One of the best ways to learn industrial hydraulics is the methodical troubleshooting of hydraulic malfunctions. And the best teachers are what I call “the impossible failures” - i.e. failures when something that can't be happening is happening. In today's post, I'll show you one, and hopefully, we'll learn something.
So, a client called the other day, saying that a brand new Rexroth A6VM107 motor that we'd sold him was not working. The motor was driving a drilling head of a medium-sized core-drilling rig, and according to the mechanic, there was something really wrong with it, because while drilling perfectly fine, it wasn't able to turn in the opposite direction. Like at all. Even with an empty chuck. They could hear a clank and they would see the head “attempt” to turn, but then it would lock with a lot of oil coming out of the drain line.
Even five-year-old children know what an excessive drain flow means for an axial-piston hydraulic motor... But I wasn't convinced. Because brand-new hydraulic motors don't do that, especially in one direction only.
Still, the mechanic, who was working on the rig, was a very experienced fellow who had his fair share of hydraulic malfunctions, and so he did the good old “swap-em-lines” test to get a clue about what was going on, and swapping the motor lines didn't invert the problem - the motor would keep on locking in the same direction. He was also savvy enough to check the pressure and the drain line flow rate - and saw that the pressure was lower than normal and the drain flow rate was (extremely) excessive, and then, when he plugged the reverse direction hose just to test the system - the pressure jumped back to its normal value.
Ok, so now let us talk about why I called this malfunction an impossible malfunction.
First, as I said - brand-new motors don't do that. But OK, I suppose we could get a very unlikely lemon from Rexroth, but let's theorize a bit more.
Any type of rotary group failure that causes locking and tons of drain flow is bi-directional and usually final (i.e. fatal), almost always accompanied by other signs of a catastrophic failure (like shavings found in the drain line). The drain line was disconnected and no signs of debris or metal shavings were seen. And let's not forget - the motor was behaving absolutely fine in the forward direction. One hundred percent.
So, this leaves us with only two suspects - the loop flushing valve (which this motor did have) and the valve plate lift (something I've already had a very... let's say... invigorating experience with).
It is very unlikely for the loop-flushing spool to lock in one of its extreme positions, but it is not impossible. So, this could explain the lower-than-normal pressure and the increased flow rate in the drain line, but it does not explain why the motor would lock (with an empty chuck, remember?).
The same goes for the valve plate. Valve plates and cylinder blocks act out under certain amounts of high pressure, and it is very hard, if not impossible, to create conditions when the motor does not turn in one signle direction and in all the attempts. At least sometimes it had to turn.
So it had to be something hydraulic (or purely mechanical).
I had this conversation over the phone, so I didn't have the luxury to inspect the failure in person, and I was already thinking about other tests they could run, but the client told me that he'd already arranged for removing the motor and sending it to our shop for a check-up. Ok, I guess the diagnostics are done then. We'll talk more when I get in on my operating table.
The first “surprise” when the motor arrived was the fact that what everybody was referring to as “the motor” was actually the hydraulic motor plus a pretty complex-looking valve manifold attached to it. I learned then that it was a common practice to store and replace the motors with the manifolds attached. Since it was always that way - everybody got used to calling the complete assembly just “the motor”.
So I wondered, naturally, if there was anything in the manifold that could be shutting off the motor, like an over-center valve or something, and the mechanic told me that there was, indeed, a reverse flow restrictor, that limited the speed of the motor in the reverse direction, but it was checked and re-checked multiple times, and no obstructions or malfunctions of any sort were found.
At least now I saw something else that could be the problem - because I still didn't believe that the problem was somehow inside the motor. I did open it - and saw nothing wrong, just as I expected - the rotary group was brand new, and the flushing valve arrangement looked intact. So - it had to be the manifold then.
A quick inspection of the manifold confirmed that there was zero chance of anything obstructing the flow and locking the motor. But there were a couple of other valves in it, and since there was not much else I could do, I decided to back-engineer the manifold, and figure out what the other valves were for.
I believe the design and execution of the manifold are proprietary, which is why I should not publish any pictures of it, but I am pretty sure that I can disclose the hydraulic principle of what I found because this is a system that is used in a lot of drilling rigs.
Here's the basic arrangement of the valves inside the motor manifold (note that the pressure valve is a sequence valve, not a relief):
Any ideas yet on why there's a sequence valve connected to the reverse direction line of the motor?
If you don't have experience with drilling rigs, you may have a hard time figuring out the purpose of this manifold. Don't worry, I will explain it all, and you will see the “why” of the locking motor in a minute.
So, the drilling is done with a long string of rods that are threaded one into another, and as you go in and out with your drill string, you need to constantly make and break the threaded connections. And the rod threads (like all threads in the world) are sometimes hard to undo. The hydraulic motor, even at its max displacement and max pressure, can't provide enough torque. In some severe “thread locking” cases it is not unusual to see a couple of fitters hanging on a pipe wrench with another one hammering away at the pipe (or torching it) to undo a particularly stubborn thread.
This is why, ever since the first drilling rod got stuck in another one to the point it had to be cut out with a torch, rig designers have been inventing ingenious solutions to help break such threads. And this rig had one as well.
The mechanics of a drilling head is simple - it's a single-stage gear train - a driving gear on the motor shaft, and a driven gear on the drilling chuck. And for the rod-breaking “boost” purposes, that head had a compact two-cylinder system that would engage a small rack into the driven gear to give it tons of extra torque for a tiny amount of travel. Something like that:
I bet now you see where things are headed. But I'll explain anyway.
The sequence valve in the reverse direction line was feeding the “torque boosting” system. When a certain pressure threshold was reached - the valve would open and send oil to the two cylinders, one would push the rack into the driven gear, and the other would turn it. There was a third cylinder, always pressurized, which took care of the return spring function, but I didn't put it in the drawing for the sake of simplicity.
Like in any respectable sequence valve system, there was also a check valve to allow for an easy return flow from the sequenced actuator, and this is how its spring looked:
Yes, sir! The spring broke in several pieces, and its debris eventually locked the check valve poppet in the open position, which caused the torque boosting system to engage immediately after the reverse rotation was activated, locking the driven gear with the rack!
But how about the increased drain flow? Well - there was another malfunction. I believe that it may have happened sometime before the check valve locked. As you can see on the diagram, there was supposed to be a small orifice venting the boost system to the drain line, and it was missing as well. So not only the reverse direction line was immediately connected to the boost system, but it was also immediately vented to the drain through a pretty large 5mm wide hole. The drain hose ran from the motor to the manifold, and then there was a common drain line running from the manifold to the rig - and it was there where the mechanic would detect the tons of drain flow coming out of the “motor”!
Here are the malfunctions:
It was an easy fix in the end. Install an orifice and replace the spring - good to go. But since probably not all hydraulic techs are used to working with drilling systems and rod-breaking arrangements, I thought it would be a nice story to share.
Some points to outline: