Parallel vs Serial Kinematic Flexure Stage Design

When designing piezo flexure stages two examples that dominate mechanical design include parallel and serial kinematics. At nPoint we employ both strategies to meet customer requirements. Testing and calibration are always done with external interferometric sensors regardless of flexure design and feedback sensor choice. Specifications tested in this manner are actual, not theoretical sensor or parasitic motion numbers.

Serial Kinematic Flexure Design

In a serial kinematic design, one axis of the stage sits on top the other in a nested in-line or stacked fashion. This design will produce lower resonance frequencies and reduced system dynamics. This occurs most notably on the bottom (outer) axis which carries all axes of the moving platform.

Sensors in these systems are located in each independent axis. Crosstalk will not be compensated for because the sensor of the second axis is independent the first. In a sense, each axis operates separately from the other.

Serial kinematic design approaches are more cost efficient to produce, and can be compact (when stacked), but are not as useful in high-end applications. Exact position is dependent on off axis guidance, which is not compensated for with a sensor. The simplistic design also allows for only one high-speed axis.


Example serial kinematic flexure design. Note how each axis is independent from the other, so feedback from one can not compensate for changes to the other. This creates inferred, not exact position.

Parallel Kinematic Flexure Design

In a parallel kinematic flexure stage the actuators for two (or more axes) will actuate one platform. In this configuration it is possible to minimize the moving mass, which allows higher resonant frequency and better system dynamics. Either axis can be chosen for high-speed scanning because resonance frequencies are typically similar.

In order to achieve the best metrology, capacitive sensors can be used for all axes between the static frame and the moving platform. The sensor of the parasitic axis ensures the position of both axes during positioning of the system.

Parallel kinematics are used for high-end systems with greater degrees of freedom to achieve the best position accuracy. Direct metrology and parallel kinematics used in combination with digital control electronics, provide the best nanopositioning systems currently available. A calibrated, parallel kinematic system may have a higher associated cost however; in the world of nanopositioning actual position often takes precedence over theoretical position and sensor resolution often used with piezoresistive sensor technology and serial kinematics. 


Example parallel kinematic flexure design. Note how each axis works together allowing for off axis changes to be accounted and compensated for. Position is measured directly in each axis.

Integrated Parallel and Serial Kinematic Designs

nPoint offers combinations of these two principle design approaches to meet customer’s specifications. An example of this would be a parallel kinematic design in the X and Y axis with the Z axis integrated in a serial design. The Z stage is often directly driven (no amplification) for speed considerations. Since it does not carrying any other axis, it is able to achieve the highest resonance frequency of all system axes. Configurations like these are mostly used in scanning applications where a 3D profile is required such as in Atomic Force Microscopy.

Pin It on Pinterest