Insane Hydraulics

Site theme image

Results of One Brave Hydraulic Experiment

If you work with axial-piston hydraulic pumps of "common industrial" sizes, you know that they are mechanically bound to a single direction of rotation. Furthermore, you know that their valve plates are optimized for the respective single direction of rotation as well, and in the case of open-loop pumps there's a distinct difference between the suction and the pressure ports:

For example, one can clearly tell that this valve plate is built for the left-hand rotation (counterclockwise if facing the sealing surface of the plate) - because the pressure-relieving grooves are widening in the direction of rotation to assure a gradual transfer between the high and low-pressure areas, and the difference between the inlet and the outlet is obvious - the suction port is wide open - to facilitate the inlet flow, and the discharge kidney is divided into several smaller openings because it needs to be tougher to be able to keep its shape under high pressure that is literally trying to "tear it open".

Now, what do you think would happen if, despite all mechanical (and common) sense, we would run this plate "the other way around" - i.e. clockwise, subjecting the "weakly" suction port to all the might of the system pressure, conditioning the intake by the additional restriction of the pressure kidney, and, obviously, subjecting this whole "pumping contraption" to a negative pressure carry-over angle and zero damping?

I bet you had to suppress the urge to cry "Heresy!" out loud now, didn't you? I agree. This does sound like a complete desecration of axial-piston principles we all love and respect, but if you think about it - what would really happen? It is obvious that the pump would be put in terrible conditions, but would it fail immediately? Would the suctions kidney deform under pressure and cause an abnormally high leakage or even blow out? Would the noise be so loud that the operation would have to be stopped immediately? Would you even notice that something is wrong?

I bet every hydraulic tech who repairs axial piston pumps has thought about this "experiment" but has never had the courage or the "financial immunity" to deliberately conduct such an eccentric trial. Well, I have good news, my friends, for I was lucky to run into a pump that (most likely unwittingly) demonstrates exactly that - the conditon of an open-loop axial-piston unit that had to work with its valve plates "switched around" for at least 750 hours!

Here she lies in all its glory:

This is a 295-9663 CAT pump from a 345D excavator. But we all can tell it's a Kawasaki K5V212DPH that was painted yellow. The client bought the machine second-hand, and it ran problem-free under his command for 750 hours doing heavy excavations with zero issues. The main reason the pump ended up in our shop was the appearance of small external leaks, and the maintenance manager decided it was a good time for an overhaul. But once again - they had no complaints about the operation.

And, as I lifted the end-plate, I saw something strange:

Yes - you guessed it - the valve plates are switched! Look at this picture: the suction ports are in the middle, the pressure ports are at the sides. I am holding an LH valve plate in my hand, and I removed it from the driving rotary group - i.e. the RH one! I triple-checked everything and even pinched myself to make sure I was not dreaming - and then ran to fetch my camera. Such cases must be documented!

Obviously, this answers the question "will the pump work if you use the wrong plate?" - at least for the Kawasaki that uses steel plates. The answer is yes, it will. I know for a fact that it lasted for 750 hours. The valve plate even left a new mark in its seating place. Look at this picture - three are too mirrored marks now! One from the correct valve plate, and one from the "experimental" overhaul. How not confusing at all!

Check out the erosion on the valve plates. This is what a typical cavitation wear pattern of an open loop pump valve plate looks like. The barrel port kind of "projects itself" in a horseshoe pattern as it leaves the suction side and "touches" the pressure groove. On this image you can see the "sample" valve plate and the "experimental one" lying next to each other. The Kawasaki exhibits the same wear pattern, albeit in the wrong place and facing the wrong way (don't forget that the suction side is the pressure side, and the real rotation direction is clockwise) - as soon as the barrel kidney was "touching" the pressure side (the open suction port that became the pressure port) the imploding bubbles left the same horseshoe erosion pattern on the sealing surface. By the way - the last two pictures show the other valve plate, i.e. the "right-hand" one that was working on the "left-hand" side.

Obviously in such atrociously un-optimized conditions cavitation was an issue, and the state of the barrel clearly shows it:

But what surprised me the most was the very severe cavitation wear of the piston slippers. Complete chunks of material were missing:

I am glad some unknown pump technician gave us this unique opportunity!

Now we know that steel valve plates are much tougher than most think, and even the open suction port can withstand "excavator-level" pressures. Who knew?!

And we have another proof that cavitation can "bend corners". Can you imagine all of these tiny bubbles imploding under the piston slippers and eating the bronze away?

This piece of knowledge may be of little practical value, but you have to agree that it is very exclusive.