Bench Degree·FLUID POWERchapter

Chapter 25: What You Now Know

The ledger. What Chapter 1 promised, whether it was delivered, what you can now do on a bench, and what this volume left out.

Chapter 1 made six general promises and one specific one. It said that by the end you would be able to walk up to any machine with hoses on it and read its circuit diagram, compute its forces and speeds, say which number the pump sets and which the load sets, say why the designer chose oil or air, point at where the heat is going, and point at the component that could kill you.

Then it made a claim with a number in it: that you would work out how your own foot stops your own car, and that the answer would be about 80 to 1 at each front wheel and about 14,000 N at the road, enough to stop two tons at eight tenths of gravity.

Here is that promise, checked line by line.

Section 1: The Promise, Tested

Read a circuit diagram. Chapter 14. Eight conventions generate the whole language: solid lines carry fluid and long-dashed lines carry orders, circles are rotary machines with a triangle showing which way the fluid goes and whether it is liquid or gas, squares are valves with one square per position, and you read a valve by sliding the row of squares in your head until the ports line up with the position you care about. Springs, levers, solenoids and pilots on the ends of the row tell you what the machine does when the operator lets go, and a normally-closed pressure valve is drawn with its arrow deliberately not lining up. You drew the two-syringe rig, the bottle jack and the reference circuit, and then you found a real diagram on a real machine and answered six questions from it.

Compute a force and a speed. Chapters 5, 6 and 11. Force is pressure times area, and in the units that matter it needs no conversion factor at all: newtons equal megapascals times square millimetres, and pounds equal psi times square inches. Speed is flow divided by area. And a cylinder is not symmetrical, because the rod occupies part of the piston’s face on the return, so the reference cylinder makes 19,630 N at 85 mm/s going out and 14,720 N at 113 mm/s coming back, which is 4,410 lb at 3.3 inches per second against 3,310 lb at 4.5 inches per second, and the two ratios are the same 1.33 the opposite way up. You measured that on a real cylinder with a bathroom scale and got the ratio right.

Say which number the pump sets and which the load sets. Chapter 6, and it is the sentence the field gets wrong. The pump makes flow, and the flow sets the speed. The load makes pressure, and the pressure sets the force. A hydraulic pressure gauge is a load cell. A machine that gets slower under load has run out of flow, not out of pressure, and the flow went over the relief valve. And the gauge reads low when nothing is happening because nothing is resisting, which is why the tandem centre of Chapter 12 exists.

Say why the designer chose oil or air. Chapter 7, and the answer is always a single number: oil is about two thousand times stiffer than shop air, because a gas’s bulk modulus is nothing more than its own pressure. So the same cylinder holding the same load sags 0.020 mm (0.0008 in) on oil and 43 mm (1.7 in) on air for a 20 percent load change. Chapter 16 then made the excavator’s version of that argument arithmetically: to replace one boom cylinder you could carry with a pneumatic one, you would need a bore of 902 mm (35.5 in). And Chapter 17 made the opposite argument for the factory floor, where speed, cleanliness and the fact that a trapped hand can pull free are worth paying eight times as much per joule for.

Point at where the heat is going. Chapter 8, and this was the promise most worth keeping. Every kilopascal of pressure drop that is not moving a load has become heat, in a specific component, at a wattage you can calculate. The reference circuit does 200 W of useful work while putting 1.47 kW into the oil, which is 12 percent efficient with every part working perfectly. You measured a pressure drop in a garden hose and converted it to watts with a $12 gauge and a bucket, and then you measured a real machine’s heat rejection from the outside with a thermometer.

Point at the component that could kill you. Chapter 22, and there are two. A pinhole leak, producing an invisible jet at 220 m/s that penetrates skin above about 7,000 kPa (1,015 psi), gives almost no pain, and destroys tissue over the following hours. And stored energy, in an accumulator holding 67 kJ after the key is out, in an air receiver holding 236 kJ with no warning of its own corrosion, and in a hose warmed 20 degrees by the sun, which is enough to generate full system pressure in a trapped rigid volume.

And the specific promise, which was a number. Chapter 18 chained a 5 to 1 pedal lever, a vacuum booster adding 1,885 N, a master cylinder of 387 mm² (0.600 in²) turning 3,000 N into 7,750 kPa (1,124 psi), and a caliper piston of 2,290 mm² (3.55 in²) turning that back into 17,750 N.

foot to front pad          17,750 / 220   =  81 to 1
total force at the road                   =  14,246 N   (3,204 lb)
deceleration               14,246 / 1,814 =  7.85 m/s2  =  0.80 g
stopping distance from 100 km/h (62 mph)  =  49 m        (161 ft)

Eighty-one to one, 14,246 N, 0.80 g, two tons, fifty metres, on a fifty-pound push. The promise was 80 to 1, about 14,000 N and eight tenths of gravity. It was delivered, and you measured your own car’s version of the first three numbers with a tape measure and a $10 luggage scale.

Section 2: What You Can Actually Do Now

Not what you have read about. What you can do, on a bench, with the kit in The Bench.

Build a hydraulic press for ten dollars and predict its force ratio from two diameters before you test it.

Measure the cost of one air bubble, in millimetres of lost travel, and then recognise that same feeling with your foot on a brake pedal.

Compute the force a bottle jack will make, and the effort its handle will need, from three measurements, and be right to within the honest 20 percent that seal friction accounts for.

Prove that flow and pressure are independent with two syringes and a stack of books, which is something most people who work with these machines every day have never done deliberately.

Measure the fluid power at your garden tap in watts, and discover it is about two light bulbs, and then understand from that why the London mains ran at 5,200 kPa (750 psi).

Measure a pressure drop and convert it to a wattage of heat, which is the fundamental operation of all hydraulic troubleshooting.

Find out what your compressor actually delivers, with the pump-up test, and separately what its honest continuous duty is, and expect both to disappoint.

Measure a hydraulic machine’s pump flow from the outside, with a tape measure and a stopwatch, by timing a cylinder through a known stroke.

Measure your own compressed air leak rate and convert it to dollars a year, and then find the leaks with soapy water.

Confirm the orifice square-root law with a hose, a gauge, a bucket and a sheet of graph paper.

Read a spool valve’s centre condition off the spool itself, and therefore say what the machine does when nobody is touching it.

Compute your own car’s brake system end to end, and test its brake fluid for water in five minutes for fifteen dollars.

Build a McKibben artificial muscle for fifteen dollars and plot its force against contraction, and see with your own graph paper why designing with contractile actuators is a different discipline from designing with cylinders.

And find a leak with a piece of cardboard on a stick, having practised the habit at a pressure that cannot hurt you.

Section 3: The Ideas Worth Keeping

Eight, and they outlast every specific number in this book.

Pressure is the same everywhere, and force is pressure times whatever area you give it. That is Pascal’s whole contribution and it is the reason a thumb can hold up a stack of books, a foot can stop two tons, and a hose can go round a corner without losing anything. Every force multiplication in this subject is nothing more than choosing two areas.

Flow sets speed, pressure sets force, and the load sets the pressure. Two independent quantities, controlled by two different components, confused constantly. Once separated, most diagnostic questions in fluid power answer themselves.

Stroke is the currency. Force multiplication is always bought with distance, at an exchange rate fixed by the areas, and the product never changes. A machine that appears to offer a free multiplication has taken it out of the stroke, and the place to look is always how far the input had to move.

Compressibility is the only real difference between the two halves of this subject, and everything else is downstream of it. Stiff fluid: precise, strong, holds position, dangerous. Soft fluid: fast, springy, stores energy, forgiving, expensive. Neither is better. They are answers to different questions, and you can now tell which question a machine was answering by looking at it.

Every pressure drop that does not move a load becomes heat, and the heat has an address. Which means efficiency in a fluid system is a circuit design question rather than a component quality question. You can build a 12 percent machine out of excellent parts, and every part will be working correctly.

Ask what fraction of the time the actuator is moving. Chapter 19’s resolution. Efficiency matters in proportion to how many joules actually flow, so a transmission’s efficiency is decisive and an actuator that holds still for an hour between movements is judged on something else entirely. The right first question about any actuator is not “how efficient is it” but “how many joules will it move in a year”.

A component that fails twice in the same way was never the fault. Chapter 21’s cavitating pump, replaced twice, whose actual fault was a suction line. This is a general shape and it applies well beyond hydraulics.

And a specification is not a result. Chapter 10’s compressor nameplate, Chapter 12’s “10 micron filter” with an unstated beta ratio, Chapter 16’s published pump curve which turned out to be the thing that made diagnosis possible, and Chapter 23’s three pump figures that cannot all be true at once. The habit of multiplying a datasheet’s numbers together to see whether they close is worth more than any single fact in this volume, and it takes ten seconds.

Section 4: What This Volume Did Not Cover

Named plainly, so you know the shape of your own remaining ignorance.

Fluid mechanics proper. Boundary layers, turbulence, the Reynolds number, drag, lift, and open-channel flow. This volume used exactly two results from that subject, the orifice equation and the fact that pipe friction rises roughly as the square of the flow, and it used them as tools rather than deriving them. A genuine fluid mechanics volume is a separate book, and the aerodynamic slice of it, airfoils and lift and moving air over a surface, is already covered in the wind volume of this series, which is recommended reading rather than required.

Pipe flow calculation in earnest. Darcy-Weisbach, the Moody chart, minor loss coefficients for every fitting, and the arithmetic of designing a pipe run to a pressure budget. This volume gave the velocity guidelines that practitioners actually use, 3 to 5 m/s (10 to 16 ft/s) on the pressure side and 0.6 to 1.2 m/s (2 to 4 ft/s) on the inlet, and stopped there.

Electrohydraulic servo control theory. Transfer functions, loop gain, phase margin, the oil column resonance as a second-order system, and the design of a controller that will not make a machine oscillate. Chapter 7 named the resonance and Chapter 12 named the bandwidths; the mathematics of doing anything about it belongs to control engineering.

Hydraulic system modelling and simulation. The commercial and open-source tools that predict a circuit’s behaviour before it is built, and the modelling assumptions that decide whether the prediction is worth anything.

Component design. The internal geometry of a piston pump’s valve plate, the port timing that makes it quiet, the tribology of a slipper bearing, the finite element analysis of a manifold. This volume treated pumps, valves and cylinders as components with specifications, which is how you meet them.

Sealing technology in depth, which Chapter 2 argued is the real history of this field, and which is a specialist discipline with its own literature on materials, groove geometry, surface finish and extrusion gap calculation.

Fluid chemistry and formulation. Additive packages, how a viscosity index improver is designed, oxidation stability testing, and the standard test methods behind every number on a datasheet.

Brake system overhaul as a repair procedure. Chapters 18 and 19 explained how every part of a brake system works and what each measurement means, and they deliberately did not tell you how to rebuild a caliper, hone a master cylinder or specify a replacement hose. That is a workshop manual’s job, for your specific vehicle, and a book is the wrong medium for it. The one thing this volume asks you to do to a brake system is test the fluid and, if you are confident, bleed one caliper.

Machinery safety compliance. The standards that govern hydraulic and pneumatic safety functions, performance levels, category ratings and the documentation a machine builder must produce. Chapter 17 and Chapter 20 gave the physics and the practice; the regulatory framework is jurisdiction-specific and changes.

And several genuine applications of fluid power that are not power transmission at all. Water jet cutting, hydraulic fracturing, pneumatic conveying of powders and grain, fluidic logic devices that compute with jets and no moving parts, fluid amplifiers, and hydraulic damping in buildings and bridges. All of them use ideas from this book and none of them is about moving a load.

Section 5: Where to Go

Build the two-syringe press if you have not. Ten dollars, and it is the experiment that makes everything else obvious. It is Chapter 1 and it comes back in Chapters 2, 6, 14 and 15.

Then compute your own car’s brakes. An hour, a tape measure and a $10 scale, and at the end of it you will know a number about a machine you use every day that almost nobody else knows about theirs.

Then build the McKibben muscle. Fifteen dollars, an afternoon, and it puts a completely different actuator in your hands so that you know what a cylinder’s flat force curve is actually worth.

Then find a machine and read its diagram. This is the one that turns the book into a skill. A hire shop, a farm, a garage, a plant depot or a friend with a log splitter, and the six questions in Chapter 14.

And put a gauge on something. Any gauge, on anything. A garden tap, a car tyre, a compressor line, a brake bleed nipple if you can borrow the fitting. The habit of measuring a number rather than accepting one is the whole method of this series, and in fluid power it costs twelve dollars to start.

And one thing to carry out of here that is not really about hydraulics.

The most consequential fact in this volume is that the gauge you are looking at may be measuring the wrong side of the very component you are worried about.

Chapter 13 said it about accumulators, and it is worth restating because it is the mechanism by which people are killed by machines that have been switched off correctly. The main pressure gauge on a hydraulic power unit often sits downstream of a check valve. Switch the machine off and that gauge falls to zero, honestly, because the pressure on its side of the valve has gone. And behind the check valve, four litres of oil at 21,000 kPa (3,045 psi) are waiting, with 67 kJ in them, and nothing in the room is making any noise.

The instrument is not lying. It is answering a different question from the one you are asking, and it is answering it correctly.

That is a general shape and it is worth recognising elsewhere. Wherever a system stores energy, and wherever an indicator is fitted for operating rather than for isolating, the reading that tells you it is safe to open something is almost never the reading that is on the panel. The question is never “does the gauge read zero”. It is “is this gauge on the same side of every valve as the thing I am about to undo”.

Verify, on the correct side, or treat it as live.

Bench Degree

Get the degree without the diploma.

Learn the material, not how to pass the exam.