Can you share a cut section and full view (pictures) of your current model and it'll be easy to suggest something.
I could be wrong, but it definitely sounds like you're modeling a bong. I will be interested to see this simulation.
To set the water level, try making a cut feature or split surface at the waterline. Also, make sure gravity is applied in the correct direction.
Unfortunately it is not something I can share any views of for reasons of confidentiality to our customer.
But for simplicity's sake (and comedic value following that last post) lets just say it IS a bong. haha
If it would help I can try to throw together something that doesnt divulge anything confidential... but it will just take some time.
Please ask any clarifying questions and I will do my best to answer.
I'll try and cobble something together to be representative of the problem.
I am not sure if you can do this in Flow Simulation. It does not do the interaction between liquid and gas, open channel flow.
This should offer a good generic model of what I am trying to accomplish.
The "soda straws" are completely submerged in the water which comes up about half-way up the vessel volume.
The "snorkel" is the exhaust port.
I am trying to get the pressure drop from the beginning of the straw to the end of the snorkel.
Feel free to post your model.
I must be missing something here. If you are interested in the overall pressure drop of the air, why even add the water to the model? Just add the hydrostatic pressure to your air inlet pressure drops?
The multiple inlet pipes will each have a pressure drop. The single outlet pipe will have a pressure drop. The motion of the air bubbles through the water will be driven by bouyancy, hence no pressure drop. Add up the inlet delta p, the outlet delta p, and the hydrostatic pressure, to get the overall delta p.
BTW, since you started a 'name the model' contest, my entry will be that it's a batch mode protein fractionator.
I know that is an option and it may end up being my best option if it simply cannot be done.
However, being able to see the actual gaseous flow path within the reservoir would be of value to me as well.
I recently had a thought...
What if I modeled it as a porous media with very little actual material substance and made a pressure gradient in the vertical axis to mimic head loss?