Max-i Fieldbus

Title
Go to content
Welcome to the official homepage for the Max-i fieldbus and Max-i Association
Overview
Max-i is a new, extremely powerful and efficient multi-master fieldbus, which enables the lowest automation and LED-lighting costs ever seen combined with maximum performance! It may be regarded as a combination between a highly improved CAN bus with a high-power 20-V power supply running on standard installation cables and some hardware-based I/O objects to handle the most commonly used functions like UART interface to microprocessors, SPI interface to for example sensors, displays and servo motor controllers, 4-bit Boolean (on/off) I/O, A/D and D/A conversion, lock control with keypad scanning and very advanced LED driving. Max-i may be used for virtually all low to medium speed applications (up to 25,000 telegrams per second on short lines) like:

  • Green smart house systems and building automation.
  • Intelligent LED lighting including demanding stage light, architectural lighting and traffic lights.
  • Industrial automation including motion control, robots, cobots and safety applications.
  • Transportation applications such as automotive, special vehicles, railway, ships and aerospace.
  • Internet of Things - IoT.
  • Military applications including equipment, which must survive very strong electromagnetic pulses.

Max-i has a lot of outstanding benefits:

  • It is possible to make a complete bus interface in one integrated circuit (IC), which is small and cheap enough to be built into even the smallest and most price sensitive actuator, sensor or lamp, gets its supply voltage of 20 V (15.4 - 24.6 V) directly from the bus and only needs an absolute minimum amount of external components - often only a few small ceramic capacitors.
  • Because the bus interface and protocol is handled entirely in hardware, there is no needs for an expensive and time consuming conformance test.
  • The hardware-based architecture gives a much higher reliability and failure tolerance than any fieldbus, which need a microprocessor on the sensor and actuator side, like all systems, which use high-level protocols or layers like OPC UA, TCP, UDP, Thread, Matter etc. The present Max-i implementation uses only approximately 5,800 gates corresponding to approximately 40,000 transistors. As a comparison, even the smallest ARM Cortex-M0 microprocessor uses approximately 12,000 gates corresponding to approximately 70,000 - 100,000 transistors, and if just a single transistor fails or there is a bug in the program, you can lose everything. With a single chip Max-i interface, you will only lose the function(s) that transistor is a part of and there is no software to go down. Besides, the use of a unique dual-phase clocking scheme for all flip-flops makes Max-i much more tolerant to changes in transistor data and single-event upsets (SEU) than any microprocessor.
  • It is possible to use cheap, standard, unshielded and un-terminated installation cables instead of special communication cables. This saves a lot of money and troubles with shield connection and shield currents and makes it possible to transfer much more power over the bus than any other fieldbus - up to several kW.
  • It uses bit-wise bus arbitration, which has many very important benefits compared to any other access method:
      • It is the only multi drop technology (not point-to-point), which does not need any reconfiguration when devices are added or removed, so this may happen at any time on the fly. In all other systems, a new device needs to be inserted in the communication sequence before any direct communication is possible, and if a device is removed without reconfiguration or fails, time is lost forever.
      • The response time is much faster for the same speed as a device does not need to wait for a poll, a token or a time slot before it is allowed to send a message.
      • The bandwidth utilization is much better since no time is lost on devices, which do not need communication.
      • The failure tolerance is much higher since there is no dedicated master or leader unit and no need for a centralized database. Since Max-i uses voltage clamps in each device instead of resistor termination, the cable may even be cut in two or more parts and each part can still work if it is powered up.
      • It is possible to use many different protocols simultaneously. It is for example possible to run any CAN protocol like DeviceNet or CANOpen together with the Max-i protocol and in this way save extra cables and communication channels.
  • Unlike other fieldbus systems based on bit-wise bus arbitration like CAN, the signal integrity is always very good:
      • All bits are 100 % symmetrical even under bus arbitration (dominand and recessive bits) so unlike for example CAN there is no bias distortion and no problem with ringing on recessive bits.
      • All signal levels use a virtual midpoint between the supply rails L+ and L- as reference and therefore do not change due to voltage drops as long as the cross section of L+ and L- is the same.
      • The triggering levels are always in the ideal point in the middle between the positive and negative pulse levels no matter the signal attenuation. In CAN they are typical fixed to 0.5 V and 0.9 V compared to the negative supply rail, which may cause asymmetry both in case of varying signal amplitude and in case of voltage drops.
      • There is very limited oscillation due to propagation delay between devices during bus arbitration and it is not necessary with a fairly high driver output resistance (loss) to limit this to an acceptable level.
      • It has an excellent ratio between signal and noise and between transmitted power and lost power.
      • The signal integrity does not depend on termination resistors and is therefore usually not destroyd in case of a line break.
  • It uses the very efficient publisher-subscriber model, where it is the various values, which have an address/identifier, not the various devices. There are many very important benefits of this:
      • The same value may be utilized simultaneously by any number of devices.
      • Since the majority of telegrams only contains simple, standardized values, all features of present and future devices can be utilized without any limitations. It is just a matter of reading the data sheet of a device to see which data it publishes and which data it subscribes to. It is therefore not necessary to specify and standardize device profiles for lamps, blinds, HVAC control, door locks etc. so there is no need for complex abstraction layers like OPC UA or Matter.
      • It is not necessary to add a long list of source and destination addresses when gateways are used.
If the publishers and subscribers (clients) are not connected to the same bus, MQTT brokers (see https://en.wikipedia.org/wiki/MQTT ) may be used to establish world-wide communication over TCP/IP and decouple the various clients so that they need not to be online simultaneously.
  • It is extremely efficient:
      • With the short 12-bit identifier, a 4-bit value can be transmitted in only 5 bytes including a 20-bit CRC-check to detect errors, a 7-bit Hamming code on the identifier to protect against masquerading and an optional 7-bit telegram serial number for safety applications. Even with the long 31-bit (PNS) identifier, a process value with a 20-bit mantissa and a 6-bit exponent, scaled to SI-units only requires 9 bytes.
      • It does not take longer time to poll a value than to send it event driven if the value is generated by the Max-i controller or loaded into it through the SPI interface.
      • It is possible to transmit data to more devices in the same telegram for example to stage lamps and to servo axes in CNC machines, robots and cobots. On a 25 m trunk line, Max-i is able to synchronize over 16 servo motor axes with 32-bit precision to an accuracy of 0.1 μs at an industrial state-of-the-art communication cycle of 400 μs. This is much faster than CAN and close to or even better than the performance of a much more expensive and complex Ethernet solution, which also has much lower reliability due to the necessary routers and other complicated electronics. With Max-i, it is no problem to get robots to work together as cobots as they can just use the same telegrams.
      • It uses synchronous communication with a speciel, self-clocking bit coding, which is 1.2 times more efficient than Manchester coding during bus arbitration and 1.5 times more efficient after that and cannot get out of phase synchronization.
Efficiency is a much better means to get many data through than speed because it does not reduce the signal-to-noise (S/N) ratio. If for example the probability of a bit failure (p) is 10-7, which is usually accepted in a field bus system, just a factor 2 (N) increase in speed will reduce this to approximately p(1/N) = 3.2 x 10-4, which is unacceptable. Ethernet may be fast, but for many practical applications, the S/N may be serveral hundred times lower for the same throughput due to the enormous overhead - especially if IPv6 is used!
  • It is extremely safe. It is the first fieldbus designed directly for highly demanding safety applications according to IEC 61508 SIL 3 (death of 1 - 3 people) without the use of additional protocol layers or safety monitors, and it is fully deterministic. A "babbling idiot protection" prevents high priority devices from taking over the bus and if all devices want to transmit simultaneously, they just get through one by one.
  • It combines the outstanding performance with a remarkable and unusual simplicity that will please everyone from makers to professionals. There are just a few registers to setup and it is not necessary to add complicated layers like OPC UA to make it possible to interpret the data. All values have a specified data types and may be scaled to SI-units and a new numbering system (PNS) makes it possibe to identify them directly as properties of the equipment to which they belong like for example HX127AT4 for Heat Exchanger 127A Temperature 4. The Max-i specification fills only approximately 220 pages where approximately half is used for background material and annexes. As a comparison, most other fieldbus systems have specifications way over 1000 pages and OPC UA consists of 14 documents with a total of 1250 pages!

"Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away."
Antoine de Saint-Exupéry

The name Max-i means Multiple Access Cross (X) coupled interface. It refers to Max-i being a multi-master bus, which for industrial applications, railway applications, stage light control and long distance communication uses a cable with 4 conductors connected as a balanced 4-wire line (X-coupled) to prevent mutual coupling to other cables. The name also refers to the Max-i-mum performance, which by far exceeds comparable fieldbus systems on most fields.

Released specifications can be downloaded from the Technical Description page and used free of charge for all non-commercial use. Commercial use requires membership of Max-i Association.

News
Interface/control module Max-i IF2 has been released.



Specification 13.0 has been released.
  • Added possibility for quadrature encoder input to LAMP object for "analog" adjustment.
  • Maximum smoothing time increased for better usability for inrush current limiting.
  • Serial number format modified and enabled to be used as auxiliary device identifier.
  • Out-of-the-box UART speed selection added.
  • 3 lower speeds added (1.2 kbps, 2.4 kbps and 4.8 kbps) to enable long distance communication on thin cables.
Max-i Association
C/O Innovatic
Bøgebakken 3, Gjessø
8600 Silkeborg
Denmark
Phone: (+45) 86 84 72 92

Chairman Carsten Kanstrup
Mobile: (+45) 20 15 51 60
This page is created with WebSite X5 and updated March 27th 2023

Back to content