Spring Water

How often have you heard the expression “water is incompressible”? No matter how often, it’s never been correct; nothing is incompressible (with the possible exception of subatomic particles, but let’s not get bogged down in theoretical physics). Of course, much folklore often contains an element of truth – water is in fact very stiff! Indeed, water has roughly twice the stiffness of mineral oil, but nowhere near the stiffness of, say, steel.

In fairness, for many applications, water can be assumed ‘incompressible’. In other applications however, fluid stiffness can significantly influence performance (both beneficially and detrimentally).

Here’s a useful rule of thumb: a given mass of water will change volume by about ¼% per thousand psi. I won’t apologise for using archaic units (psi = “pounds per square inch”), because the rule of thumb as it stands is easy to remember, and conversions are simple (simpler for me than remembering an equivalent rule adopting SI units).

So what does the above mean in practice? Where small volumes are involved, not much. However, for typical longwall hydraulic systems, where delivery hose volumes might be several thousand litres, and working pressures are around 350 bar (~5,080 psi), several buckets of water are ready to ‘spring’ out at you if a system breach occurs. The compressed fluid energy stored in longwall pressure hoses often exceeds 500kJ – roughly equivalent to the energy dissipated during a head-on collision between 2 light automobiles, each travelling at 70 kph! Should the breach dimensions be small, the resulting jet of fluid would be much like that used to water-cut concrete and the like. Clearly, being in the way would be bad! The situation only worsens with additional energy released from bottle accumulators, and any further energy added to the system before the pumps shut down due to a low-pressure trip.

Stored energy is the main difference between longwall hydraulic systems and more conventional hydraulic systems, and it’s more the result of the greater fluid volumes than the higher system pressures. The energies would be roughly twice as high if the working fluid was mineral oil instead of water, and the situation would become truly explosive if using a gas.  However, there wouldn’t be any danger at all if water were truly incompressible.

It’s not all bad though. Fluid stiffness ‘softens’ the system response (or perhaps fluid ‘flexibility’ would be a more appropriate description in this case). That is, fluid filled hoses behave like accumulators (albeit accumulators without a pre-charge). As such, systems with short hoses might require additional bottle accumulation to help remove pressure impulses, and/or to service immediate fluid demands before the pumps come on-load. The hose accumulation is mostly due to fluid compressibility, with contributions from hose dilation (and hopefully entrapped air) often negligible. Of course, an overly ‘soft’ system might suffer sluggish performance, so the combination of ‘hose accumulation’ and ‘bottle accumulation’ must be considered.

Another place fluid compressibility influences system performance is within hydraulic cylinders. Again, the influence might be negligible for small cylinder volumes and/or low-pressure demands (like roof support base lift cylinders and flipper cylinders, etc.). However, the fluid required to ‘pressurise’ a leg cylinder to full load can represent a significant proportion of the overall roof support demand. This is largely due to very high pressures in the upper fluid column of telescopic leg cylinders, often exceeding 800 bar (~11,610 psi). The upper column of fluid (trapped above the ‘blipper’ valve) will reduce in volume by ~3% on pressurisation to 800 bar. Dilation of the steel cylinder is negligible, and almost all the change in fluid volume manifests as contraction of the fluid column height; exceeding 30mm at leg cylinder top stage extensions greater than 1m. As such, if cylinder travel of 100mm is sought, then the fluid required for travel and generation of the desired set force probably exceeds 130% of the fluid volume required for travel alone. The additional ‘leg set’ fluid is sometimes overlooked when specifying pump system capacities.

Leg cylinder motion following contact between the roof support structures and the strata is sometimes wrongly attributed to air in the cylinders. While true in some extreme cases, most of the time, fluid compressibility alone accounts for the unexpected cylinder travel.

The elastic energy stored in the fluid ‘spring’ of a set leg cylinder dissipates rapidly when the POCV (pilot operated check valve) opens on initiation of leg lower.   Peak return flows can be several thousand litres per minute. While leg decompression ‘spike’ durations are often less than 1 second, the resulting back-pressure in the return main can result in fluid loss through return hose relief valves, and/or erratic control valve piloting.

A further matter influenced by the fluid system stiffness is the speed at which pressure pulses travel. While the speed is in the order of 1,500m/s, the propagation delays associated with long hoses can cause some very puzzling longwall system behaviours, but I’ll save that fun for a future post(s).   For now, remember that water is not ‘incompressible’, and that in cases like those cited above, its ‘springiness’ can be very important.

Leave a reply