# Could someone explain dither signal?

I know that regarding proportional valves the dither signal is a signal superimposed on PWM command that actually moves the valve.

It improves the response of the valve because the coefficient of static friction is higher than the coefficient of kinetic friction.

What I don't understand is how that could be. How the valve is moving forward per example and at the same time backwards and forward.

• Do you understand now that the driving signal is force , not position? and position is a quadratic result of a constant force? – Sunnyskyguy EE75 Feb 24 at 15:25
• @SunnyskyguyEE75 Yes, thank you – veronika Feb 24 at 17:58

By adding a vibration wave to create a small dithered motion, the mass frees itself of static friction or "stiction" so that it can move freely with only low-level dynamic friction.

Current causes force which equals acceleration (F=ma) and not position changes which are a 2nd integral of acceleration or 12dB/octave reduced amplitude with frequency. This is just enough to overcome slack in the mechanics and prevents stiction from slowing the response.

Linear superposition applies so its position would only go back & forth when idle, before a valve position-change command. The wave then just modulates the acceleration and braking to the next position with vibration so it doesn't really do what you suggested.

HDD's do the same thing before spinning up the disk by dithering the magnetic voice coil to rotary head-arm assembly. This is so the R/W heads don't stick to the disk and rip off from high acceleration to near 50~100 KPH in a few seconds. It might sound like a rising buzz-sweep, then click to unlock and then it starts spin-up.

Quality Hydraulics have a good explanation of dither.

Static friction, stiction, and hysteresis can cause the control of a hydraulic valve to be erratic and unpredictable. Stiction can prevent the valve spool from moving when given small input changes, and hysteresis can cause the shift to be different for different applications of the same input signal. In order to counteract the effects of stiction and hysteresis, small vibrations (cyclic frequency) around the desired position are created in the spool. This constantly breaks the static friction ensuring that the spool will move even with small input changes, and the effects of hysteresis are averaged out.

Dither is a small ripple frequency that is superimposed over the PWM signal to the solenoid current that causes the desired vibration and thereby increases the linearity of the valve and improves valve response.

Dither and PWM frequencies complement each other for improved spool control and are, in most amplifiers, adjustable independently. This allows the user to customize these signals to each individual application for optimum performance.

In my work as an automation engineer I have only come across a dither adjustment on a series of SMC Pneumatics proportional valve control card. These were 4 - 20 mA input and a PWM dithered output. The cards had potentiometers to adjust the span of the output but also had pots to adjust frequency and dither. When the latter were incorrectly adjusted I could hear wild oscillation of the valve spool and resultant pressure. I was concerned that the dither might cause excessive seal wear so we adjusted to minimise this while still controlling the valve. They lasted at least ten years with this setup.

A valve has a spring to force a spool to be retracted to its reference position, zero position. The electromagnetic force provided from valve coil acts in opposite direction, opening the valve. When these forces are in balance the valve stops moving and remains position that is determined with the characteristics: valve opening VS. coil current.

Superimposing to the DC driving current a small AC current will have a hammering effect. This is so called dithering. The spool shakes back and forth but it doesn't move due to dithering.

Am reminded of meter movements where a coil-driven pointer indicates current flow...old-style fuel gauge, ammeter, or even speedometer. The same stiction problem sometimes applies to these indicators. A viewer, unsure if the pointer is indicating a proper reading "taps" its face, to jog the pointer's suspension. This is a dithering process meant to overcome stiction.

In old piston-driven aircraft, many such instruments are arranged before the pilot. The constant engine vibration provides the dithering - no tapping is necessary.