Lucid vs WITS-DNP3 shootout

Both the Lucid and WITS-DNP3 protocols are managed by the WITS-PSA. Why have two protocols? What are the differences? Why choose one over the other? This article tries to answer those questions. More background on Lucid and the WITS-PSA is provided in this previous article in our series on Lucid.

Some background

WITS-DNP3 was born in the traditional world of SCADA and Telemetry Systems with Remote Telemetry Units (RTUS) and (Programmable Logic Controllers (PLCs)). It catered well for users’ systems which were generally in-house and separated from IT within the Operational Technology (OT) estate. After many years of successful use within the UK Water Industry, those users have now become accustomed to the consistent set of functions offered by all WITS-DNP3 devices.

Over the years, pressure grew for a simpler-to-implement protocol which would increase the number of device vendors, giving users more choice of IoT-like devices to monitor much more of their infrastructure. In this instance we are associating IoT-like devices with devices which are normally lower cost and high volume.

Lucid was developed as a result of these pressures and is targeted at simple IoT-like devices. As far as possible, Lucid preserves most of the functionality available from WITS-DNP3 whilst also improving other aspects, which we will cover in this article. This approach was deliberately chosen in order to allow the many users, already familiar with WITS-DNP3, to directly apply their knowledge to the use of Lucid. Thus, Lucid would preserve that investment in people’s knowledge, whilst enhancing other aspects of functionality.

Lucid was not designed as a replacement to WITS-DNP3, although, as we shall see, they have some considerable common ground. Each protocol may be the better selection in different scenarios. This article should provide you with the background to make such a selection. The WITS-PSA firmly believes that both protocols have a place within OT systems; indeed, we think that many users may end up using both protocols.

A functional comparison

Prior to the development of WITS-DNP3, UK Water organisation users developed 97 requirements for the new protocol. WITS-DNP3 fulfilled most of those user requirements.

Over a decade later, the sub-committee of the WITS-PSA which developed Lucid, assessed the protocol against those original requirements. The following figure and tables present a summary of this analysis for both WITS-DNP3 and for Lucid.

Lucid fulfils 71 out of the 97 possible requirements whilst WITS-DNP3 fulfils 83 out of the 97 requirements. Only 9 requirements are not met by either Lucid or WITS-DNP3. This is shown in tabular form below:

 Number of Requirements% of total requirements
Neither meet99%
Lucid meets7173%
WITS-DNP3 meets8386%
All requirements97100%

In the following Venn diagram and table, we extend the analysis to show which functions are implemented by both protocols and which by only one.

venndiagram
 NeitherLucid OnlyBothWITS-DNP3 OnlyTotal
Number of requirements95661797
% of requirements9%5%68%18%100%

Most of the functions, 66 as shown in the Venn diagram, are provided by both Lucid and WITS-DNP3, with WITS-DNP3 providing 17 functions which Lucid does not provide and Lucid providing 5 functions that WITS-DNP3 does not provide.

The differences between the requirements implemented by Lucid and WITS-DNP3 are summarised below:

  • WITS-DNP3 supports a fuller communication regime between the Supervisory Application (called a Master Station in DNP3 terminology) and the Field Device. WITS-DNP3 can initiate direct connection to Field Devices if they are capable and it can hold continuous communications sessions with devices. It also supports the configuration of complex communication regimes with detailed fallback capabilities. As Lucid is based on MQTT, which is by its nature asynchronous, these same capabilities do not exist, with Lucid defining that the Field Device initiates communications to the MQTT broker.
  • WITS-DNP3 supports more secure point control. As direct connections are supported by WITS-DNP3 it is capable of things like Select – Checkback – Execute control of point values. The use of MQTT in Lucid prevents this.
  • WITS-DNP3 supports pulsed digital point control whereas Lucid does not contain this functionality in its first release.
  • WITS-DNP3 supports accurate methods of time synchronisation. Lucid expects devices to synchronise their clocks independently of the MQTT protocol, using standards such as SNTP, NITS or GPS. However, Lucid does provide a method of time synchronisation, which only works if both the Supervisory Application and the Field Device are connected to the broker at the same time. WITS-DNP3, on the other hand, can use the DNP3 provided methods of time synchronisation which are likely to be more accurate than Lucid’s simple method.
  • Lucid supports various types of non-destructive log reading. WITS-DNP3 does not support this.
  • Lucid supports encryption. This is not provided natively by WITS-DNP3.
  • Lucid supports options for enhancing the protocol. As WITS-DNP3 is based on DNP3 we are normally dependent on developments within that protocol to allow extension for WITS-DNP3.

Other notable differences

Some other notable differences that it is worth considering are:

  • Lucid does not require a Configuration Application (CA) to share configuration with the Supervisory Application. Instead, all configuration is provided as part of the protocol, leading to full “discoverability” and plug’n’play capabilities.
  • Lucid can handle multiple Supervisory Applications communicating with and configuring the Field Device, whereas WITS-DNP3 supports only one Master Station for any given Field Device.
  • Lucid provides a method of suggesting user interface configuration to a Supervisory Application to assist in a sensible configuration GUI layout.
  • Lucid provides a wider range of deployment scenarios dues to its use of MQTT. So, for example, Field Devices that do not implement Lucid natively, could appear as Lucid devices to a Supervisory Application through an interim server.

In summary

Lucid provides a new protocol suited to smaller devices with IoT-like functionality. It enables users to mass deploy devices whilst maintaining control of installation and configuration costs. Despite being aimed at smaller devices, Lucid still provides most of the functions available from its slightly more capable cousin WITS-DNP3. Lucid also shares terminology and its data model with WITS-DNP3, thus permitting staff with existing experience in WITS-DNP3 to easily adapt to Lucid.

One might imagine that with WITS-DNP3 implementing more requirements that Lucid, is the better protocol? However, such a comparison should be considered too simplistic. Both Lucid and WITS-DNP3 implement requirements that the other does not and also share in implementing a large majority of the requirements. The small differences in requirement implementation speak in the main to the different ways in which the devices would be deployed. Some scenarios in which a user may wish to deploy Field Devices may better suit one protocol and its capabilities rather than the other; in those cases, users should select the more appropriate protocol. However, for many scenarios the user may have a free choice as to which protocol they can select. One protocol is not better than the other, just more appropriate in certain situations.

The few differences between the protocols have been discussed in this article and that will hopefully allow users to select the protocol most suitable for a given application. One important difference not considered yet is that of product availability. WITS-DNP3 has been around for a good number of years and several products which implement it are available to procure right now. Lucid is a new protocol and we are yet to see the first devices come to market at the time of writing. On the other hand, Lucid has been designed to be an easy-to-implement protocol which will hopefully mean it sees adoption with many vendors, in relatively short timeframes, after its introduction.

Dave Howarth (Northumbrian Water) and Mark Davison (Terzo Digital)

January 2024


Please see our Lucid reading page for a collated list of other articles on Lucid.