Hardware Reference
In-Depth Information
Table 3.9: MASSC for different control schemes
Control scheme MASSC in µin
Case 1
10
Case 2
> 1000
Case 3
> 1000
for P
M
, the MASSC can still be greater than 1000 µin when the DC gain of
P
M
fluctuates in between ±50% of its nominal value.
30
25
20
15
10
5
−60
−40
−20
0
20
40
60
Estimation Error in Percent (%)
Figure 3.105: MASSC for linear control scheme with reduced order micro-
actuator model.
It is observed from the above results that for the linear control scheme of
Figure 3.105, a lower value for P
M
increases the MASSC. However, lower value
of P
M
decreases the gain of the VCM path and affects the hand-off properties
of the dual-stage loop even though the microactuator is not saturated. The
proposed nonlinear modification, on the other hand, has a much larger MASSC
with a saturating secondary stage. Furthermore, it retains all the properties
of the original linear design with the microactuator not saturated.
In a dual-stage servo controller, the secondary stage actuator helps to
achieve higher servo bandwidth but not necessarily lower sensitivity peak.
MEMS based microactuators with frequency response similar to that of a dou-
ble integrator 1/s
2
will produce the same bode integral, and hence the water
bed effect. Similar to the discussion on VCM, MEMS actuators having veloc-
ity and/or acceleration sensor can help to achieve a lower Bode integral value,
and hence, more vibration rejection with the same servo bandwidth. The fre-
quency responses of PZT actuated suspensions and sliders have constant gain
