpm or rad/s. Also i have placed stator walls for the casing that would not rotate, but you had mentioned that the fluid flow would also rotate when I have placed a global rotation. Could you please tell me how I can stop this?
Greatly appreciate all your help.

 

[019342806]

Hello,
Yes, you need to put a negative in front.
You cannot stop the fluid rotating in this instance, if you want the fluid to flow uniformly into the impeller i think you would be better using the local region. You have to create a simple part which covers all the dimensions of the impeller and assign this as the rotating region and make sure the axis of the impeller and rotating region are aligned (mated). (see the attached example i've created).
The rotating region part must be disabled in component control the analysis then treats this area is if it were fluid. I usually then hide it.
The attached example uses an external analysis, even though it flows internally, i've found that this way you may not need to assign real walls and therefore the analysis is much quicker. You also don't need lids.
Good luck
Steve

 

[Minura]

Hi thanks for your help,
Unfortunetly I am now encountering another problem when i run floworks. The solver tell me that there is a vortex crossing the opening and I believe that this leads to negative pressures given in my goals. Do you have any ideas on what is causing this?

 

[019342806]

Hello,
I think this is quite common and is due to part of your design.
The COSMOS Tutorial PDF file explains this on page 4-10
the tutorial is usually located in: C:\Program Files\SolidWorks\COSMOS\FloWorks\lang\english\Docs

Also did you let the analysis run for the complete duration? These messages can sometimes dissapear in the late stages.

Regards
Steve

 

[altanos]

Thank you Steve,

Your rotating region example is quite helpful. But I just wonder that, what was your boundary conditions, and goals in your forward curved centrifugal fan example.

I want to analyse volume flow rate of air at different pressure differance between inlet and outlet, is there any way to find out this without lids or without totaly closed volute.

When I used totaly closed volute model , and I defined inlet and outlet face as a pressure openning ( enviroment pressure), and I defined volute as a real wall (filtered out outer face and inlet/outlet face), even I have used external analysis, all flows stay in volute, they can not flow out of volute unlike your example and results are not logical, such as; volume flow rate at the inlet ~1 m^3/s, volume flow rate at the outlet ~0,4 m^3/s, it is impossible because it is just a fan, not a compressor

Regards
Altan

 

[019342806]

Hello,
I found that when using the external analysis you do not need to assign any boundary conditions, which in turn make the analysis far faster.

In my example the scroll casing was not closed, there were opening at inlet and outlet. I had to create lids to measure the volume rate, these are position at inlet and outlet (see attached file), these were two seperate part files in this example and you MUST disable then in component control, or the flow will not pass through them. Assign the goals to the internal faces of the lids.

Hope this helps.

Steve

 

[gerdjan]

Hello,

I am working with Floworks for a couple weeks now, and I am trying to calculate the static pressure at the outlet of a centrifugal pump. I did the tutorials, and read a lot of resources, but I am not able to get any proper results..

I must do something wrong, but I don’t know what.
This is what I do to calculate:

The flow is water

I use a local rotating region for the impeller; witch has a speed of 3000 rpm
I use the following boundary conditions:

At the inlet – a static pressure of 1 bar
At the outlet – a volume flow of 40 m3/h

I set the goals for pressure drop

Is there anything I forgot to do? Do I have to use real walls when I use a local region?

The results are way out of performance of the pump. (over 100 bar, while the pump is doing 12 bar)

Note: when I calculate only the impeller and the diffuser (without the pump casing) the results seems more accurate.

I hope someone can help me out.


I also tried the solution from steve, but that didn't work neither

See attached picture

 

[gerdjan]

bump

 

[MikeCool]

You are doing it backwards. You need to:
Set a suction and discharge pressure boundry conditions.
Your goal should be the flow rate.

At a particular differential pressure, you will get a particular flow rate.

If you want to set the flow rate, you must also set a dischare pressure, and then your goal would be the suction pressure.

FloWorks has no idea what your system's resistance is, therefore, it cannot give you a static head when you only specify a suction pressure and a flow rate.

Mike Cool
Mechanical Engineer
American Stainless Pumps, Inc.
Los Angeles California USA

http://www.aspumps.com

 

[gerdjan]

Thanks Mike

I already tried to set a pressure boundarycondition at the discharge and a volumeflowrate at the suction inlet. this didn't work either.
I will now try to set pressureconditions at suction and discharge.


1 question just because i'm curious....; why is it not possible to set 2 boundaryconditions at 1 outlet? (for example a pressure and a flowrate ?)

 

[gerdjan]

Mike

Are you using floworks to calculate the centrifugalpumps that are on your website

http://www.aspumps.com

?

And if so do you get accurate results?

 

[MikeCool]

Yes, we get accurate results sometimes, but on greatly simplified models. I have just had to upgrade my system so I can study more complicated systems. I'm in the middle of that mess right now.

You can set two boundry conditions at one outlet. You could set both the discharge pressure and flow rate at the discharge nozzle, and then solve for the suction pressure. This would give you the differential across the pump. In addition, we also solve for torque on the impellers to derive pump efficiency and input horsepower.

Mike Cool
Mechanical Engineer
American Stainless Pumps, Inc.
Los Angeles California USA

http://www.aspumps.com