Skip to main content

Cascade lag compensator for 50K/[s(s+5)(s+10)]

Short answer

K=30; lag compensator Gc(s)=(1+4.71s)/(1+59.9s) lowers the gain crossover to 2.14 rad/s. This yields Kv=30 s⁻¹ and a phase margin of 49.6°, stabilizing the originally unstable system.

Design a suitable cascade lag compensator network for the given system G(s) = 50K/[s(s+5)(s+10)] such that the requirement of velocity error constant of 30 sec⁻¹ and phase margin of ≥ 45° are met.

A cascade lag network Gc(s)=1+Ts1+βTsG_c(s) = \dfrac{1+Ts}{1+\beta Ts} (β>1\beta > 1, unity DC gain) is designed with the Bode plot. A lag network is suitable because the plant has enough phase at a lower frequency; we only need to lower the gain there.

Step 1: Gain for KvK_v

Kv=lim⁡s→0s 50Ks(s+5)(s+10)=50K50=K=30K_v = \lim_{s\to0} s\,\frac{50K}{s(s+5)(s+10)} = \frac{50K}{50} = K = 30 G(s)=1500s(s+5)(s+10)=30s(0.2s+1)(0.1s+1)G(s) = \frac{1500}{s(s+5)(s+10)} = \frac{30}{s(0.2s+1)(0.1s+1)}

Step 2: Uncompensated margins

  • ∣G(jω)∣=1|G(j\omega)| = 1 at ωgc=9.77\omega_{gc} = 9.77 rad/s; ∠G=−90∘−tan⁡−19.775−tan⁡−19.7710=−197.2∘\angle G = -90^\circ - \tan^{-1}\frac{9.77}{5} - \tan^{-1}\frac{9.77}{10} = -197.2^\circ, so PM=−17.2∘PM = -17.2^\circ.
  • ωpc=50=7.07\omega_{pc} = \sqrt{50} = 7.07 rad/s, ∣G(j7.07)∣=2|G(j7.07)| = 2, GM=−6.0GM = -6.0 dB.

The uncompensated system is unstable.

Step 3: New gain crossover frequency

Allow 10∘10^\circ for the lag of the network (needed for this large shift):

∠G(jωgc′)=−180∘+45∘+10∘=−125∘\angle G(j\omega_{gc}') = -180^\circ + 45^\circ + 10^\circ = -125^\circ −90∘−tan⁡−1ω5−tan⁡−1ω10=−125∘  ⇒  ωgc′=2.12 rad/s-90^\circ - \tan^{-1}\frac{\omega}{5} - \tan^{-1}\frac{\omega}{10} = -125^\circ \;\Rightarrow\; \omega_{gc}' = 2.12\ \text{rad/s}

Step 4: Value of β\beta

∣G(j2.12)∣=15002.12×2.122+25×2.122+100=12.72  (22.1 dB)  ⇒  β=12.72|G(j2.12)| = \frac{1500}{2.12\times\sqrt{2.12^2+25}\times\sqrt{2.12^2+100}} = 12.72 \;(22.1\ \text{dB}) \;\Rightarrow\; \beta = 12.72

Step 5: Corner frequencies

1T=ωgc′10=0.212 rad/s,T=4.71 s1βT=0.21212.72=0.0167 rad/s,βT=59.9 s\begin{aligned} \frac{1}{T} &= \frac{\omega_{gc}'}{10} = 0.212\ \text{rad/s},\quad T = 4.71\ \text{s} \\ \frac{1}{\beta T} &= \frac{0.212}{12.72} = 0.0167\ \text{rad/s},\quad \beta T = 59.9\ \text{s} \end{aligned}

Step 6: Compensator and check

Gc(s)=1+4.71s1+59.9s=0.0786 s+0.212s+0.0167G_c(s) = \frac{1 + 4.71s}{1 + 59.9s} = 0.0786\,\frac{s+0.212}{s+0.0167} Gc(s)G(s)=1500(1+4.71s)s(s+5)(s+10)(1+59.9s)G_c(s)G(s) = \frac{1500(1+4.71s)}{s(s+5)(s+10)(1+59.9s)}
QuantityUncompensatedCompensated
Gain crossover9.77 rad/s2.14 rad/s
Phase margin−17.2∘-17.2^\circ (unstable)49.6∘49.6^\circ
Phase crossover7.07 rad/s6.86 rad/s
Gain margin−6.0-6.0 dB15.5 dB
KvK_v30 s−1^{-1}30 s−1^{-1}

A practical RC lag network (R1R_1 in series, R2R_2–CC in shunt) gives T=R2CT = R_2C and β=(R1+R2)/R2\beta = (R_1+R_2)/R_2; e.g. C=10 μC = 10\ \muF gives R2=471R_2 = 471 kΩ\Omega and R1=5.52R_1 = 5.52 MΩ\Omega.

Answer: K=30K = 30, Gc(s)=1+4.71s1+59.9sG_c(s) = \dfrac{1+4.71s}{1+59.9s}; Kv=30 s−1K_v = 30\ \text{s}^{-1}, PM=49.6∘ (≥45∘)PM = 49.6^\circ \ (\ge 45^\circ).

✓ Answer checked by Claude Opus 5.5 (high effort)

All Control Systems chapter 7 questions →