In January of 2018 I may have created the ugliest makeshift hydraulic system on the planet, but since it has worked up to this day, I believe I've had my redemption...
I want to use the following story to:
a) stress the importance of having a detailed understanding of how an old hydraulic system operates before you re-design it with modern components
b) show that under certain circumstances using high precision instruments like angle grinders on valve spools can actually be not a terrible idea
c) give praise to the most ingenious hydraulic system in the world - namely the control valve arrangement of the Atlas Copco Diamec 250 core drilling rig
Once upon a time a copper mine had an old Diamec 250 diamond core drilling rig winch was old, rusty and leaky, and the controls didn't seem to work that well any more, so the production called a hydraulic technician from a respectable company to investigate for possible solutions.
As the tech saw the operation of the aged three spool control valve, and heard lots of hissing noises inside it, he deemed it worn out beyond repair and suggested replacing the original components with their modern equivalents. An honest standard path anyone would take under the circumstances.
The re-design task was handed over to the engineering department, which inspected the hydraulic diagram and the original three spool control panel (something very similar to this), and after deliberating for a while decided to improve and simplify the outdated design by replacing the original controls with a three spool Danfoss PVG32 valve with manual controls and friction detents. Again - nothing alarming so far.
When the renewed panel was delivered, however, the operators complained that it was impossible to work with, to which the supplier replied that the system was "within the quoted specs" and although "slightly different" from the original, still perfectly capable of delivering the same amount of function control as the old one, and the only thing the operators would have to do was learning the new way to operate it and that once they learned the new procedures the work would flow as easily as it had before.
The guys honestly tried to use and master the new system for a couple of weeks, but it didn't go that well, and since the production numbers were steadily going down, my opinion was asked if there was something that could be done with the hydraulic system to make it "usable" again.
So, I went and saw the cutting edge system, and really appreciated all of the engineering work that went into it, and then after changing a couple of hoses around and adding a 50 buck valve, craftily installing it in the most durable way you can fasten a hydraulic valve in an industrial system (cable zip ties, of course!) - the system was running again as it should, and the small rig drilled happily ever after with now very ugly, but fully functional, hydraulics.
I am not criticizing anyone here. I perfectly understand how an engineer who has never seen a core drilling rig operate before can suggest a technically sound hydraulic solution that is not "workable". I happened to have an advantage in that particular case simply because I had worked with clients with core drilling operations for years, and therefore had a decent idea of how a core drilling rig operated.
Let us dive into the technical details. You can see the schematics of the original system here and here. In fact, comparing both diagrams is a good exercise on its own, because it allows you to see how different designers can address elaboration of a hydraulic diagram in two distinct ways. This is a simple system, but I can assure that it accomplishes a lot! For those who don't know how a conventional core drilling rig operates I explain:
The objective of a core drilling rig is to take rock samples by drilling into the rock with a hollow drill bit attached to a pipe and then pulling out the resulting center cores. As the hole becomes deeper, more pipes are screwed one onto another creating the so called drill string. The length of the inner tube that holds the sample core is less than two meters (for our small rig), and every time it's filled you have to pull the string out, undoing (breaking) each and every pipe thread in the process, take the core sample out, and than insert the pipes back into the hole again, one by one. As you can imagine, after a certain depth, making and breaking the drill string in order to obtain samples becomes a lengthy process, and the more efficient your machine is in doing it, the faster the drilling process becomes. You'll never find these small rigs do deep holes, but even a 60m hole with 3 meter pipes means 40 makes and breaks when you pull the string out and put it back in.
Mechanically speaking a small core drilling machine is simple - you have a hydraulic motor to turn the drilling head, you have a so-called chuck inside the drilling head, which grabs the pipe when pressurized - and for all intents and purposes can be considered as a single acting cylinder with spring return. Then you have the rod holder in the bottom, which is nothing but a single acting cylinder that opens a spring applied clamp keeping the drill string from falling down into the hole. And then you have a double acting feed cylinder that moves the head up and down. That's it - four hydraulic actuators, one rotary and three linear. Quite simple right?
I am not going to discuss the drilling part here, I want to concentrate on the drill string running part, which was the main complaint of the operators. If you consider the running of the drill string up and down the hole, you realize the two main challenges for the hydraulic system. First - you have to make sure that you time the opening and closing of the chuck and the rod holder in a manner that when you move the drilling head up and down it pulls the string out of the hole or pushes it into the hole without loosing it. And second - when you operate the drilling head motor to make or break a thread - you have to make sure the rod holder is holding the string tight (contrary to what happens during drilling - when you need the clamp to be open).
To assure this function you have to inject the oil in both cylinders (the chuck and the rod holder) when you move the feed cylinder in one of the directions only (depending on whether you are pulling the string out or pushing it back in), and then inject the oil into the chuck alone when you're turning the motor. If the sequence is still kind of hard to grasp - picture your self holding a drill string with your two hands, one of them being the chuck and the other the rod holder. Now picture a sequence of moves that you would have to do if you had to pull the string up by moving just the upper hand up and down.
Atlas Copco did it all with a single orifice and the third directional valve section, which they cunningly turned into a 3/3 directional valve, with middle position closed. Thus, when the spool is in one of the active positions, the oil from one side of the feed cylinder is injected into both the rod holder and the chuck, thus freeing the drill string to be pulled or pushed as the head is moving up or down. And when the rotation is engaged, the oil is injected in the chuck line, but since there's an orifice in the rod holder line - the pressure can't rise too high to open the rod holder because it's drained through the three way valve and the open center feed section spool to tank. By the way - this is the reason for the hissing noise.
I, personally, love everything about this solution - the way how they transformed a 4 way directional spool into a three way three position valve is pure genius. They simply tapped into the cavity where you would normally find a load drop check valve in a "normal" directional valve section, and ran an external connection from it. Very clever!
With the "original and old" design all the operator had to do was use the third lever to select if he wanted to run the pipes in or out, and then operate the up/down lever to move the string in the desired direction, and use the rotation control to break or make the pipe threads.
In case of the the "improved modern version" the operator now had to control the rod holder and the chuck manually with the third lever. Which is perfectly fine, when you have to run one or two pipes per hour, but becomes impossible to work with when you have something like 50 pipes down the hole. Apart for having to constantly be thinking about the right sequence, all it took was a moment of distraction to send the drill siring flying down the hole...
I can't say that I invented anything new here to solve the problem - I simply copied the original schematics and installed an external 3/3 valve that made the operation automatic. The only thing that I needed for the system to operate properly was a "real" open center spool, and the PVG32 open center spools are actually quite closed. They are OK for a a piloted check or an overcenter valve, but weren't open enough for our case. Again - the "makeshift" solution was simple - when you have a closed center spool and need an open center what do you do? You grind the extra material away with a grinding wheel and a steady hand! So long as you don't become enthusiastic and grid away too much spool - you're OK! Definitely not pretty, but absolutely functional!
Actually there was one more thing the new system didn't get right, and again this stems from the fact they didn't have "drilling knowledge". There's a chuck delay valve that delays chuck closure when the drilling head is operated in the left hand rotation. This is needed to help breaking tight threads - the drilling head gains some speed before biting into the pipe - another ingenious solution done purely with hydralics. The problem was solved with a simple sequence valve, and again - I don't blame the authors - they were asked to design a system that can move four hydraulic actuators, they sent a quote for a system that can move four hydraulic actuators, and someone approved the quote for a system that can move four hydraulic actuators.
Be it as it may - this is a perfect example of how a seemingly simple circuit can ingeniously perform multiple functions and how easy it is to overlook some of them when you don't know all of the details of a system's operation. The best way, of course, is to see a machine operate or, even better, operate one yourself for a couple of shifts before attempting a re-design.
I wish all of the hydraulic problems I come across could be solved with a 50 buck valve and a couple of zip ties...