Stand-alone motion controllers can take the place of PCs in some types of control systems. And there are a few different configurations possible. One is that the motion card can be housed in box with I/O and networking connections and mounted directly to the machine or process it controls. Control programs can be uploaded to the cards via USB link or Flash drives. This type of setup works well with single axis control and where the motion is fairly simple and doesn’t require interfacing with other controllers.

Another configuration is either rack mounted or DIN-rail mounted. The well-established PCI bus is the most common form factor for motion control cards in rack systems and so is readily available.

The PPCIe8443 controller board from Nippon Pulse can generate high-speed pulses to control input-type servo and stepper motors. The PCI Express board uses an updated PCI-express bus interface that can control up to 4 axes of coordinated motion.

Regardless of configuration, stand-alone motion cards can be networked together into larger, more complex multi-axis control systems. Typical options include USB, Ethernet, or other Ethernet-based connections.

However, stand-alone motion cards can also reside on PCs in some cases. The way it commonly works is that the motion control card, with a dedicated processor to handle motion control tasks requiring fast processing speeds, handles motion-critical tasks while offloading less demanding tasks to the PC processor.

Drives and controllers are key components in any motion system. A drive accepts commands from a motion controller and converts these signals into high current signals to drive a motor. So, how to connect a drive to a controller?

Manufacturers offer a range of options for drive and controller connectivity. These typically break down into stand-alone components that can be connected together to meet specific design requirements as well as integrated products, with the drive and controller integrated into one package.

A diagram of a motion control system shows a controller and drives connected via EtherCAT with individual drives powering separate motors. (Image via ACS Motion Control)

The fairly common integrated drive/controller (or drive/controller and servo motor) has internal connections between all the components. There is usually one external connection, often to a network bus, that ties individual units together. Common examples of this are such integrated units controlling individual axes on a machine.

The SmartMotor from Moog Animatics is a programmable, integrated servo motor system. The servo motor is integrated together with an encoder, an amplifier, a controller, and RS-232/RS-485 communication and I/O. (Image via Moog Animatics)

The most common way to connect stand-alone drives to controllers is via I/O connections or common networks for motion control applications such as EtherCAT, SERCOS, PROFINET or others. Which network is used depends on factors such as the speed of the application, system complexity, or the need for highly synchronized motion across multiple axes.

Non-uniform memory access (NUMA) is a kind of memory architecture that allows a processor faster access to contents of memory than other traditional techniques.

In other words, in a NUMA architecture, a processor can access local memory much faster than non-local memory. This is because in a NUMA setup, each processor is assigned a specific local memory exclusively for its own use. This eliminates sharing of non-local memory, reducing delays when multiple requests come in for access to the same memory location.

The illustration shows the differences between a NUMA and UMA architecture. The NUMA layout has local memory assigned to each processor, while in the UMA setup multiple processors share memory via a bus. (Image via Database Technology Group)

NUMA architecture is common in systems with multiple processors. For example, in industrial controls, particularly in systems with multiple processors, this is beneficial because it speeds up overall system performance times. NUMA allows for fast and easy access to memory for the processors, as opposed to shared memory architectures where access times may be longer, thus slowing down execution of key processor and system tasks.

NUMA architecture was developed largely due to the advent of modern microprocessors that are faster than memory speeds. As a result, local memory is placed closer to the processor, shortening signal path lengths and reducing latency times with fewer delays in accessing memory.