Insane Hydraulics
Bold and audacious blog about fluid power
Today I want to share a curious troubleshooting case involving a Rexroth M4-15 load-sensing directional control valve. I believe it has the potential to save you a lot of time if you ever bump into the same malfunction that I did this week.
I am team Danfoss all the way (for obvious commercial reasons) and I don't know Rexroth stuff that well, but a load sensing system is a load sensing system, so I felt pretty confident when I received a call from a good client of ours asking if I could have a look at the hydraulics of one of their diamond drilling rigs.
The mechanic who called me had years of experience, and the very high and consistent production rate of their rigs was a testament to how good he was. He said: "I am out of ideas, man! I've double-checked all the usual stuff, and the rig's hydraulic system is dead in the water!" Now, when an experienced tech tells me something like that - I drop everything and go there to check it out because there's always a good chance for me to learn something new!
So, I get to the mine, we descend half a mile underground, and I see an all-Rexroth core drilling rig that has no LS signal coming out of the DCV, and therefore, quite obviously, no way of doing what it is supposed to be doing - drilling!
As I said, the technician did a very thorough job of checking all the "usual culprits" before calling me. After all, when you work with the same equipment years in a row, you tend to know the weak spots inside out. Unfortunately, the man's extensive knowledge was more practice than theory-based, so he didn't think of checking the pressure at the LS port of the DCV, but he did reach out for help when the usual fixes didn't produce results.
No problem! I am here, happy to help. So, what do I see? I see a four-section load sensing DCV (the Rexroth M4-15), in which the first section works perfectly fine (in the sense that it feeds a correct LS pressure into the LS port when its spool is shifted), and the other three sections have no LS communication whatsoever.
A classic multi-section load-sensing directional control valve (no "flow-sharing shenanigans") with one section LS-ing, and the next three not LS-ing - what is the first thing that comes to your mind? I am sure that you are thinking about the LS shuttles. That's what I thought too, and said: "Looks like there's an obstruction in one of the shuttle valves, man, and since I am not sure where exactly the shuttle is located, we'll have to remove the DCV from the rig, crack it open, and see if we can find and, hopefully, clear it (my imagination was already drawing an image of me presenting a clogged orifice to the audience, silently gaping at me with admiration and awe).
The rig was very compact, and 90 percent of the valve connections were made with steel pipes. Yikes! But we bit the bullet and managed to remove and disassemble the DCV with what tools we had on the spot. Now, underground diamond drilling rigs don't go to the best places. They go to the worst ones! Often close to extraction fronts, which means "generally unpleasant underground conditions" only with less ventilation, more heat, and terrain best described as "if you drop something small - it's gone forever".
As you can imagine, one must be extremely careful when disassembling something full of small yet important and not easily replaceable parts like springs, spools, orifices, and shuttle valves with tiny (but very fast) balls. But don't worry - this is not a tale about me dropping and losing another small part - I just want to convey how doing repairs underground takes like ten times the surface amount of work!
But let's get back to our M4 now. When the tie-rods were removed, I pulled the first section out, and it became immediately apparent where the shuttle valve was located - under a small round plug at the underside of the section's body. So far so good. I removed the shuttle, pretty confident that I was about to find something "cool and unexpected" stuck in the LS shuttle (that's exactly what I said to everyone around me - "We are about to see something cool and unexpected!" - as I pulled it out - "Ta-da!!!" Oh, wait, there's nothing there....
In fact, the LS gallery holes in the body were huge, about 3mm in diameter. One would require something really nasty and definitely visible to block such a wide passage - and yet - there was nothing there! Clean, unobstructed passages. In all sections! Bummer. I could tell from the facial expressions of the "audience members" that they were disappointed with where the troubleshooting process was heading, because the DCV, apparently, was not the problem (and, of course, because they didn't get to see anything "cool and unexpected" as promised).
Hmm. Definitely not what I was expecting. Now, even though I didn't know this particular valve model that well, I know an LS DCV when I see one. Let me explain what I mean by that. Here's a cutaway view of the M4-15's work section:
This LS valve design is classic. Everybody uses it. And every hydraulic tech should know it. When I come across a "move-pressure" spool shaped like this:
(if you wonder what the expression "move-pressure spool" means, check out this article), with small holes next to the pressure-metering slots, that connect to a side gallery via the central hole in the spool - I already imagine how the valve is built. A benefit of addressing repairs with a back-engineer's mindset, I guess.
And since I can more or less imagine how a valve is built, I an also tell where the "clogging risk spots" are located. In this design it is simple - you have the obvious small hole in the LS groove of the spool:
then you have the tiny orifice in the compensator:
and then you have the shuttle valve, which, as I just saw, had the biggest holes and therefore was the least likely place to ever get obstructed.
But the thing was - I had three consecutive sections failing, and only one working. Three! I can imagine one single section failing because of an obstructed spool orifice. This can (and does) happen. But not three! When three sequential sections fail, it must be the LS shuttle! And yet - I removed and checked all the shuttles and found nothing. Then I disassembled the rest of the sections and thoroughly checked all the spools and compensators, and found no clogged orifices and no contamination at all, for that matter. No wonder - these rigs are well maintained, and the oil and filters are top-notch.
Here's a troubleshooter's dilemma for you. You know you have a malfunctioning component in front of you, but you can't prove it! I knew the valve was faulty, but I had to find a bullet-proof way of proving my theory to the "audience", which was highly reluctant to accept my malfunctioning DCV hypothesis.
Luckily, there was a similar rig parked on the surface, and I suggested that we repeat the same LS pressure test on it, and so we did, and - sure thing - the LS pressure was being perfectly communicated by all the four sections, and now I received a green light to replace the malfunctioning DCV with the good one - which we did during the next shift, and the rig "came back from the dead"! This sounds easy, by the way, but I assure you it was not, and the bruises and cuts on my forearms can attest to that.
But what about the malfunctioning DCV then? Why wasn't it working?
I took it to our shop, and hooked it up to the test bench before doing anything else - and, once again, saw that only the first section was operational.
Good. At least when a malfunction is there, I can find it. And it is so much easier to work on a component when it is on your cozy workbench as opposed to the back of a truck half a mile underground. The shop air gun helped me find the "ls-blocking spot" quickly - the air was not passing from one side of the section to the other:
I wish I had compressed air around when I disassembled the valve underground - it would have made my life so much easier!
Once again - I removed the seat of the shuttle, pulled out the tiny ball, and once again found these large, open, and completely clear galleries. How the hell can this be obstructed so well?
And then, when I peeked inside the hole where the shuttle valve sits, I saw a hexagon shape in the center of the seat and it dawned on me - the freaking bottom seat of the shuttle valve is removable!
It felt as if I was struck by a lighting. I grabbed a 2.5mm Allen key and - sure thing - the freaking seat was loose! It got loose and eventually unscrewed itself to the point that it locked the ball tight between to two seats turning the shuttle valve into a very tight plug:
Damn! I have no idea why I didn't see this in the mine. I'll blame the bad lighting conditions for that!
I managed to find a Rexroth file stating this part should be assembled with Loctite. Found no traces of it, though. Be it as it may - now I know another pretty treacherous way such a valve can fail, and, a quick way to fix it. And now you do too!
