Insane Hydraulics
Bold and audacious blog about fluid power
There are two (extremely subjective) concepts that always spring to my mind whenever I receive a hydraulic cylinder labeled as creeping. I named this post "Drifting Cylinder Classics" because they represent situations that I see often: "The Incomplete Tale" and the "Piston Seal Paradox." I'm not even sure if the word "concept" is the right word here, but I want to try and explain what I mean anyway (at the risk of sounding like a total weirdo), and I will do it with the help of this stabilizer cylinder from a Caterpillar TH360B telehandler, which was just "cold-dropped" at our shop with a note "Leaks and Drifts":
As you can see, this is a pretty robust double-acting hydraulic cylinder with a flanged counterbalance valve.
Unfortunately, the person who knows the exact symptoms is unreachable at the moment, but the large oil puddle means the rod seal is busted. The fact that the paint on the gland end was stripped and, by the looks of it, the grooves on the gland nut were smacked pretty hard indicates someone attempted to open this cylinder before sending it to us. All I can say is that I feel their pain - I know first-hand how stubborn these can be, especially when you don't have the right tools. But the leak means I am opening it no matter what, and so, today we'll deal with my (weird) theories and the CAT's "super-protected" counterbalance valve, and next week, when I get to the disassembly, I'll show you how you can make a pretty capable wrench for such glands from scrap.
As always, let's begin with gathering reference materials - in other words, let's google for "Caterpillar TH320B hydraulic diagram, service manual, parts manual..." - you know, the usual. I did that and found this hydraulic schematic, in which, alas, in a very Caterpillar fashion, there was no mention of the settings for the counterbalance valve. The only thing it had was a part number: 204-6924.
A pretty subtle way of saying - "Something wrong with your valve? Get a new one!" Let us remove and inspect it to see what we are dealing with:
Apparently, this is an Oil Control part, coded 08352703563501B. Of course, this is a non-standard reference, and despite being represented in the hydraulic schematic as an adjustable valve, you can clearly see that it has no adjusting screws, which means it has fixed settings.
OEM model or not, if you have at least some experience with Oil Control and you don't mind spending a couple of minutes going over their range of dual counterbalance valves, you will quickly discover that this is, essentially, a 08.44.11 (A-VBSO-DE-30-FC2) with an asymmetric mounting hole pattern and one of the V ports on a different side. You can easily recognize it by the absence of separate check valves, which are present on other models, and, of course, by the dimensions of the body.
Now, fixed settings mean "all you need to know is that it works", and the asymmetrical hole pattern means that the settings of the two valves most likely differ, which makes total sense for cylinders that are supposed to hold their primary load in a single direction - i.e. stabilizers, outriggers, etc... We can also safely assume that the "03" after the first six digits stands for the 4.2:1 pilot ratio, and the customary "35" stands for the adjustable pressure range of 100 - 350 bar (1450 - 5000 PSI).
Now that I mentioned the load-holding, I want to address the first "drifting cylinder classic" - "The Incomplete Tale" - which refers to the information that you usually get when such a cylinder is delivered.
Let me explain what I mean by that. Whenever you have a double-acting cylinder with a label "it just drifts" - you should always confirm with folks who actually saw it happen which way exactly it was drifting, and never assume that it was drifting in its primary load-holding direction.
For example, in our case, it would be natural to assume that the "declared drift" was happening in the closing direction (these cylinders extend to lower the stabilizers and lift the front end of the handler in the air), but it should never be taken for granted. When the stabilizers are raised, they are supported by the rod-end chambers, and a major leak through a rod seal, for example, would be enough for a stabilizer to drift down, extending the cylinder. Very often, a person who relays the "it creeps" message does not know the direction detail, which is crucial for correct diagnostics. In our case, for example, if I knew for sure that the main complaint was "stabilizers dropping down when raised" and "oil leaking out," I would say that most likely it's the leaking rod seal. If, however, I was told that the machine could not hold itself in the air, I would have to look into other causes.
Unfortunately, the tale of this cylinder is, indeed, a classic incomplete, at least for now (I should be able to get an update next week and I'll make sure to report if I do), which means I'll be checking everything. Since we discovered the model of the counterbalance valve, we can check out its construction, conveniently provided by the Rexroth catalog:
Beautiful, is it not? I especially like the pointy shuttle in the middle that "switches" (for lack of a better word) the pilot pressure. The wide lines on the drawing make it look like a tight piston, but it is not - it is a loose shuttle that seals against the holes in the noses of the two poppets. You can also see all the potential leak paths: the poppet/seat interfaces, the seals around the check valve sleeves, the seals at the pilot ends of the poppets. Any wear or damage in these areas could cause a leak and, consecutively, the creeping. So, let's take it apart and inspect:
All's good here - all the seals are intact, and the poppets and seats look brand new. Now you can clearly see the difference in the settings of the two sides - while the rod end side has a 1.6 mm stack of washers, the blind end side has a 4 mm one - and that's a big difference! Let us put it back together again and see if we can figure out the settings using an interface plate improvised from a piece of flat bar and a lever-operated pump (you can read more about using hand-pumps as diagnostic tools here):
Testing showed that the low-pressure (rod) side cracks at 95 - 100 bar, and the high-pressure one at 345 - 350 bar. I must say that using a lever-operated pump for such tests is extremely rewarding. I can pump the valve to 350 bar and keep this pressure up without any noise, and I can count the drops per minute - in other words, perform the ever-so-dreaded open-port test - in absolute safety. Once again, I want to point out the importance of using a hand pump with a significant displacement for this purpose. In my case, it's a 12 cc pump, and it requires substantial physical effort to get to 350 bar even with a meter-long lever, but it also provides a strong feedback and a relatively high flow rate, allowing you to determine the cracking pressure and even the override with relative ease and consistency. You can't possibly achieve this with a low-displacement porta-power because of the lack of proper muscle feedback. Believe me, you have to try it for yourself to see what I am talking about.
So, what does this mean for our cylinder? Well, the fact that the rod side counterbalance is set to the relatively low 95 bar means that if there's an internal leak through the piston seal, given the rod is 55 mm in diameter, you would only require mere 23 kN (2.3 ton) to begin closing it! This is nothing - especially considering the mechanical advantage provided by the geometry of the stabilizer. Essentially, this means that an internal leak through the piston seal would cause this load-holding arrangement to creep even under the weight of an unloaded machine. And this, my friends, is the place where I reveal my second "drifting cylinder classic" - "The Piston Seal Paradox."
The paradox lies in the fact that the exact same phrase used to describe the exact same situation may or may not be correct, depending on who says it.
I admit that this statement is subjective to my person, but I want to let it out anyway - so, here it goes:
In my head, when a person not versed in hydraulics says that a "cylinder is creeping because of a leaking piston seal" - he is not correct because his intuitive mental image is incomplete. All he's imagining is the hydraulic fluid flowing past the damaged seal, but he's utterly unaware of the load-holding valves, and he skips the fact that when the rod goes in, the oil must go out. Now, when an experienced hydraulic tech states exactly the same thing - i.e., "this cylinder is creeping in because of a leaking piston seal" - he is correct because he fully realizes that the leak transformed the double-acting piston into a ram, amplifying the load-induced pressure and causing a counterbalance valve to open. Maybe we should start using a secret signal when we say this phrase? Like, "It's a piston seal leak for sure!" - wink-wink... or, maybe, a double tap on the nose?
Anyway, I haven't opened the cylinder yet, but following the "tradition", I can quickly test it with compressed air to see if I can detect anything major:
Indeed, when I injected compressed air into the tail end (with the rod fully extended) - I saw an infrequent but steady stream of air bubbles appear from the rod-end hole. Hmm... Did I find the bypass? I wouldn't be so sure. Low-pressure compressed air is different from hydraulic oil, and I know that such cylinders often use segmented piston seals (seals that have a stepped cut for easy assembly). Such seals, theoretically, can bypass low-pressure air. The bubbles I am seeing right now are small and infrequent, so I am postponing my "piston seal verdict" until I have the cylinder disassembled and the seal physically inspected.
So, I leave you with this cliffhanger until next Sunday, and I sincerely hope that today's blog made at least a little bit of sense.