We already know that Komatsu's excavator hydraulic system from Dash 6 onwards is designed according to the CLSS principle. The very special point of this system is that the main distributor has pressure compensation valves to "divide the flow" when the machine performs many operations at the same time. The other manufacturers, such as Caterpillar, John Deere, Hitachi, Kobelco, Volvo, Doosan...they do not use pressure compensation valves for the main distributor valve (main control valve), so how do they solve the "big problem of controllability": the flow flows all to the low-pressure action branch. (for example: lifting the BOOM while swinging): Hydraulic oil will always take the path of least resistance. If the load of the swing is lighter (for example, 50 bar) and the load of the BOOM is heavier (for example, 200 bar), all (or most) of the pump flow will "run" to the swing path. Result: The swing moves very fast, while the heavy BOOM moves slowly, jerkily, or doesn't move at all. They solved the "flow sharing" problem differently: they did not use a pressure-compensation valve for each operation. Instead, they used very complex "flow compensation grooves" right on the main spool. When you move the two control levers, these grooves automatically balance the pressure and flow between the two functions. It is not "perfect" like Post-comp or Pre-comp, but it is much cheaper. I'll stop here for now because I'm afraid that if I write too long, you'll be too lazy to read. (to be continued).
Please continue! I love detailed and in depth explanations
The truth is: All major manufacturers (Cat, John Deere, Hitachi, Kobelco, Volvo, Doosan...) have to deal with the "flow sharing" problem, and they DO use pressure-compensating valves. The main difference is not "to use or not to use", but the type of valve they use and the installation location. Komatsu (with CLSS) uses one architecture. Most of the rest (usually using components from Bosch Rexroth, Kawasaki, Eaton...) use another "rival" architecture. You will never find the term "flow compensating groove" in mainstream hydraulic textbooks. The reason is because it is not an academic concept or a standard component like "Pressure Compensator". Instead, it is an "engineering trick" or a "design philosophy" of the manufacturers. Details of "Spool Profiling / Metering Notches". 1. What is the correct name? If translated from the manufacturer's technical documentation, it is often called "Metering Notches", "Tapered Spool" or "Spool Profiling". I use the term "flow compensation groove" to describe exactly what it does, to help you visualise what it is "fighting" (which is "flow sharing"). 2. What is it? Where is it located? It is not an additional sub-valve. It is the physical shape of the main spool – the spool that your pilot hand pushes back and forth. Instead of turning a simple "step" for the oil to pass through (like an on/off switch), manufacturers mill/turn these grooves to extremely complex shapes: • Can be V-notch. • Can be U-notch. • Could be cross-drilled holes. • Can be tapered grooves. non-linear relationship between the slider travel (how much you push the control lever) and the actual oil flow (how much oil passes through). 3. How does it solve the "flow sharing" problem? (This is the "magic" part) Let's go back to Scenario: Squeeze 100% BOOM (heavy, 200 bar) and 100% SWING (light, 50 bar). LS System (Komatsu): Use the Pressure Compensation Valve (PC Valve) to "command" the SWING valve (light) to "tighten", giving oil to the BOOM. This is an "Active" solution. Compensation Groove System (Hitachi, Kobelco...): There is no separate pressure compensation valve. It solves in a "Passive" way, based on the design of the slider itself: 1. Light Load "Squeeze" Design: The notch on the SWING (light) slider is intentionally designed so that it is susceptible to low pressure. 2. When the system is running, the pump pressure (Ppump) will increase to serve BOOM (heavy)—for example, Ppump = 220 bar. 3. At the BOOM valve (heavy): o Pressure drop across the spool is: ΔP = Ppump – Pload BOOM = 220 - 200 = 20 bar. o With a small Delta P (20 bar), the groove geometry allows a flow rate Q1 to pass through (e.g. 100 L/min). 4. At SWING valve (mild): o The pressure drop across the spool is: ΔP = Ppump – Pload SWING = 220 - 50 = 170 bar. o With a very large ΔP (170 bar), the oil will tend to "break through" and flow through here. o This is where the "magic" happens: The SWING slider's V-groove is designed so that when ΔP (pressure drop) exceeds a certain threshold, its ability to let oil through (flow coefficient) will no longer increase linearly, but will "squeeze" back. o It acts as a passive "Pressure-Dependent Variable Orifice". It "tightens" itself, allowing only a flow rate of Q2 (e.g. 100 L/min) to pass, even though the pressure drop is very large. 5. Result: o Q1 (BOOM) = 100 L/min. o Q2 (SWING) = 100 L/min. o Total flow = 200 L/min (assuming this is the max flow of the pump). o The oil was "shared" passively, simply by the geometric design of the two spools. Summary: • "Flow compensation notch" is a practical (not academic) descriptive term for "Metering Notches" on the main slide. • inexpensive, passive, mechanical solution to replace expensive, complex pressure compensating valves (PC valves). • It works by designing the slider profile in a non-linear manner, causing the light load functions to automatically "squeeze" the flow when the pressure drop across them gets too large (during combined operation). • This is the "secret" of the Control systems of Hitachi, Kobelco, Sumitomo... before they completely switched to electronic LS. (to be continued).
Wow, little did i know of spool notch geometry but i was unaware that it can also eliminates entirely additional flow distribution valves. I came to light when CNH/Case made this paper: https://patents.google.com/patent/US6450194B1/en but i was unaware of this. Thank you for sharing, you packed my weekend for research. Please continue. Also one question as you said [The SWING slider's V-groove is designed so that when ΔP (pressure drop) exceeds a certain threshold, its ability to let oil through (flow coefficient) will no longer increase linearly, but will "squeeze" back.] meaning if let's say boom circuit have relief setting of 250 bar and if machine needs to go beyond let's say 300 bar, then spool grove geometry also need an update?
Thank you for very interesting write up! Do you know anything about these valves on older machines with open center systems? Do they also use valves with cutouts etc?
All right, friend. I have to explain your question at length. Your question was calculated and designed by the manufacturer from the beginning. The most expensive, most secretive, and "soul" element of an excavator: Application Engineering . An engineer cannot design "in general". They are required to follow the steps below exactly. Here are the detailed factors they have to take care of: 1. Duty Cycle Analysis The first step is: Hitachi, Kobelco... engineers will attach hundreds of sensors to the machine and give it to the best operators (in Japan, the US, and Europe) to "torture" the machine for thousands of hours. They collect data on: Frequency of coordination: Which operations always go together? "Dig deep": Boom-up + Arm-in + Bucket-curl. This is the classic trio. "Unload": Swing + Boom-up + (Sometimes) Arm-out. "levelling": Boom-down + Arm-out + (Sometimes) Bucket. Typical Loads: When combining "Digging Deep", what is the Boom (lift) load, what is the Arm (pull) load? Speed Requirement: When "Unloading", does the operator want to swing the machine fast or raise the Boom Faster? (Usually, both should be smooth). 2. "Spool Profile Customisation" From the above data, they will design the profile of EACH spool DIFFERENTLY . A control valve set does not have two identical spools (unless they have two identical functions, for example, two moving motors). A. Designed for "Dig-Up" function (Boom-Up + Arm-In) This is the most profitable combination of mining operations. Problem: Boom-up (lift) is usually a heavy load (high pressure). Arm-in (pull) is a lighter load (low pressure). Consequence (if the design is stupid): The oil will "rob" all the way to the Arm-in, causing the arm to retract very quickly, but the boom cannot rise (or jerk). Solution (V-notch design): Arm-in Slider: Designed with a V-notch with a "variable-orifice" effect that is extremely sensitive to ΔP (pressure drop). When ΔP through it increases rapidly (because the pump pressure must increase to overcome the Boom load), this notch will automatically "squeeze" very strongly, slowing down the Q Arm flow. Boom-up slider: Designed with a more "tough" V-notch. Its "variable throttling" effect is less sensitive to ΔP. It is "favoured" to receive a larger Q Boom flow even when the ΔP through it is low (because its P load is already high). Engineers calculate (by CFD ( C omputational F luid D ynamics) and testing) that when the operator pushes 100% of these two arms, the QBoom / QArm ratio must be at a "golden" number (e.g. 60%/40%) for the machine to have a smooth and powerful digging motion. B. Design for "Swing" function The most special function because it is a motor (with inertia), not a cylinder (static). Requirement: Must be "smooth" (anti-jerk) when starting and stopping, but must be "priority" (not dead) when coordinating. Solution (V-notch design): Fine Metering: The Swing's V-notch will have a very narrow V-angle at the top. This allows the operator to "roll" the swing arm to rotate the carriage just a few centimetres accurately. Anti-"FOLLOWING": Swing car is usually lightly loaded. When combined (e.g. Swing + Boom-up), it will be "crushed" by the Boom slider (heavy load). The "variable throttle" design of the Swing groove will be very sensitive to the self-brake, giving oil to the Boom. Additional (Important): Because Swing is so important, manufacturers often design an additional swing priority valve or a separate hydraulic logic valve to ensure it always receives a minimum amount of oil, regardless of how the V-notch is "divided". Conclusion: The design of these V-notches was not a single hydraulic problem, but an extremely complex system optimization problem , based on statistics, CFD (Computational Fluid Dynamics), and thousands of hours of real-world testing with operators. It is the unique "quality", the "driving feeling" that makes one operator like Hitachi machines, another like Kobelco machines.
The academic literature on these grooves is not usually focused on the “flow sharing” that I am talking about. They are focused on a more “headache” problem: “Flow Forces” . • What’s the problem? When oil flows through a narrow slot (the groove you see) at very high speeds (hundreds of bars of pressure drop), the flow itself (due to the Bernoulli effect) will create a huge hydrodynamic force pushing the slide – usually pushing the valve closed. • The result: This makes the control lever “heavy” or “jerky”, as the operator (or pilot pressure) has to “fight” with this flow force. • The solution (That’s the groove in the picture): Those intricate grooves, notches, and holes are also cleverly designed to create opposing flows, creating opposing pressure zones to compensate for the “Flow Force”. Therefore, this "compensation groove" is a 2-in-1 "expert": It both meters the flow to "share the oil" passively, and balances the force to help the spool move smoothly and accurately with a very small pilot control force. • On the "OPEN CENTER" circuit in particular or on the "NO PRESSURE COMPENSATION VALVE" circuit in general: The V-notch must take on 2 tasks: 1) Passively share the oil (Flow Share), and 2) Eliminate the Flow Force. • On the CLSS valve: There is already a dedicated Pressure Compensation valve to handle task (1). Therefore, the groove on the spool is still VERY IMPORTANT to perform task (2), which is to eliminate the Flow Force. Without these complex grooves/holes/chamfers, what would happen? 1. "Jerky" valve: Extremely high pressure (Ppump) and large flow (Q) when flowing through the valve edge will create an extremely strong hydrodynamic (Bernoulli) force that "hits" the slider, usually in the direction of closing the valve. 2. "Stiff" valve: Pilot control pressure (e.g. 30-40 bar) will not be strong enough to "overcome" this Flow Force smoothly, causing the control to lose its smoothness. Therefore, the grooves on the spool of the PC200-6 and later (and all machines using the CLSS system) are still there, and they are optimised by engineers for 2 main purposes: 1. Force Balancing: Create reverse flows, opposing pressure zones to "cancel" Flow Force, helping the slider move smoothly and stably. 2. Fine Metering: Create a perfect "flow-stroke" characteristic curve. The V-notches help when the operator just moves the spool slightly; a very small and precise amount of oil flow will pass through. This is what creates the ability to "drive silk", "roll" the bucket centimeter by centimeter. Figuratively speaking: • Spool valve without pressure compensation valve: "I am a multi-tasker, having to divide oil and prevent jerking." • Spool valve with pressure compensation valve (PC valve) of CLSS system: "I am the expert. The PC valve boss takes care of the oil distribution. My job is to make sure everything is as 'smooth' and 'precise' as possible (anti-jerk + fine metering)."
Wow, I had never thought about the forces on the spool created by a big pressure drop/high velocity flow. But ofc when you tell it it makes perfectly sense. There is a lot to it when designing spools, thats for sure.
CLSS= Closed-center Load Sensing System Thank you for great explanations. Us operators only know that this machine is smooth in most functions, this other machine can hog OK but don't try fine things with it, etc. Now there is a possible " why" and also maybe something that could be fixed because it is indeed malfunctioning. Sometimes we just make things work when they are far from optimal.
@LACHAU Is there any comparison available with this spool groove system with 3PC System?
Of course not, my friend! But there are related events with which we can make an indirect comparison. 1/- Compare the working speeds of Komatsu machines of the same weight class and same year of manufacture with other brands like Kobelco, Hitachi, John Deere, etc. 2/- Even Komatsu, with its large machines today, still uses a design without a pressure-compensation valve, but only divides the flow through grooves on the spool when coordinating multiple operations simultaneously.
Let me explain further to make it clearer (although this explanation may be unnecessary for many of you). I used the word "magic" in (4) above, not because it is supernatural, but because it is an extremely clever application of a basic physical principle. It is the "flow coefficient" (Flow Coefficient Cv or Cd ) . Theory (simple): The basic formula is When teaching, it is often assumed that Cv (flow coefficient) is a constant and A (open cross-section) is linear with the spool travel. Reality (The V-groove): A (Cross Section) is non-linear: The V-groove is designed so that when the spool moves for the first 1mm, the A section opens very small (for fine metering), but when it moves for the last 1mm, the A section opens very large. Cv (Flow Coefficient) is not constant: This is the "Tech secret" point. Cv varies a lot depending on the 3D shape of the profile and the Reynolds Number of the flow (which in turn depends on ΔP and oil viscosity). The "magic" I'm talking about is that the manufacturers have actively designed that groove profile (both 3D geometry and surface roughness) to create a desired Q (flow) characteristic curve. They take advantage of the change in Cv when ΔP increases dramatically (causing turbulence and possibly cavitation) to actively slow down (squeeze) the flow of the lighter load. It's not magic, it's materials science and applied fluid dynamics at a very high level. It's a micron-level calculation on that groove, not luck.
This tread does my head in. So on behalf of those who can get their heads around this but haven't responded, thank you LACHAU.
Well. Im aroused
This is quite a bit of useful information! Thanks for posting