In this application example we will look at a design project to create a controller board (PCB) for a three axis gantry controller. We will look at two different design approaches: building the board with MC58113 Motion Control ICs and building the board with ION/CME N-Series Digital Drives.
Figure 1: Three axis Gantry
In this application example, we will look at two different approaches for building an embedded motion control board that controls a three axis gantry. There are many types of gantries, but we will focus here on gantries used with laboratory and scientific equipment.
Figure 1 shows a representation of such a mechanism. It has an X and a Y-axis mechanism each driven by a motor and a third motor to control motion in the Z plane. Together these motors can therefore be driven to access any point in the gantry's contained XYZ operating space.
Although an impressive mechanism, by itself such a gantry can do very little. That's because the XYZ gantry is really a framework for carrying all manner of actuators to accomplish a specific (or perhaps several specific) functions in the workspace.
A common gantry actuator type is a liquid pumping system, consisting of one or more syringes for aspirating and dispensing liquid. But the possibilities for gantry actuators are endless and include pick and place mechanisms, cutting tools, glue applicators, 3-D printing extruders, and more.
For this design comparison example though we won't worry about the entire machine including the actuator(s), and instead we will focus just on the XYZ gantry controller itself.
To give our gantry controller some specific performance metrics to meet we have detailed a few of the important design parameters. These include the motors to be controlled, the gantry operating modes, and a few other details. These parameters can be seen in the table below.
Figure 2 shows the overall control architecture. There are three axis controllers in total but there is only one connection between the controller PCB and the Ethernet host network. The CPU or control unit on the controller board interfaces with the host network and parses received commands. It then coordinates the movement of the axes under its control according to this received command.
Figure 2: Overall architecture of three-axis gantry controller
For example, if a host command is sent requesting the gantry move to location XYZ, the controller board's CPU receives this command and then determines the specific profiles that each axis must execute to affect that command. This might include generation of special profile paths that avoid obstacles if there are physical areas within the XYZ space that are off limits.
OK, with the design parameters and control architecture specified, we are ready to look at each design approach and then compare and contrast the results. We will start with the IC-based controller board.
For the IC-based design we will use three MC58113 Motion Control ICs from Performance Motion Devices, Inc. The MC58113 is a popular general purpose single axis motion control IC. It does profile generation, servo loop closure, signal generation for digital switching amplifier control, and many other timing and motion signal processing related functions.
Figure 3: MC58113 Connections Diagram
Figure 3 shows typical connections when using an MC58113. Note that the "8" version of this product means the motor type is software programmable. For dedicated brushless DC, step motor, or DC Brush control versions different members of the same family part numbers are used: the MC53113, MC54113, or MC51113 versions, respectively.
The other major control element we are going to need is a microcontroller that can talk by Ethernet to the host network. For this function, we will assume a dedicated Ethernet IC driver (called a PHY) is used.
On the motion IC side, we will need to communicate with the three MC58113 motion control ICs. The MC58113s support several interfaces including SPI (Serial Peripheral Interface), CAN, and serial, but in this example, we will use SPI. SPI is efficient, fast, and a great choice for on board communications between ICs.
The overall control architecture for the IC-based controller board is shown in Figure 4 below.
Figure 4: Control Architecture for IC-based Controller Board
The table below details, roughly in chronological order, the project design steps. These include designing the schematic for this board, creating the board layout, producing the prototype, and then testing. We have allowed time for one re-spin with the above steps repeated (although hopefully with only minor tweaks, and no big changes needed to the design).
To anyone who has designed a control board such as this before, the above sequence will sound very familiar. The timeframes are estimates and should not be taken too literally. Some teams will get it done faster, and some slower. But from this list it's clear that the overall project is significant in its scope and requires the right EE design resources to be successful.
| #1 | Select Connectors | 1 week |
| #2 | Determine Board Form Factor | 1 week |
| #3 | Create Schematic | 16 weeks |
| #4 | Send Board Out for Layout | 3 weeks |
| #5 | Send Board Out for Fab | 4 weeks |
| #6 | Debug Board & Correct | 4 weeks |
| #7 | Send Board Out for Layout | 2 weeks |
| #8 | Send Board Out for Fab | 8 weeks |
| TIME TO PRODUCTION-READY BOARD | 39 weeks |
Although there is a solid amount of work here working with dedicated motion control ICs makes the overall design task much, much easier. Not only because the IC itself does so much of the motion control heavy lifting but because these ICs come with a wealth of example design schematics and are available as developer kits. The MC58113 ICs which we have used in this project have an available developer kit called the DK58113.
Motion control IC developer kits are hugely important because they let the software developers start writing motion control sequences and spinning motors almost immediately, well before the new controller board design is ready. So, developer kits allow both the EEs and the Software engineers to work in parallel, greatly lowering the project time.
Figure 5: ION/CME N-Series Digital Drive
For the N-Series ION-based controller board design we will use (surprise) three ION®/CME N-Series ION Digital Drives from Performance Motion Devices. N-Series IONs are integrated single axis motion controllers that are packaged in a rugged, compact, PCB-mountable format.
Figure 5 above provides an image of the N-Series ION, and Figure 6 below shows an Internal Block Diagram for the N-Series ION.
Figure 6: Internal Block Diagram for the N-Series ION
The overall control architecture for this board design is shown in Figure 7 below.
Figure 7: Control architecture for N-Series ION-based Controller Board
There are a few key differences between the motion control IC-based and the N-Series ION version of the gantry controller board. For one, there is no need for a separate Ethernet controller IC. N-Series IONs come in a choice of four different host interfaces one of which is Ethernet. So the N-Series ION unit can receive commands directly via Ethernet.
In addition, there is no need for a separate microcontroller because N-Series IONs have an internal microcontroller available to the user called the C-Motion Engine. The C-Motion Engine accepts C-language code downloaded to it and can execute code at over 400 MIPs.
To unpack the received commands user-written code is loaded onto the host network-attached ION. This ION unit directly controls one axis, the X axis, but also sends motion commands to the Y and the Z axes controllers. This is achieved via a second network connection called an expansion network. All N-Series ION support both a CAN and an SPI expansion network.
The design task to create a board with N-Series IONs is not so much creating a schematic, it's creating a wiring diagram. And the resulting board is not 10, 12, or 14 layers (as would be typical for the IC-based board), but 2 or at most 4 layers. The table below lists the project steps and estimates the effort involved for each:
| #1 | Select Connectors | 1 week |
| #2 | Determine Board Form Factor | 1 week |
| #3 | Create Schematic | 1 week |
| #4 | Send Board Out for Layout | 1 week |
| #5 | Send Board Out for Fab | 2 weeks |
| #6 | Debug Board & Correct | 0 weeks* |
| #7 | Send Board Out for Layout | 0 weeks* |
| #8 | Send Board Out for Fab | 0 weeks* |
| TIME TO PRODUCTION-READY BOARD | 6 weeks |
* Assumes a respin is not necessary.
It is time for the big reveal. Let’s start with the boards themselves. Figure 8 shows the IC-based controller board design, and Figure 9 shows the N-Series ION based controller board design.
Figure 8: IC-based Controller Board
Figure 9: N-Series ION-based Controller Board
As one would expect for boards that are built for the same project, much of the design elements are similar. In particular, the location and numbers of connectors are more or less identical.
But there certainly differences too. One is that the heat sinking scheme for the IC-based board is quite different than the N-Series ION based board. The IC-based design requires traditional air cooled heat sinks glued directly onto the amplifier switching ICs, which in this design are MOSFETs.
The N-Series ION design does not seem to have any heat sinks, but that is because the N-Series IONs are below the board and are intended to mount directly to the unit chassis. This approach is called substrate mounting and has the advantage of achieving a robust mechanical mounting of the printed circuit board as well as and a robust heat sinking function all in one simple approach.
Figure 10 below illustrates the two different N-Series ION mounting approaches via a simple one-axis design. Substrate mounting is to the left and traditional 'upward' mounting (called component mounting) is shown to the right.
Figure 10: Substrate versus Component Mounting Approaches
Below in Figure 11a and 11b another difference between the two boards emerges, which is that the N-Series ION board has a somewhat smaller footprint.
Figure 11a: IC-based Design Footprint
Figure 11b: N-Series ION-based Design Footprint
The reason is that with the N-Series ION the circuitry needed for the motion control function is on several internal levels, whereas in this IC-based design everything is on one plane.
The tables below summarize the key differences between the two design approaches, one using motion control ICs and the other using N-Series IONs.
| Design effort | 39 weeks |
| Board area | 26.3"2 (170 cm2) |
| Mounting & heatsink method | Hex standoffs, air heat sinks |
| Components cost per axis | $150 |
| Design effort | 6 weeks |
| Board area | 20.6"2 (133 cm2) |
| Mounting & heatsink method | Substrate |
| Components cost per axis | $300 |
While there are some differences in the designed controller boards themselves, ultimately the overwhelming difference in these two development projects is the design time. The design time for the N-Series ION-based board is 6 weeks, while the design time for the IC-based design is 39 weeks. This factor of six improvement is ultimately the "special sauce" that the N-Series ION brings to the table - Allowing engineers to create fully custom control boards in much less time than possible before.
Is there more to consider? Yes. As can be seen in the tables above the motion control IC based design has a significant cost advantage in terms of per unit cost compared to the N-Series ION based approach. The IC-based design is less expensive for a few reasons, but mainly because the extra design effort allows the engineer to include only the electronic functions needed for the project at hand.
So, in the end both design approaches have merit, and for a given controller board project the designer must decide what is more important - convenience and fast time to market, or per unit cost. This decision will be affected by the expected production volume of the machine, the engineering design resources available, and the importance of time to market.
Of course, a hybrid approach is also possible - quickly creating a functioning controller board using N-Series IONs for early prototype machines, while in parallel undertaking a motion control IC-based design effort to access lower per unit cost when ready. This approach is made convenient by the fact that all PMD products use a single motion language, C-Motion. So, code written for the N-Series ION will also work with the MC58113 ICs and vice versa.
PMD has been producing ICs that provide advanced motion control of DC Brush, Brushless DC, and step motors for more than twenty-five years. Since that time, we have also embedded these ICs into plug and play modules and boards. While different in packaging, all of these products are controlled by C-Motion, PMD's easy to use motion control language and are ideal for use in medical, laboratory, semiconductor, robotic, and industrial motion control applications.
Performance Motion Devices’ MC58113 Motion Control IC is ideal for controlling small Brushless DC, DC Brush, and step motors, providing high-performance motion control features including field oriented control and closed loop stepper operation. In addition, the MC58113 IC provides profile generation, servo-loop closure, commutation, current control, and direct PWM (Pulse Width Modulation) motor output command to drive MOSFET switching amplifiers. It is ideally suited for portable and mobile medical, scientific, robotic, and automation applications and is available as a single, one-axis IC.
N-Series ION Digital Drives combine a single axis Magellan IC and a high performance digital amplifier into an ultra-compact PCB-mountable package. In addition to advanced servo and step motor control, N-Series IONs provide S-curve point to point profiling, field oriented control, downloadable user code, general purpose digital and analog I/O, and much more. With these all-in-one devices building a custom controller board is a snap, requiring you to create just a simple 2 or 4-layer interconnect board.