2014 Texas Instruments Motor Control Training Series To

2014 Texas Instruments Motor Control Training Series To schedule a motor control seminar for your company, or in your area, please contact me at dave@kappaiq. com www. kappaiq. com Dave Wilson

Sensorless Sensored FOC P + + I Commanded iq (torque) I Reverse Clarke-Park Transform + + Vq Va Vb ∫ I Vc θd ∫ iq id Observer ia d/dt TI Dave’s Control Center + - + Rotor Speed ∫ P P + PMSM + - Commanded Rotor Speed + Vd Forward Clarke-Park Transform ib ic Phase C Current Calculation θd Speed Commanded id = 0 θd Shaft position sensors are VERY expensive ($1, 500 in some cases). Many applications cannot afford the cost of a shaft sensor. Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Model Based Filtering Noise Process Σ Error feedback Measurement + - Mathematical Model of Process Dave Wilson Σ Estimate TI Spins Motors…Smarter, Safer, Greener.

Tracking Filters ^ Dy correction ^ correction y + Σ β - α Z-1 + + Σ Integrator Better tracking is obtained when α and β are high Better filtering is obtained when α and β are low Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

The Tracking Filter…Unmasked! The tracking filter is revealed to be a simple 2 nd order IIR filter as shown below. Accumulator a+b X(n) Y(n+1) + + Delay X(n-1) -a 2 -a-b Y(n) Delay a-1 Dave Wilson Y(n-1) TI Spins Motors…Smarter, Safer, Greener.

Cascaded Representation This form of the filter reveals the derivatives of the tracked variable. Measured Position Error β + Σ - α Z-1 + + Σ Integrator Estimated Position Estimated Velocity Estimated Acceleration Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Parameter Estimation with Observers By providing an additional feedforward input, the tracking filter can make better output estimates. It then takes the form of an OBSERVER. Can be designed to have zero (or near zero) estimation lag. Source: Motion Controller Employs DSP Technology, Robert van der Kruk and John Scannell, Phillips Centre for Manufacturing Technology, PCIM – September, 1988 α β Integrator Model of H(z) Observers are used to “observe” a quantity which is difficult to measure by mathematically modeling the system. Observers literally recreate the desired signal mathematically (great noise decoupling). The “guess” is corrected by comparison with an observable signal. Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Servo Performance with Velocity Directly from Encoder vs. Observer Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Sensorless Sinusoidal PMSM Control applied voltage Assuming no saliency, stationary frame equations Rotor with surface-mount magnets Non-salient design (magnetically round))Back EMF component Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Stationary Frame Back EMF Observer i Back EMF Observer Vin i + PI Vin - emf + - 1 ^R s Low Pass Filter ^i emf Vin -t æ Vin (t )- emf (t ) öç ÷÷ç 1 - e t i (t ) = çç Rs è øè ö ÷÷ ø + - 1 Rs Low Pass Filter i emf Dave Wilson TI Spins Motors…Smarter, Safer, Greener. -1 emf

Back-EMF Observer Performance i 0. 416Ω Back EMF Observer Vin 1. 365 m. H 120 VAC 60 Hz One of three phases of Baldor PMSM motor i + - emf ? ? ? + 1 ^ Rs - Low Pass Filter PI -1 EMF estimate ^i 120 V Observer simulation Observer sampling frequency = 10 KHz Back-EMF Estimated Back-EMF -120 V 25 ms 0 Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

Stationary Frame State Observer for a Non-Salient Machine Texas Instruments Dave’s Motor Control Center PMSM Motor (2 -phase representation) + LPF + - Σ PI cos -1 Back EMF α + Σ Σ K 1 + Σ Dave Wilson K 2 + - Σ PI -1 sin Back EMF β TI Spins Motors…Smarter, Safer, Greener.

Sliding Mode EMF Observer Iα Vα + - 1/Ls ^ Iα + + - + Iαerror SGN Kslide - Rs/Ls Zα LPF EMFα “A Position and Velocity Sensorless Control of Brushless DC Motors Using an Adaptive Sliding Observer”, Takeshi Furuashi, Somboon Sangwongwanich, Shigeru Okuma, 1990 IEEE Proceedings, 087942 -600 -4/90/1100 -1188, pp. 1188 -1192. Dave Wilson TI Spins Motors…Smarter, Safer, Greener. Zα

Low Speed Saliency Tracking Observer P I regulator results in steady state value servoing to zero High Frequency Current Measurement Demodulator LPF High Frequency Voltage Injection (about 500 Hz) q axis st) (e xis qa s d axi P I Zq Zd t) s si (e x da θ actual θerr Disadvantages: Only works on motors with a strong saliency signal. Saliency graph shifts when the motor is loaded. Works at low speeds, but falls apart at high speeds. Dave Wilson TI Spins Motors…Smarter, Safer, Greener. LPF

ACIM Sensorless Control Note: Still in stationary frame Stationary Frame Synchronous Frame Source: Zero-Speed Tacho-less I. M. Torque Control: Simply a Matter of Stator Voltage Integration, by K. D. Hurst, T. G. Habetler, G. Griva F. Profumo, IEEE paper, 1997 Dave Wilson TI Spins Motors…Smarter, Safer, Greener.

ACIM Stator Flux Referenced FOC roto r cu rren t stator q axis roto xis (easily measured)! However, D-axis stator current no longer independently controls flux (i. e. , the axes are now COUPLED). TI Spins Motors…Smarter, Safer, Greener. roto Dave Wilson stator flux axis r flu x ax is rqa

DTC: A Peek under the Hood Vector Selector Flux Error Hysteretic Band increase decrease Texas Instruments Dave’s Motor Control Center Torque Error Hysteretic Band Commanded Torque - + Commanded Flux + Dave Wilson - increase decrease SVM Torque Calculator Stator Flux Calculator TI Spins Motors…Smarter, Safer, Greener.