Precision motion systems are becoming more and more widely used, so many
engineers are having to consider designing them. In some cases it may be
better to call in an expert servo designer, but whether you do it
yourself or brief a consultant it is best to have some ideas about
procedures and consideration. Andy Sumner concludes his two part series
with some basic dos and don’ts
There is a misconception abroad to the effect that you can solve any
actuation or motion problem simply by sticking a great big servo-drive on
the end. This is not true and it is possible to waste a lot of time and
money proving it to yourself. It is far better to work through a few
logical steps and, if necessary, get some expert help from a motion
specialist. The first thing to realise is that no matter how good your
control system, it cannot mask fundamentally bad mechanics. So when
designing a machine or motion system, start with the mechanics and do not
move on to the controls until they are optimised.
In practise there are only three basic mechanisms that can be used:
rotary inertia or flywheel, belt drive or rack and pinion and ball screw.
There are well known equations associated with each of these for
calculating their net dynamic effect or ‘inertia reflected to the
servomotor’s shaft’. These equations will not be considered here; most
engineers will either know them by heart or know ehere to look them up.
Suffice it to say that the most common mistake at this point is to
concentrate only on the obvious load and forget to add in the mass of the
drive mechanism and the dynamic effects of speed and acceleration.
Inertias are additive, so all you have to do is work them out for each
element of the total load and add them together to get the load on the
motor shaft.
One quick aside, though, before proceeding. There are other mechanisms,
such as cranks, but these are best avoided because as they go over centre
they decouple the load from the drive or induce a lost motion effect.
this sends the servo controller into an unstable state. It is possible
(theoretically, at least) for the controller to recover stability, but it
is better to avoid the problem all together by using an alternative
mechanism.
Inertia
With the mechanics in place, you have one or more dynamic loads for your
control system to address. This is achieved through four simple questions
for each load: how much, how far, haow fast and how often? The answers to
these questions are the starting point for the control system design, but
don’t go rushing ahead and size the motors yet - there is more to
consider first.
To a servo engineer, load is not simply the dead weight, as could be
measured on a pair of scales. It is the inertia as experienced at the
motor shaft, so you must allow for the actual load, plus the load induced
by the mechanical elements of the drive mechanism, plus the dynamic
effect of the load and mechanism as they move in time and space; plus a
couple of other little factors like thermal capacity that we will come to
later. If you have ever got to this point then worried about having a
nervous breakdown - don’t; it’s easier than its seems. And if it’s your
first time designing a servo-system, the advice is to get expert help
from here on, anyway.
Once you have the load’s inertia, calculate the ratio between this and
the inertia of the motor’s rotor. The ratio can then be optimised by
selecting an appropriately sized motor and by using a belt or planetary
gearbox to reduce the load’s inertia. A good ratio to aim for is between
3:1 and 5:1. if you require motions with very high dynamics, you need to
approach closer to 1:1; peak loads will be considerable. If you go
significantly beyond 5:1 you will begin to lose accuracy and dynamics -
key reasons for specifying a servomotor in the first place. This may
appear restrictive, but inside the motor’s control