|Product Performed to Expectations:||10|
|Specifications were sufficient to design with:||10|
|Demo Software was of good quality:||8|
|Product was easy to use:||8|
|Support materials were available:||8|
|The price to performance ratio was good:||7|
|TotalScore:||51 / 60|
New Year’s Grab Bag RoadTest – Eaton SmartWire-DT
By Gough Lui – March 2019
Doesn’t time pass by quickly – it’s a new year and there’s more RoadTesting to be done! Thanks to element14 and rscasny for selecting me as one of the New Year’s Grab Bag RoadTesters. I’ve been doing my best to keep it a secret – I’m the lucky one applicant that received the Eaton SmartWire-DT system featuring a number of components. While I have no direct experience in industrial automation, it’s something I’ve been aware of through watching others at work and it’s something I’ve wanted to experiment with. This RoadTest is the perfect excuse to devote some time to PLCs.
A little bit of research shows that plans to RoadTest the Eaton SmartWire-DT system were already in TD9, markav, jeastwood and cghaba (Part 1, Part 2). All except the last review were somewhat disappointing in their contents – a reflection of where the RoadTest program was six years ago., with reviews delivered in early 2013. Those reviews included those of
Even though the product is now the better part of six years old, I’m still excited to get my hands on it and give it the RoadTest it deserves. It would be much appreciated if you would leave a comment, like, rate or bookmark the document if you found it interesting or helpful.
Information about SmartWire-DT seems to be rather sparse, without even a Wikipedia article covering it. It seems that the story starts with Moeller’s EASY line of PLCs. Moeller was acquired by Eaton in 2008 and in 2011, a rebranding saw Moeller products rebranded Eaton. From there, it seems that the SmartWire-DT family of products was introduced under the original EASY name. The earliest documentation I have that refers to the equipment supplied is from 2009 which refers to the bus as “SmartWire-Darwin”. I suspect the initialism DT stands for “Darwin Technology” as per the logo.
To understand the motivation behind SmartWire-DT, we first have to understand how things were done without SmartWire-DT, namely a conventional control panel.
In a conventional control panel, you will often have a PLC of sorts with a lot of inputs and outputs. For each contactor, you would wire in the power connections but also wire up the coils and auxiliary contacts to the inputs on the PLC. Every switch, similarly, would have additional connections to the PLC for button lighting. Ultimately, what you would get is a “rats nest” of wiring with a mixture of both control (low-voltage) and power (high-voltage) cabling running together. Ensuring there are no wiring errors and troubleshooting such systems could prove difficult as some types of errors may only manifest in certain conditions.
In contrast, SmartWire-DT aims to make things easier by instead having “smart” modules distributed at each contactor and switch which interface with an 8-way green ribbon certified for running with power wires.
The ribbon carries a differential RS-485 multi-drop serial data stream along with auxiliary power for contactors and a 15V DC supply to operate the bus. Bus lengths up to 600m are supported, with up to 99 nodes. The system offers automatic address configuration, automatically scanning for connected devices and ensuring correct connections, while also offering possibility for monitoring data from motor drives as well.
The SmartWire-DT system can also be extended with “round” cable and be used in machine wiring as well. It can also be integrated with other controllers through the use of ProfiBUS, CANopen or Ethernet/Modbus-TCP gateways at additional cost. It can also integrate with other EASY controllers through EASYnet to build larger “distributed” solutions.
While I can see how the SmartWire-DT system can simplify panel-building, reduce wiring errors and make for neater panels, it is what is “unspoken” that is of concern. The SmartWire-DT system is limited mainly to products from Moeller/Eaton and a select few other vendors, limiting choice and increasing initial cost. Being a digital bus system, the system may be more vulnerable to disruption in case of a fault with any of the components taking down the bus and perhaps increased latency could factor into some applications. But more than that, troubleshooting the system and bringing it back online entails a completely different process compared to that of working on conventional panels (where probing with a meter and jumpering out connections are everyday occurrences). Replacement parts may also be less widely available, which could factor in when you do have a breakdown. Additionally, specific software, programming cables, connectors, ribbon and crimping tools are necessary which add costs that could eat into the supposed savings and reduce the amount of personnel who may be able to repair the system – perhaps if you were building many panels, then you might realise some level of savings.
Another concern with the SmartWire-DT system is that it seems to have a slight “identity crisis”. On the one hand, it bills itself as a “lightweight” and “lean” solution but on another, it tries to perform some of the tasks commonly reserved for “heavyweight” field-bus solutions with distributed controllers. Whereas some of the more “heavyweight” competitors have “modular” PLCs based upon running field-bus right up to each control point with much greater configuration flexibility in integrating with any other device capable of operating on that particular standardised field-bus protocol (e.g. ProfiBUS/ProfiNET), the SmartWire-DT solution is notably less capable. While SmartWire-DT can interoperate with such field-buses via a gateway, the SmartWire-DT portion of the network is still very much proprietary and closed.
It is for these reasons that I suspect SmartWire-DT did not achieve the level of popularity that Eaton may have hoped for – while it undoubtedly worked and had its benefits, it was too closed, potentially too limiting and a potential liability when it came to servicing.
When I received the package, I really scratched my head. Was it indeed the Maxim RoadTest that was my final preference? Surely an ECG evaluation board wouldn’t weigh several kilograms …
Just one more layer and the truth is unveiled …
Indeed, it is not the Maxim RoadTest (almost got me there rscasny & danzima) but the Eaton Smartwire-DT RoadTest. Interestingly, it seems Eaton had prepared a special package along with Newark for the RoadTest, although the picture is a bit deceiving as it illustrates motor starters, an EasyNET capable controller, a traditional EASY PLC and pilot switches that are not the supplied units.
Inside, we get a wealth of components needed to RoadTest the system. There is one thing missing from this package – apparently originally the kit contained a mini screwdriver, but I suspect it was either lost or repurposed at one stage. This kit has probably been sitting around for six years …
A look at the “flatter” objects include some documentation for the SmartWire-DT interfaces for the contactors, a printed graphical colour RoadTest guide by Ed Ku of Eaton, a printed SmartWire-DT brochure, the easySoft-Pro V6.90 CD for programming the controller and an updated USB driver CD for the programming lead. The inclusion of a RoadTest guide is a little unusual although not entirely unwarranted, as it could be a little daunting for those who have no experience with the equipment to set it up and get running – but I chose to ignore it as I felt that following a manufacturers’ suggestions for RoadTests risks missing out on key points of the user experience by holding their hands and perhaps “glossing over” some of the less polished parts.
The DILM7-10 contactor on its own can be used as a regular 24V DC 3-phase 400V 3kW contactor. The label on the side provides a wealth of ratings.
But if you mate a DIL-SWD-32-002 to the contactor, it becomes SmartWire-DT enabled. The unit has prongs which reach into the coil through slots in the front of the contactor, and latches which grab into the channels in the front of the contactor to secure it in place. Each SmartWire-DT device features a bi-colour LED for diagnostics which indicates connectivity status and operating status. Additional contacts are available on the unit for interfacing with motor starters for control and monitoring purposes – perhaps these can be repurposed for other uses but I did not investigate. The resulting combination does have a rather deep profile which may perhaps put some level of leverage onto the DIN rail.
Many of the pilot switches are built around the M22-SWD series of switch modules, this particular variant being the K22LED-W indicating white LED with two switches.
The button hardware is supplied in a separate bag with its mounting hardware included.
The switch mechanism is attached to the button hardware through a tool-less “clip” design. It seems possible to clip in the switch either way, which could lead to reversal of inputs if a direction convention is not adhered to.
A 24V linear DC-power supply is provided with pre-crimped ferrules for powering the unit. Unfortunately, as I live in a 240V/50Hz country, this power supply is not appropriate, so I instead used a Manson HCS-3102 switchmode lab power supply I had spare.
Included is an Eaton EU4A-RJ45-USB-CAB1 (01) for programming and diagnostics. The cable itself is surprisingly difficult to buy and is quite pricey now – an absolute necessity if working on EASY-series PLCs. It has just three wire connections and is based on RS-232.
The EASY802-DC-SWD comes in its own box with its own documentation. This particular unit claims to consume 5W with an input from 20.4-28.8V. It is shipped with OS V1.00. The unit itself is very slimline with inputs for auxiliary power (to power contactors), power (to power devices), a COM connection for programming, some status LEDs and a configuration button to auto-assign addresses to devices on the SmartWire-DT bus. Aside from this, there are no digital/analog inputs/outputs.
This is, with the exception of the SmartWire-DT bus itself where other modules can be added. In the current state, this does make operation and diagnostics slightly different as other PLCs may have LCD screens for feedback and diagnostics, but this unit relies on watching LEDs on the devices or using a computer with the programming lead.
The SmartWire-DT cable is clearly identified with its green colouration and logo. It is a special type of ribbon with a 600V insulation rating, designed to be run alongside mains wiring in the same ducts.
SmartWire-DT device plugs feature a “clip” locking mechanism to prevent accidental unplugging or loosening due to vibration. The connections are keyed to prevent incorrect connection, with the black stripe of the wire indicating the +15V supply for SmartWire-DT devices which also gain power from the bus. The plugs are crimped using an Eaton-branded specific tool.
Cable ends are dressed in flat IDC connectors, with the far end of the bus being connected into a terminator (SWD-RC8-10) which dampens signal reflections on the bus ensuring transmission integrity.
Additional close-ups and partial-teardowns can be found in my New Year’s Grab Bag RoadTest Extras blog.
As I would like to test the equipment in a modern setting, I decided to test it on a laptop running Windows 10 Professional 64-bit edition.
To get started involves first installing the easySoft Pro software. Being that the shipped items were quite old, I felt the easiest way to get started would be to download the latest versions from Eaton’s website. While easySoft 7.00 is now available, it requires licensing and the lack of the “pro” naming suggests it does not support EASY800-series devices. As a result, I had to install easySoft Pro V.6.90 from disc first, and then install the downloaded easySoft Pro V6.96 update to bring my installation up to date.
The next step was to install the drivers for the programming cable, as it is not automatically supported.
I tried installing the drivers supplied on the original CD but they failed due to being unsigned.
Downloading the latest drivers from online, version 2.20 was able to install correctly.
Once this is done, then it becomes possible to use easySoft Pro to design, program and diagnose EASY-series PLCs. By default, opening the software results in a blank project with no devices.
From here, it is possible to build the project by adding a device. Interesting to see that security PIN code functions are available to lock the controller against certain operations, although I did not try it. It suggests that a technician could make it difficult for other people to service the system.
When connected to the PLC via the communications tab, it is able to read out information about the device and connected peripherals. This can make commissioning easier as it automatically verifies what is connected to the bus.
From the communications tab, it is also possible to load a new program into the PLC, verify if the program in the PLC is equal to the program on the computer, read a program from the PLC, erase the device and configure the time on the device.
In my case, attempting to load a program for the first time resulted in an error, as easySoft Pro was pre-configured to expect a device with firmware 2.00 or greater, whereas the device I have has 1.00. A change to the project settings to build for 1.00 or greater resolved the issue, allowing for program load to the PLC. The process takes close to a minute as the PLC must be first stopped, program loaded, verified, bus enumerated/configured and then program rerun.
When it comes to building your project, after adding a PLC device, it comes time to add the devices on the SmartWire-DT bus. A selection of library components from Dajo, Eaton, Falk, Phoenix Contact and Wohner are available, although you do require the actual product attached to the bus in order to add it to the project.
Once devices are added to the project, it is then necessary to assign bit inputs and bit outputs to oprands, which is analogous to wiring up switch inputs or auxiliary contacts to a PLC’s terminals, except done electronically. There is an autoassign feature which will assign all devices sequentially, although it does not assign every possible input or output in all cases and in other cases of revising the oprand table, old comments can also be left behind.
The program can then be designed in ladder logic on the circuit diagram screen. This can be imagined as the two voltage rails on either side, with switches (inputs) on the left, leading to coils (outputs) on the right – if a path can be drawn through, a circuit is created and the coil is energized. It is, in essence, a representation of relay-logic schematics which requires thinking somewhat differently to sequential programming which I am more comfortable with. The rule-based logic is arranged in “rungs” with up to five inputs/outputs in each rung and up to 256 rungs for this particular controller. Despite appearing to operate in real-time with parallel inputs, rungs are rapidly executed sequentially.
When executing the program on the PLC with it attached in online mode, various diagnostics are possible. This includes red highlight on the rung diagram to show the current logic status based on current inputs.
It is possible to show a virtual LCD screen as well with inputs and outputs, although more helpful is the inputs page which allows you to view all possible inputs and their current status.
It is also possible to view the SWD bus live and see which contactor is pulled in (green) and which switch LED is on (white).
An oscilloscope function is also available, but first the configuration needs to be set which maps which oprands to graph on which channel. An example of this is shown in the next section.
easySoft Pro also features a simulator, although in the case of using it with the EASY802, this did not seem to work as the simulator only supports simulating changes to the first 16 I/Os which represent the “reserved” internal I/O that is not fitted to the EASY802. I could not find a way to simulate input changes to higher numbered I/Os which are those assigned to SmartWire-DT devices, as a result, I worked in “online” connected mode with the PLC to debug my designs.
I wanted to see what I could get the system to do that would be remotely interesting and perhaps make some use of the “up to 256-rungs” capability of the controller (with each rung capable of having up to five I/Os referenced). Owing to a lack of direct inputs and outputs which limited the possibilities, I came up with an example that used 13 rungs (leaving rungs 7-9 empty). Given this is my first real attempt to do anything with ladder logic, forgive me if my designs are in any way deficient.
I discovered that the way the order that the rung diagram is drawn is very important as it affects the order of operation (as the PLC scans sequentially through the rungs) but also where wires are drawn crossing other rung lines are assumed to be “connected” which resulted in some strange issues at first. Luckily using the select tool and cutting/pasting the circuits allowed for rearrangement to create some free space to add in additional rungs and interconnections – but leaving some free rungs next to sections could be advisable.
In the above program, rungs 5 and 6 are seal-in logic which performs the industrial “start-stop” function, switching the contactor on or off depending on the button pushed. As I wanted the LED to flash at a fast rate when on and a slow rate when off, Q1 state enables the two relevant timers in rung 3 where if Q1 is not energized, slow-flash T01 is enabled and rung 4 where if Q1 is energized, rapid-flash T03 is enabled. In rungs 1 and 2, the outputs from both timers T01 and T03 are connected to the switch LED Q19.
Rungs 10 through 16 implement what I call a “lamp torture test”. Rungs 13 and 14 implement an industrial start-stop function with seal-in logic but using the virtual relay marker M01 to hold the state of whether the test is running or not. Rung 15 tests if the marker is set, if it is, then the cyclic relay timer T04 is enabled which through rung 16 directly drives the contactor output Q18. Rung 12 enables T02 which provides the LED flashing when the test is not running via rung 10, but if the test is running, rung 11 causes the LED to flash in synchrony with the contactor output Q18.
It may seem somewhat cryptic at first – but if you watch the video, things should make sense.
The easySoft Pro software also allows you to view the function block plan which shows the four timers with their key variables.
It is possible with the oscilloscope feature to monitor the status of the outputs and inputs in quasi-realtime on a connected computer. At first, the oscilloscope has to be configured with the oprands to monitor, but once started, statistics on cycle time appear in the bottom and a display of the logic levels is visible in a scrolling graph.
The test program and report is attached as a .zip file to this review. The following video shows the functioning of the example program I built above:
While I had intended to do some testing of the response times and cycle times of the PLC, unfortunately, this didn’t prove possible as there are no integrated digital or analog I/O on this unit nor were any digital/analog I/O interfaces part of the test package. With only two pilot buttons and two contactors, there isn’t much else that could be done.
I was interested to know what is happening on the bus itself while the SmartWire-DT system is working. I know from experience that the bus is fairly robust – removing the terminator on such a short bus does not stop the bus from working which was a surprise, but also removing any SmartWire-DT peripheral causes an almost immediate halt to operations (if the device is mandatory) and reattaching the lost device results in a very short delay before the bus is bought back to full operation. We know from the specifications that it is RS-485 based, however, details of the protocol appear to be proprietary to Eaton.
I thought I’d set my Rohde & Schwarz RTM3004 with the latest V.1.400 firmware a good challenge by using it to sniff the bus with an adapter I built to break-out the bus (which, incidentally creates a “stub” of wire on the bus which would also affect transmission integrity).
It seems that only the Data A and Data B lines are of much use – the Node Select line appears always to have 15V on it. Indeed, it seems that RS-485 is being used, as a differential signalling scheme can be seen. The bursts of data happen continuously and repeatedly, lasting about 1.2ms and spaced about 2.25ms apart at the most, but sometimes even less.
Looking at a single train, it seems that the bus “idles” at around +3.3V and +1.6V. Once the transmission is to start, the transmitter pulls the voltage up to +4V and down to +1V on the pair before starting a “packet”. During the packet, it seem that other peripherals are transmitting bits/bytes in their respective slots with a difference in transmission levels seen at about 850us.
Using the UART decode feature, we can see it seems to consist of a train of data beginning with 0x55555555 which may serve as a “reset” or “synchronize” preamble for all the downstream peripherals.
Looking closer at the details and we can see there seems to be some “jitter” as to where bytes begin and end and the spacing between them. The bus levels appear to return to +3.3V/+1.6V when the bus is undriven, which is then promptly pushed back up/down to +4V/+1V when the associated peripheral begins its transmission. It seems very much to be a master-initiated polling system which happens presumably at each cycle of the PLC.
Additional details including the semi-full trace of I/O during initialisation can be found in my New Year’s Grab Bag RoadTest Extras blog.
The Eaton SmartWire-DT system de-clutters control panel wiring by utilising an RS-485-based digital bus with custom ribbon wiring and distributed logic to replace point-to-point wiring. The idea is valuable as the reduction in cable clutter, ease of installation, reduction in installation time, ability for the system to self-diagnose and be expanded are real benefits.
The downside of the SmartWire-DT system was the seemingly proprietary nature of the system including the cabling, crimp connectors, crimp tools and modules. While Moeller/Eaton weren’t the only vendor of compatible modules (there was also Phoenix Contact, Falk, Dajo and Wohner), using the SmartWire-DT system really restricted component choice and potentially increased costs due to a lack of competition. Crucially, in order to make your investment in programming software/hardware and tooling pay off, you would have to be installing a fair number of these systems.
But even without this barrier, it didn’t seem that panel builders and technicians would have fully appreciated the capabilities of the SmartWire-DT system. Being different from directly-wired PLCs in system architecture and troubleshooting, servicing and maintenance requires a different approach. A simple workaround (e.g. a jumper wire across an input) could be much more difficult to implement and require modification of the programming of the PLC (requiring the software, cable and loaded code). Likewise, a failure of a component would lead to swapping another into the system, but until its address is configured and unless it is an exact replacement (or the program specifies compatible devices are permitted), the whole program may be stuck in an error condition. Some diagnosis and simulation tools are available, but few technicians are likely to have been trained on it. Where industries cannot afford downtime, this is a particular disadvantage, as with a conventional system any failing contactor could just be swapped out wire-for-wire in five minutes with no technical expertise required.
The 125/250kBit/s RS-485-based serial nature of the SmartWire-DT bus also potentially causes performance issues if very high speed/synchronised control is necessary. Traditional “directly wired” systems may be able to process I/O more rapidly (or in parallel) as they are not bound by the rate of the bus. One upside to the distributed bus system is that the controller keeps a constant watch over the network to ensure it is configured correctly and that all required elements are connected. This way, there is no way the system can operate with critical elements missing (whereas, a disconnected wire in a directly-wired PLC would not trigger an error).
easySoft Pro was quite intuitive to use, after doing a quick “walk around” and thinking in “ladder-logic” mode. The program load times do take a minute or so, but the software is relatively stable and works with Windows 10. Creating a design is as simple as enumerating the devices, adding them to the design, assigning the I/O oprands, drawing a ladder logic diagram and loading it to the PLC. While there are simulation tools available, they were not able to simulate I/O for the SmartWire-DT bus making it impossible to simulate designs intended for the EASY802 which as no direct I/O which can be simulated. The availability of the oscilloscope tool allows for a connected PC to directly observe the status of inputs and outputs to verify correct operation of the PLC and aid in troubleshooting. I was able to implement a simple lamp torture test, industrial start-stop with flashing LED indicators on the switch utilising a number of virtual timer relays and a marker without any great difficulty and learnt a lot about ladder logic in the process. That being said, there are some subtle bugs with the oprand autoassign features and simulation does not appear to work for SmartWire-DT based inputs as the simulation only features input toggles for the first 16 input oprands (i.e. the onboard I/O that the EASY802 does not have).
As of 2019, the Easy800 series products are not widely available and stocks of SmartWire-DT products are low to non-existent with fairly steep prices. Those who may have bought into the SmartWire-DT system and need spare parts may eventually revert to a conventional system as a result. Luckily, some parts can be reused in a conventional system by removing the adaptor module. While the documentation for SmartWire-DT remains online under Eaton’s webpages, Swire-D.com is no longer accessible and there isn’t even a Wikipedia entry about it.
From this, I can only conclude that SmartWire-DT was an attempt by Eaton to see if they could popularise their own light-weight fieldbus for panel wiring. It seems that the industry ultimately was not too interested and the product is in the process of fading away having not displaced the traditional point-to-point wiring methods (as they had hoped) in any significant way. It seems that for larger installations, there is a trend towards “modular” PLCs that competitors (such as Siemens) sell, running on standardised ProfiBUS/ProfiNET, CANopen or some other field-bus, enabling them to interoperate with any other bus-compatible partner.
As always, thanks to element14 for the opportunity to RoadTest the Eaton SmartWire-DT system - if you found this review interesting or informative, feel free to leave a comment, like, share, bookmark or rate the review.
Updated 6 Mar 2019 - Incorrect reference to RS-422 due to typo fixed, missing paragraph break fixed.