I really enjoy talking about unique malfunctions of hydraulic systems because looking into rare events "broadens horizons", but even more than that, I love discussing (and documenting) recurring failures - i.e. identical breakdowns of particular hydraulic components (or component series) that I "catch" multiple times. I find them especially interesting because the very fact that they are "chronic" means that I've found a bug - i.e. a manufacturing defect or even a full-blown design flaw, and the exclusive knowledge of the possibility of running into such a defect becomes the secret weapon that makes my diagnostic more efficient whenever I troubleshoot a system that employs such a component.
Today I want to "make it official" for another part - the Rexroth M4-15 sandwich load-sensing valve and its "runaway shuttle seat syndrome." This week I received yet another one of these with the inner seat of one of the LS shuttle valves (part NÂș8 in the sectional view) loose and dislocated from its normal position, causing blockage of the LS gallery and making all the downstream sections "dead":
I've already described this failure the first time I caught it - so do check this article out if you want more details. Blockage of the LS path is a common malfunction to load-sensing DCVs, but this one is especially sneaky because, unlike a clogged orifice, a seat that is only slightly lifted from its normal position does not give you an obvious visual clue that something is wrong when you disassemble and inspect the shuttle valve. The only way to check if the seat is in the correct position is to insert an Allen key in it and check if it is loose - but if you are not expecting this, you will most likely never think of doing something like that.
Let me show you a couple of fast ways to diagnose such a malfunction. There are, obviously, about a thousand more ways to do this, but I just want to show you how easy it can be when you know what to expect.
Let us begin by looking at the valve:
So, what do we see here? We see a multi-sectional DCV with hydraulically piloted spools - which is great - no fuss with them electric proportional valves, pilot supplies, and whatnot - the simpler the better, right? Then we have two side modules - one of them has the ports marked with the letters P, T, M, and LS, and the other one has T1, T2, P2, and X.
If you have worked with proportional load-sensing DCVs before, I am sure that you are able to recognize that the module with the markings P, T, and LS (and M - for "manometer" routed directly to the P) is the inlet module.
Let us leave the other module be for now. If you look at the selection of standard end plates listed in the M4-15 catalog, you will see that port X is universally described as the pilot oil supply port, but not in this case - this particular valve is proprietary and is manufactured exclusively for OEMs (core drilling rig application) hence the additional (and very wide) tank port and the non-standard port marking of the end plate (which is a shame, IMO, because I like to see standard parts everywhere - best for clients and mechanics). But let us pretend right now that we don't know what the X port does. The chances that the P port would be the pump port, the T ports would be the tank ports, and the LS port would be the load-sensing pressure port are about 100%, so let us concentrate on them for now.
The only things that we need to begin our LS diagnostics are a blow gun and a way to move the spools (and a source of compressed air, of course). I will be using our test stand for the spool shifting, but there are other ways - you could use, for example, a hand pump to pilot the spools (obviously respecting the fact that "over-enthusiastic piloting" can blow the pilot caps off). Why do that? Since we already know from what our client told us that it most likely is the LS shuttle network issue, all we need to do is command the first spool (starting from the inlet module) and then blow compressed air in the work port and see if the air comes out of the LS port. You do this on the first spool, then on the second, and then on and on till you get to the spool that does not communicate the air to the LS port - and you just found the blockage spot. Then you investigate further.
Why do I use compressed air, and not the proper "wet test"? Because when your "primary suspect" is the LS shuttle network - it only takes a couple of minutes to check it with compressed air. Now if the compressed air didn't reveal any apparent faults - then I would need the wet test, but I really like to start with simpler things. Also - seeing the spools physically shift by looking in the work ports is a good way to visually confirm that they are indeed moving and doing the complete travel.
Here, I am injecting compressed air in the work port of the first spool (the spool is shifted by the pilot pressure supplied by our test stand), and the air is coming out of the LS port all right, and when I shift the second spool and blow compressed air in the respective work port - I see no air coming out of the LS port, so the blockage is most likely in the shuttle valve of the second section.
Take a peek inside the work port - if you zoom in, you can see the small orifice in the spool that channels the load-sensing signal to the shuttle - that's how I know which is the active work port just by looking at the spool through the work port, but of course, you could also blow compressed air into the P port and see which of the work ports the air is blowing out from (note that in this case your compressed air source must have enough pressure to shift the compensators or check-valves, if any - so this method may not work for sections with compensators with especially strong springs).
Since this is "not my first rodeo," I already know where the shuttles live, so I turn the valve on its head, remove the shuttle of the second section - and sure thing - the inner seat is loose again, which I easily confirm with an Allen key:
All I need to do now is check if the other seats are tight and then re-mount the seal in place with some Loctite - and the vale is good to go. And, by the way, props to the Rexroth engineers who designed a sectional directional valve with LS shuttles accessible for inspection and service without the need to open the stack - it can be a lifesaver in the field!
Now, let us go back to the X port on the end module. In this case, it is the inlet of the LS shuttle valve of the last section - for feeding in an external LS signal:
If you know this, locating the blockage is even easier: first - you blow compressed air in the X port - and confirm the blockage of the shuttle network (no air coming out of the LS port), then you turn the DCV on its head and remove the shuttles one by one starting from the first section till you get to the shuttle with the air coming out. Super easy!
It took me a shift underground and then more hours in the shop to figure it out the first time I ran into this malfunction. Now, knowing what I learned - it took me maybe a couple of minutes, and the only tools I used were an Allen key set and a blow gun!
To resume - the M4s have been having "Loctite issues" on the LS shuttle valve assembly line, apparently - so bear the "runaway shuttle seat syndrome" in mind should you ever work on one of these, and always give these little seats a "tightness check" with an Allen key.
Also - the compressed air test is obviously a hack and not a replacement for a proper revision - when you disassemble the DCV, clean it, inspect all the parts, and replace all the seals, but I still defend the use of compressed air for quick verification of shuttle valve networks because of how quickly it can confirm and pinpoint blockages.