Insane Hydraulics
Bold and audacious blog about fluid power
In this article, I would like to discuss two troubleshooting episodes that involved classic closed-center load-sensing systems. The two cases are especially interesting when put side by side, because both of the malfunctions had the same symptom of insufficient function flow, and were diagnosed similarly, but had completely different causes.
A forest crane, equipped with a closed-center LS system, was getting slower as the oil temperature rose. The malfunction was initially attributed to the pump wear, however, after the consequent overhaul, the symptoms remained unchanged. Further analysis revealed a faulty pressure-reducing valve that was supplying pilot pressure to the proportional solenoids of the crane's DCV. As the oil temperature increased - the pilot pressure decreased, resulting in insufficient spool travel. Therefore the low speed (i.e. insufficient oil flow) was being caused by the normal response of the pump's load-sensing control to the flow constriction in the DCV.
The malfunction was discovered by monitoring the delta P between the pump outlet and the LS line - which was constant and equal to the preset delta P at all times, which in turn meant that the pump control was working perfectly fine.
Had the insufficient flow been caused by the pump's low efficiency, the delta P reading would fall below the delta P setting (due to flow reduction caused by the excessive leakage in the pump, and not by the correct response of its control module), and also - overheating would most likely be present as well.
Equipment "in question" - an Atlas Copco core drilling rig, running a Rexroth A10VO pump with a standard DFR (LS plus PC) control. The rig was powered by an electric motor, and the assistance call was triggered by an excessive current consumption that was causing periodic shut-downs of the electric board.
Initial tests revealed that the excessive power draw was caused by the load-sensing regulator of the pump that was set to an unusually high value (above 80 bar). It is interesting to note that even though the gigantic delta P was causing a lot of energy go to waste, no overheating symptoms were present, because of the oversized heat exchanger and the unlimited supply of cold water at the drill site.
The pump was disconnected and flow tested and the test showed nominal efficiency and good controllability, however when the delta P was set to a normal value (around 20 bar) and the unit was reconnected to the system - something strange happened. Whenever the pressure reached a certain level, the flow would start to decrease. The current consumption, on the other hand, would stay almost at the same level - the hydraulic system behaved as if the pump had been equipped with a torque limiter, the only thing was... it wasn't! It also became clear why the load-sensing delta P setting had been set to such an unusually high level. When this initial symptom (the reduction of the drilling head speed with the increase of pressure) emerged, the problem was "amended" by sky-rocketing the delta P (most likely the first adjustment screw the "fiddler" was lucky to "fiddle with"). Just like in the first case, reading the pressure differential between the pump outlet and the LS line (near the pump control module) showed that a constant delta P of 20 bar was present at all times.
As you can see - the two episodes have similarities. In both cases, we have closed center load sensing systems, which use the same type of pump. In both cases, the main symptom is insufficient actuator speed, caused by insufficient flow. And in both cases, we read a stable delta P between the pump outlet and the LS line.
It would seem that in the second case, just like in the first one, one would need to look for a strangulation between the pump outlet and the place the LS signal "comes from" - and it was looked for and was not found. Everything was OK - the hose and the pressure filter at the pump outlet were introducing virtually no pressure drop, the DCV spool made the correct travel, and would maintain its position - yet the flow would steadily decrease with the pressure rise...
Before going into the causes of this malfunction, let us consider the classic closed center load sensing pump control for a minute. It is natural to think that it controls the flow, and it does, but I prefer thinking that it actually does not! Not directly anyway. In my head the classic load sensing control is nothing but a pressure control, that simply tries to keep the pump outlet pressure X bar above the pressure fed to the LS port! The math is simple - feed the LS port with five bar, and you get five plus X at the outlet. Feed it with a hundred - and you get a hundred plus X, or less, of course, if there's no pressure demand from the system, as in the end it is the actuator's resistance to oil flow that defines the pressure in the circuit, but - not more! Then we use this quality to keep the flow at a required level, by introducing a restriction (in this case a DCV spool), and feeding the LS port with the pressure signal taken downstream the restriction. Maintaining a steady delta P across a fixed restriction equals stable flow, change the restriction - and the flow will also change, as the control will try to maintain the pressure differential. Of course, when the restriction is wide open, and the pump hits its maximum displacement - the delta P drops, but whenever the flow is being modulated by the control - the set pressure differential is present and constant.
Now, having said that, let us look again at our drilling rig - What is happening?
- Drill motor speed steadily decreases as the system pressure rises.
Now, let us answer the question "What causes what?"
Therefore - since we already saw that the DCV spool in its working position, the pressure filter, and the pump pressure hose introduce restriction that remains unchanged, the only explanation for what is happening is the increase of the pressure drop in the LS line between the DCV and the pump control, occurring with the rise of the system pressure. How can this be possible? The right question here would be - what can create a pressure drop that increases along with the pressure? The answer is - oil flow through a choke point! In our case, it means that somewhere in the LS line there is an oil leak to tank, which results in a pressure-dependent oil flow, and the pressure drop in the line, created by this flow is adding to the pressure drop already existing between the pump outlet and the DCV LS port. When the sum of the two overcomes the preset delta P - the pump control starts cutting the displacement, as it should.
Allow me to "lift the covers" now. The system was equipped with a remote pressure control valve, connected to the LS line. The cartridge seals were completely gone, which created a significant oil leak between the LS line and the drain - causing the above symptom. This schematic explains how this was happening.
In the end, the seals were replaced, and the rig drilled happily and problem-free ever after!