Insane Hydraulics
Bold and audacious blog about fluid power
Today I want to talk about the swashplates and servo-mechanisms of inline axial piston pumps. Doesn't sound too attractive, does it? I agree. This is exactly why a lot of techs who repair axial piston units and understand well the axial piston "pumping principle" never come to look into why the swashing systems are the way they are in different models of pumps. I believe that if you examine the way designers "tweak" the geometry of the axial piston swash-plate mechanism to meet the "tilting challenge", the boring topic becomes quite interesting.
Let us start by looking at the thrust pattern (for lack of a better term) that the system pressure "projects" on the swash-plate surface via the piston slippers. When a piston chamber of the rotating barrel connects to the pressure port of the valve plate, the respective piston exerts a force on the swash-plate that is equal to its area multiplied by the system pressure, and when it disconnects from the pressure port, the force, obviously, disappears - thus the thrust pattern is more or less the kidney-shaped projection of the valve plate pressure port on the swash-plate surface.
It is also important to note that the net force the slippers produce is quite significant, and can easily equal several tons even for relatively small units. For example, 4 to 5 pistons with a 1.5 square centimeter area running at 280 bar will be dishing out almost two tons!
In a perfectly symmetrical system, at least in theory, the system pressure would have no effect on the swashplate, because all of the pressure-induced forces would be balanced out, but in real-life pumps the axial piston swashplate mechanisms are always asymmetrical. Let us have a look at what exactly can be "misaligned to our advantage" then.
There are two things that can be "tweaked" in the swash plate tilting mechanism - the valve plate timing angle (sometimes also called pressure carry-over angle), and the offset of the swashplate axis. Let us consider them one at a time:
First - the timing angle of the valve plate:
The valve plate kidney ports can be advanced, or retarded in relation to the direction of rotation of the barrel. But since our valve plate projects the thrust force on the swashplate, advancing the timing angle (i.e. turning the valve plate in the direction of rotation) will turn the thrust projection as well, and since the swashplate is rotating around an axis, turning the projection will unbalance the distribution of the force acting over it and create torque which will either oppose or assist the tilt.
In some closed-loop pumps with automotive control (for example the A4VG Rexroth units with DA control) the timing angle is pretty advanced and is even adjustable, which allows such a system to function as a rough replacement of a torque-liming control because the system pressure acting on the swashplate can de-stroke the pump when the servo pressure, supplied by the DA valve is low. Rexroth refers to this system as the Anti Stall Control that works due to the "pressure sensing capability of the transmission pump's rotary group". Well said, Rexroth! So, in such pumps, advancing the angle will cause the transmission to become "softer" and, inversely, retarding the angle will give it "more power".
I remember how on certain older models you could get valve plates with two positioning notches, and one of them was placed in such a way that if one used it to assemble the pump, the valve plate would end up with a negative timing angle and I have seen, first hand, how such pumps could then aggressively tilt the swashplate by means of high pressure alone and then keep it tilted even when the control signal was removed! With everyone looking at the pump like "What the hell?!!" This "mysterious" failure is very easy to troubleshoot when you understand how the system pressure acts on the swashplate and how changing the timing angle affects this.
To recapitulate - a positive valve plate angle (i.e. angle in the direction of rotation) - opposes, and a negative - assists the tilting of the swashplate.
Now let us have a look at the swashplate axis offsetting - another trick pump and motor engineers often use in their designs. As it turns out, simply offsetting the swash-plate axis of rotation to one of the sides of the rotary group axis (so that the shaft and the swash-plate axes are no longer intersecting) also shifts the thrust force distribution and thus causes the system pressure to have either a negative (de-stroking) or a positive effect on the swashplate tilt.
This trick, obviously, can't be used on bi-directional closed-loop units, where both of the valve plate ports are pressure ports and the swash-plate has to tilt both ways (meaning the swashplate and the shaft axes must intersect) but for open center pumps the "gloves are off", and the designers a free to shift the swashplate axis about as they please, using the system pressure-induced forces to assist (or oppose) the tilt.
For example, the Parker PVAC pumps have the swash plate axis offset in a way that the piston-induced force straightens out the swashplate (de-strokes the pump). This arrangement allows them to use a single servo-piston to control the displacement. If you were to measure the servo pressure - you'd see that you need to pressurize the servo piston to increase the displacement, and vent it to de-stroke the pump.
Another example - the Nachi PZS open-loop pump, in which the swash-plate axis is also offset - but in the other direction, i.e. the pressure-induced forces help tilting the swashplate and the servo piston de-strokes the pump.
This design allows you to use a relatively weak bias spring because the system pressure alone works as pressure-assisted bias. Very clever, in my opinion.
This article, is, of course, very basic, and we haven't even started scratching the surface of the dynamic forces acting on the swashplate, but I still hope that the next time you look at a disassembled axial piston pump, you'd be able to think something like - "ah, now I see why the bias spring is so flimsy in this pump! It's the offset swashplate axis design! I am so clever and beautiful!"
Well, maybe ditch the "clever and beautiful", but the rest is perfectly valid!
Food for thought, gentlemen, food for thought!