Simscape Fluids - hydraulic cylinder with inlet port above fluid level

I am trying to model a hydraulic cylinder whose inlet port is normally submerged (0-3m), but can sometimes be exposed to air at atmospheric pressure. I want the model to account for air sucked into the cylinder chamber on the extension stroke and the subsequent compression of the air on the retraction stroke. The pressure in the cylinder chamber and the flow out of the outlet port should be calculated based on the fluid / gas ratio in the chamber. The outlet flow goes to a reservoir at constant pressure (30bar).

I am currently trying to model the cylinder using a function block where I calculate the change of pressure and change in gas/liquid ratio based on piston position and velocity, inlet submergence and integrated pressure and ratio changes from the last time step. This approach is not giving me stable solutions in the simplest of configurations yet so I wanted to ask if there might be better ways of modeling this system.

Would it be possible to use partially filled pipes or other Fluids blocks to model the system? It would be a lot handier if I could use standard blocks as that way Simscape can handle fluid compressibility, entrained air, etc. which it's good at and I don't particularly want to re-implement myself.

I'm not asking for a full model, but more so insights into the best method of approaching this problem.

Many thanks

[Yifeng Tang]

Quick clarifying question: the rod is being moved by some power source to pull in liquid and sometimes air in this application, is that right? It's not a typical hydraulic actuator where the hydraulic fluid pushes or pulls on the cylinder. This is like a ... piston pump?

[Joseph van 't Hoff]

That is right, the piston is moved by an external force on both the extending and retracting stroke, so the cylinder is being used as a pump. On the extension stroke fluid is sucked in from just below the external fluid surface or air is sucked in if the inlet becomes exposed (submerged/exposed inlet state is set by an external time trace that is independent of the piston motion). On the retraction stroke, if there is gas in the chamber this will be compressed first until pressure exceeds 30bar and then fluid (or the fluid/gas mixture) is pumped out of the chamber.

I am mainly interested in the effect of intermittent gas intake on the piston motion (not necesarilly the exact fluid composition, i.e. how the gas dissolves)

Answer 1

[Yifeng Tang]

Hi @Joseph van 't Hoff,

Here is an idea. Looks like your model needs to be able to do (at least) 3 things: (1) pull in liquid under certain conditions, (2) pull in gas under certain conditions, and (3) keeping the liquid and gas inside the "cylinder" at the same pressure. I think (1) and (2) can be achieved quite easily by using single-acting actuators or simply mechanical converters. For item (3), I tried a Double-Acting Actuator (G-IL), and that seems to work fine. A Tank (G-IL) may work, too, but I didn't investigate further.

I used a "lever" to link the two rods of the actuators with the "actual" rod. When either end is locked because of no fluid can come in, the other end will be 2x the rod speed, so I guess both actuator area should be half of the actual area. The Double-Acting Actuator (G-IL) in the middle will enforce nearly the same pressure between the gas and the liquid.

The submerging can be mimicked by opening and closing valves to the actuators. In this proof-of-concept model, I arbitrarily set the gas passage to open when the input on the left is greater than 1, and liquid to open when it's less than -1. The openings transition between -1 & 1 with partial opening areas. The velocity signal on the right just keep pulling the rod to pull in liquid or air, and I'm seeing alternative increasing in gas and liquid volume. Sort of making sense.

Hope this helps. Definitely not a finished product, but maybe it's reasonable enough :)