Insane Hydraulics
Bold and audacious blog about fluid power
If you use restrictors as pressure-inducing elements in your hydraulic tests, you are probably used to solutions that look like this:
Common needle valves are, indeed, pretty good at throttling flow at will, but if you conducted enough tests at "serious" pressures and flow rates - i.e. tests that would require needle valves with ports equal to or greater than 3/4'' BSP - you also know that operating such valves above certain pressure requires significant muscular input to the regulating knob - often to the point when you need both hands or even a tool to turn it. This, of course, stems from the construction of a typical needle valve:
Pressure will always "push" the sealed popped out, and on larger valves, the unbalanced area exposed to the test pressure can be quite large - hence the increased axial force on the threads and, consecutively, the large torque required to turn the knob.
This is perfectly OK for a valve that is set once in a lifetime, but not ideal, to say the least, for a valve that is constantly being adjusted. But if you ever got to play with a "proper" loading restrictor, like for example, the one you would find in a Webtec hydraulic multi-meter - you know what a "butter-smooth" needle valve feels like. But do you know how these valves differ from "normal" needle valves? Well... I don't! But I sure as hell am going to find out because a pretty battered Webtec DHM403 hydraulic flow tester just fell in my lap:
The board and, most likely, the pressure sensor are shot, so we will be sending the meter to Webtec for an overhaul and calibration service, and since it can't possibly be broken more than it already is, the client authorized me to use it for "educational" purposes - and I want to start this back-engineering session with the loading valve. While Webtec brochures advertise that these valves give "smooth progressive pressure control in both flow directions" and "feature unique built-in Interpass® safety protection system", they never published a single decent cutaway view of their "invention" - and I, as a hydraulic technician cursed with curiosity, need to know what's inside of things I come about!
Let us begin with inspecting the patented protection system, based on thin burst discs that are supposed to reliably rupture above a certain pressure:
The burst discs sit under the hex plug threaded into the "nose" of the throttle valve - you can see in the outlet port of the tester (the tester is, actually, bi-directional, so calling this port an outlet port is technically incorrect, but the manual does state that it has a preferred direction of flow which provides better accuracy). Quite a lot of our clients use such testers, and, believe it or not, even with the instructions written on the lid and all, I often find myself having one of the following conversations:
- Oh my God! Our tester broke!
- No, your tester did not "suddenly break" - it was the safety disc that ruptured!
- Oh my God! Now what? Where will I find the new safety disc?
- Spare discs are stored under the slotted cap at the bottom of the meter - just screw it out and look inside!
- Oh my God! We installed a new disc but it broke again!
- It's not a disc - it's two discs!
There are two burst discs "hugging" a hollow spacer - this construction makes sure that a disc is always forced in the same direction. Discs get deformed by pressure - so I imagine a single disc would "fatigue out" in a second in a bi-directional tester, which is why they (cleverly) used two.
There are a couple of things I need to mention in regards to the very nice official video that shows how to replace the burst discs. First - at 2:47 they mention neutron-meters, and I would love to know what "neutron-meters" are, and second - they are tightening the plug with the throttle valve in the flow-tester - and I, personally, don't like it. The throttle valve is prevented from rotating by a pretty flimsy screw that works in the slot machined into the side of it:
When you are loosening and tightening the disc cap in place - you are putting unnecessary stress on it. I believe the best way is to remove the complete throttle valve by undoing the four screws that secure it in place and then work on it in a vice with soft jaws.
And now, let us get back to the important part - the "butter-smooth throttle valve":
As it turns out - the valve is super easy to disassemble - all you need to do is remove the plastic handle by pressing out the spring pin, and then undo the nut that secures the bearing assembly in place - and voilá - it's all in bits and pieces:
First impressions - the two very capable thrust roller bearings and a high-quality pressure seal (10x17.3x3.2) are obviously major contributors to the smooth operation of the valve. And, by the way, the five-pointed handle is very ergonomic - much better than those skinny knobs that never offer enough grip. Obviously, there's some balancing going on here because there's a hole that connects the nose end to the tail end of the throttle, and the tail end is sealed with a nice PTFE/bronze + NBR piston seal (I didn't remove it, but judging from the external dimensions I would bet on 28x20.5x3.2):
But in order to see the whole picture I needed to do better than that! So I thought - "I should spend a few minutes to make a quick sketch that will help me explain the valve's function" - but then I got carried away, and somehow ended up modeling the entire tester. At least I can show you some detail now, and who doesn't like a good 3d rendering, right?
So, this would be the throttle piece with the said safety discs:
The two section views made at different angles show the channels that lead to the burst discs and the connection between the nose and the tail.
Here is the complete assembly with the knob, the shaft, and the thrust bearings:
This two-plane cutaway view is perfect because it reveals both the safety disc channels and the connection between the nose and the tail.
And now, let us insert this assembly into the body to finally see what is going on:
Don't you love it when things fall into place? So - here's how this loading valve operates:
First, let us forget about the safety discs and the respective channels - they do nothing during normal operation. The only things that matter are these dimensions and the channel that connects the nose to the tail (the channel is not shown in this sectional view):
If you bear in mind the connection between the nose and the tail - you will see that when the oil flows from the side to the bottom of the throttle valve (the recommended direction) - the area that is pushing the valve out is the difference between the areas with the diameters of 28 mm (tail end) and 27 mm (nose end), which amounts to roughly 0.4 cm² - and this force is supported by the outer thrust bearing. When the direction of the flow is reversed - the throttle piece is being pushed in the closing direction by the same area differential, but the pressure in the tail section also pushes the 10-mm shaft out (area of 0.8 cm²), so the net force supported by the outer thrust bearing is still 0.4 cm² times the pressure.
Thus you get a valve with a wide throttling section but with a relatively small unbalanced area affected by the pressure, even when it is fully closed. I actually think the area differential was purposefully engineered into this valve to provide some positive resistance to pressure increase and locking of the adjusting knob. Pretty neat design, don't you agree?