On February 5 this year, a new standard for 10-Mbit Ethernet was approved. Yes, you read that right: ten megabits per second.
Why is such a “small” speed needed in the 21st century? To replace the zoo, which is hidden under the capacious name "field bus" - Profibus, Modbus, CC-Link, CAN, FlexRay, HART, etc. There are too many of them, they are incompatible with each other and relatively difficult to configure. And you just want to plug the cable into the switch, and that's it. As with normal Ethernet.
And soon it will be possible! Meet: “802.3cg-2019 — IEEE Standard for Ethernet — Amendment 5:Physical Layer Specifications and Management Parameters for 10 Mb/s Operation and Associated Power Delivery over a Single Balanced Pair of Conductors.”
What is interesting about this new Ethernet? First of all, it works on one twisted pair, not four. Therefore, it has fewer connectors and thinner cables. And you can use the already laid twisted pair cable going to the sensors and actuators.
You might argue that Ethernet works up to 100 meters, and the sensors are located much further. Indeed, this used to be a problem. But 802.3cg works at a distance of up to 1 km! One pair! Not bad?
In fact, it’s even better: power can also be supplied through the same pair. That's where we'll start.
IEEE 802.3bu Power over Data Lines (PoDL)
I think many of you have heard about PoE (Power over Ethernet) and know that 2 pairs of wires are needed to transmit power. Power input / output is made at the middle points of the transformers of each pair. It's impossible to do this with just one pair. So I had to do it differently. Exactly how is shown in the figure below. For example, classic PoE has also been added.
Here:
PSE - power sourcing equipment (power supply)
PD - powered device (device at the far end that consumes electricity)
Initially, 802.3bu had 10 power classes:
Three conditional gradations of the source voltage are highlighted in color: 12, 24 and 48V.
Legend:
Vpse - power supply voltage, V
Vpd min - minimum voltage on PD, V
I max - maximum current in the line, A
Ppd max - maximum power consumption PD, W
With the advent of the 802.3cg protocol, 6 more classes were added:
Of course, with this diversity, PSE and PD must agree on the power class before applying full voltage. This is done using SCCP (Serial Communications Classification Protocol). It is a low speed protocol (333 bps) based on 1-Wire. It works only when the main power is not applied to the line (including in sleep mode).
The block diagram shows how power is applied:
- a current of 10mA is applied and the presence of a 4V zener diode at the other end is checked
- power class is negotiated
- main food is served
- if the consumption drops below 10mA, sleep mode is activated (standby power supply 3.3V)
- if the consumption exceeds 1mA, the sleep mode is exited
Coordination of the power class can be omitted if it is known in advance. This option is called Fast Startup Mode. It is used, for example, in cars, because there is no need to change the configuration of the connected equipment.
Both the PSE and the PD can initiate a sleep transition.
Now let's move on to the description of data transfer. It is also interesting there: the standard defines two modes of operation - long-range and for short distances.
10BASE-T1L
This is a long range option. The main characteristics are as follows:
- range - up to 1 km
- conductors 18AWG (0.8mm2)
- up to 10 intermediate connectors (and two terminal ones)
- point-to-point operation mode
- full duplex
- symbol rate 7.5Mbaud
- PAM-3 modulation, 4B3T coding
- signal with amplitude 1V (1Vpp) or 2.4V
- Energy Efficient Ethernet (“quiet/refresh” EEE) support
Obviously, this option is intended for industrial applications, access control systems, building automation, elevators. To control rooftop chillers, air conditioners, fans. Or heating boilers and pumps located in technical rooms. That is, a lot of all sorts of applications besides industry. Not to mention the internet of things (IoT).
It is worth mentioning that 10BASE-T1 is only one of the Single Pair Ethernet (SPE) standards. There are also 100BASE-T1 (802.3bw) and 1000BASE-T1 (802.3bp). True, they were developed for automotive applications, so there is only a range of 15 (UTP) or 40 meters (STP). However, long-range 100BASE-T1L is already in the plans. So in the future they will add auto-negotiation of speed.
In the meantime, coordination is not used - a “quick start” of the interface is declared: less than 100ms from power on to the start of data exchange.
Another option (optional) is to increase the transmission amplitude from 1 to 2.4V to improve the signal-to-noise ratio, reduce the number of errors, and counteract industrial interference.
And, of course, EEE. This is a way to save electricity by turning off the transmitter if there is no data to transmit at the moment. The diagram shows what it looks like:
No data - send the message “I went to sleep” and disconnect. Occasionally we wake up and send the message “I'm still here”. When the data appears, the “I'm waking up” warning is given to the opposite side and transmission begins. That is, only receivers are constantly working.
Now let's see what they came up with the second version of the standard.
10BASE-T1S
Already by the last letter it is clear that this is a protocol for short distances (short reach). But why is it needed if T1L also works at short distances? Reading the specs:
- range up to 15m in point-to-point mode
- duplex or half duplex
- проводники 24-26AWG (0.2-0.13мм2)
- symbol rate 12.5Mbaud
- DME, coding 4B5B
- signal with amplitude 1V (1Vpp)
- up to 4 intermediate connectors (and two terminal ones)
- no EEE support
It seems to be nothing special. So what is it for? But for this:
- range up to 25m in multipoint mode (up to 8 knots)
And this:
- PLCA RS (PHY-Level Collision Avoidance Reconciliation Sublayer) operation mode
And that's much more interesting, isn't it? Because it helps significantly reduce the number of wires in control cabinets, machines, robots, and cars. And there are already proposals to use it as an I2C replacement in серверах, switches and other electronics.
But the multipoint mode has disadvantages. The main one is a shared data transmission medium. Of course, collisions are resolved using CSMA/CD. But it is not known what the delay will be in this case. And for some applications, this is critical. Therefore, in the new standard, multipoint was supplemented with a special PLCA RS mode (see the next section).
The second drawback is that PoDL does not work in multipoint. That is, the power will have to be supplied via a separate cable or taken somewhere on the spot.
However, in point-to-point mode, PoDL also works on T1S.
PLCA RS
This mode works as follows:
- nodes distribute identifiers among themselves, the node with ID=0 becomes the coordinator
- the coordinator issues a BEACON signal to the network, indicating the beginning of a new transmission cycle and transmits its data packet
- after the transmission of the data packet, the queue for transmission passes to the next node
- if the node has not started transmission within the time required to transmit 20 bits, the queue moves to the next node
- when all nodes have transmitted data (or missed their turn), the coordinator starts a new cycle
In general, it resembles TDMA. But with the peculiarity that the node does not use its time frame if it has nothing to transmit. And the frame size is not hard-coded, because depends on the size of the data packet transmitted by the node. And it all works on top of standard 802.3 Ethernet frames (PLCA RS is optional, so should be compatible).
The result of using PLCA is at the bottom of the graphs. The first is the delay depending on the load, the second is the throughput depending on the number of transmitting nodes. It is clearly visible that the delay has become much more predictable. And in the worst case, it is 2 orders of magnitude smaller than in the worst case of CSMA / CD:
And the channel capacity in the case of PLCA is higher, because not spent on resolving collisions:
Connectors
Initially, they chose from 6 connector options offered by different companies. As a result, we settled on these two options:
For normal use, the IEC 63171-1 LC connector from CommScope was selected.
For harsh environments, the IEC 63171-6 (formerly 61076-3-125) connector family from HARTING. These connectors are designed for degrees of protection from IP20 to IP67.
Of course, connectors and cables can be either UTP or STP.
Other
You can use a regular four-pair Ethernet cable, using each pair for a separate SPE link. In order not to pull four separate cables somewhere far away. Or use a single-pair cable, and put a single-pair Ethernet switch at the far end.
And you can connect this switch directly to the local network of the enterprise, if the network has already been extended to far distances via fiber optic. Stick sensors into it there, and look at the readings from them here. Directly over the web. Without interface converters and gateways.
And it doesn't have to be sensors. There may be video cameras, intercoms or "smart" light bulbs. Actuators of some valves or turnstiles at checkpoints.
So the prospects are interesting. It is unlikely, of course, that SPE will replace all field buses. But he will bite off a fair amount of them. Certainly in cars.
PS I did not find the text of the standard in the public domain. The above information was collected bit by bit from various presentations and materials available on the Internet. So there may be inaccuracies in it.
Source: habr.com
