Dr Nicola Symonds is the Director of nC2 Engineering Consultancy, part of the University of Southampton. Her expertise is in tribology – the science of lubrication – and particularly the measurement of friction. She says: “The term ‘stiction’ has been used in different ways by lots of different people.” Sometimes stiction describes static friction itself, or else it describes the effects of that static friction.
Symonds gives one definition: “Stiction is the resistance to the start of motion, usually measured as the difference between the driving values required to overcome static friction in either direction” – and then another, slightly different: “Stiction is the tendency to stick-slip due to high static friction.”Whichever the definition, the effect is clear: “In a control valve, you’re giving it an input, but it ain’t moving!”
The coefficient of friction (Cof) is only consistent between particular combinations of materials in particular conditions – and static and dynamic coefficients of friction are not necessarily the same. If it’s easier to keep the box moving than to start it moving, and the weight of the box hasn’t changed, then the static coefficient of friction (S.Cof) with the floor is higher than the dynamic coefficient (D.Cof). This is true for almost all combinations of materials.
Not quite all, though, she adds. “We’re working with companies looking for material combinations which have a steady value – where the dynamic and static friction is about the same, and therefore the designers can pick loading knowing what friction you’ll get out.” This consistency is a particular issue for reciprocating mechanisms. She explains: “In the middle, it’s going at its maximum speed, but then it slows to a stop and comes back again. Velocity is zero, so effectively you’re under static load again.”
Symonds quotes another useful definition of the concept of stiction as a property of an element such that its smooth movement in response to a varying input is preceded by a sudden abrupt jump, called the ‘slip jump’, which is expressed as a percentage of the output span. She adds: “5% stiction means you have to increase your control input by 5% to overcome this jump –but even 1% is enough to cause problems.”
Where does the excess static friction come from? “In pure tribology it’s an effect of adhesion,” according to Symonds, adding that this can be particularly significant at very small scales. “There’s a lot of research trying to remove it from microelectromechanical systems (MEMS) because the static forces and even capilliary forces from moisture become magnified. There is a lot of work on specific coatings for MEMS.” Applications include the accelerometers used in instrumentation and even smartphones.
The approach is different at larger scales: “Most of the research in plant and systems is about identifying and controlling [stiction] – modifying the software to account for it”. And this can be complex, because most valves do not have a direct measurement of their position. “You know what the input is, and you know the output in your flow”.
Stiction makes accurate proportional control of valves more difficult, leading to problems such as increased energy consumption, decreased product quality and excessive wear. Control-based measures to compensate for valve stiction include ‘dithering’ (see below) and ‘the knocker’ – a series of short control pulses to free the valve – but these can themselves lead to excessive wear. ‘Feed-forward’ boosts the current to a servo valve if the speed is low, although correcting an overshoot can cause oscillations in the control loop. Now more sophisticated approaches to fighting stiction and oscillation are generally known as ‘friction compensation’.
Symonds is more concerned with treating the causes of stiction at the source, starting with measuring friction characteristics of particular combinations of materials. As she says: “It’s an empirical subject – you need to do the actual test. You can’t just look up a data table.” Materials are tested with or without lubrication, and over a range of temperatures. “In some industries you are actually looking for high static friction [that is, grip] but in those circumstances, you want it to be controllable or understood: [ideally] when you look at the graph that comes out, it doesn’t have a high peak and goes across steady.”
What are the general principles for reducing stiction? Symonds says, “Get the right material pairs together. Differences in hardness are a problem because of wear – you can’t talk about friction without talking about wear.” A soft material that wears around a harder part will quickly align, but as it continues to wear risks causing a leak. “Or you can have a pair of hard surfaces, but then you need very good geometric alignment”. Nevertheless, she adds, “I’d always advocate going for the materials selection solution.”
Although in theory friction is proportional to perpendicular force, Symonds advises: “Reducing the contact area over which adhesion will occur will always be a benefit.” And there are other issues: for instance, put a valve in a high-temperature environment and differential thermal expansion rates of dissimilar materials may mean it no longer works.
Are there any particular lubricants or additives which can eliminate stiction? “Lots of lubricants have additives which help,” says Dr Symonds, “but you can make them bespoke for one environment, and they won’t help in another.” However, she adds, “Having a lubricant over not having a lubricant will always help.” (As one example, OKS 477 is a food-safe grease for valves in process industries.)
So stiction is a concept which needs to be defined precisely in a particular application – valve suppliers should be clear about whether they mean the cause (static friction) or the effect (stick-slip). Dealing with stiction means examining a number of factors. She concludes: “There are so many variables out there – but that’s what makes it such an interesting subject!”
BOX: MECHANICAL DITHER
During World War II, US aircraft used complex mechanical computers for navigation and aiming bombs. These devices reportedly worked better in the air than on the ground, due to the constant vibration on board, which imparted some stiction-reducing motion to every component. That phenomenon, known as ‘mechanical dither’, has since been used deliberately, with vibrating actuators smoothing train suspensions, for instance, and to make inertial navigation gyroscopes rotate more freely. It has also been used on the control systems of valves, with a varying oscillation (‘dithering’) superimposed on the control signal — but this doesn’t work on pneumatic valves, which tend to filter out high frequencies.
Another example is ‘powered’ safety razors, which are said to stimulate the skin’s hair follicles, though it’s more likely that the vibrating motor in the handle induces enough motion to reduce the stick-slip of the blade.
Dr Nicola Symonds says that mechanical dither may have its uses, but “as a tribologist, I’d say it’s better to pick materials that don’t have high static friction in the first place.”