Results 1 to 5 of 5
  1. #1
    Join Date
    Oct 2014
    Location
    Columbus, Ohio
    Posts
    1,500
    Post Likes

    Ethernet as a fieldbus

    Hey controls guys, I'm just starting to learn this stuff, and am confused and intimidated by the vast array of protocols and differing physical layer options for BAS and industrial control. Coming from a computer and networking background, () can you guys school me as to why we don't just use Ethernet for everything? It's well known and understood by many people and every building has an Ethernet network anyways.

    Here's all the pros and cons that I could think of, from my perspective as a Internet networking kind of guy. Feel free to let me know where I am off base or if I am completely wrong on some of this.

    Pros for Ethernet and TCP/IP control:

    • Standard, well understood physical layer connections (any network guy could crimp you up a replacement cable if yours broke).
    • Possible to get power via PoE, if necessary, although I kind of assume most controllers have their own source of power.
    • Essentially unlimited addressing - you will never run out of points. IPv4 when routed internally can make use of the 3 private IP blocks of 172.16.0.0/12, 192.168.0.0/16, and 10.0.0.0/8 as well as 4096 VLANs. IPv6 has literally billions of internal IP addresses and you could even get a /48 or /64 subnet from an ISP and have a public IP for every sensor and control device if you wanted to.
    • No collisions if you implement a switched network, which is possible with the plethora of inexpensive switches from many manufacturers.
    • Fast and high bandwidth transfers.


    Cons for Ethernet and TCP/IP control:

    • Short distance limitations unless you implement repeaters, switches and/or fiber optics (which are much more difficult to terminate and are more fragile).
    • Not deterministic, if your machine needs extremely tight control or timing.
    • Daisy-chain topologies are not possible.
    • Physical layer takes up more space because of additional wires needed (4 twisted pairs).
    • Somewhat more susceptible to noise and crosstalk (except fiber, but see point 1).


    So what do you guys say? I think most of the limitations are less of a problem for a BAS environment but may be an issue for industrial process control.

  2. #2
    Join Date
    Feb 2005
    Location
    Edmonton, AB Canada
    Posts
    838
    Post Likes
    Good day rjk_cmh,

    Indeed, Ethernet has a lot of benefits and is extremely well supported by a huge number of companies that provide a lot of equipment. However, pushing Ethernet all the way down the smallest controller/sensor level I can see a number of issues:

    1. Home-run (back to a switch ... etc) wiring
    One of the larger project costs is the labor/effort to install wiring and/or conduit. If one has 100 or so controllers and each one needs to be wired back to a switch, then I can see this cost going through the roof and potentially a lot of space (conduit), as in many cases all wiring has to be placed in conduit. Secondly with the price of wiring these days, the physical amount of wiring needed would also be quite a bit larger... which also adds to cost.

    Daisy-chaining minimizes this overall installation cost dramatically.

    Using wireless like Wifi is an option, however, at the Wifi frequencies uses there are limitations with regards to signal penetration and integrity in the areas where a lot of controls are installed... The higher the frequency the harder it is for the signal to penetrate concrete, stone, etc... unless bumps up the transmitted power... but the problem here is that this is in the Microwave region and too much power results in some serious issues with regards to living creatures.

    2. Communication length limitations.
    As you stated, Ethernet is technically limited to 100m unless one uses Fibre or Wireless, etc, whereas RS485 is spec'd at 4000 feet (assuming the data rate is < 100Kbaud).

    3. Ethernet adds a reasonable amount of $$$ to a product. There are a number of components that are necessary for a wired Ethernet interface (Magnetics, physical layer interface, etc) as well as the additional resources (memory, clock speed, etc) needed by the controller's processor... translation... bigger and more featured processors costs more $$$. This also adds to higher assembly and testing costs, higher support costs, but also the increased effort and cost of the operating firmware (Ethernet require some form of a protocol stack, etc). Another cost is the increased Governmental testing cost, as the bigger/faster processor means that the product will most likely radiate more electrical emissions and so add more costs to deal with all this.

    4. Adding wireless Ethernet is an option... but this tends to be costly from both a component and testing perspective, but also time to implement. Governmental radiated emission testing on devices that actually transmit wireless information (i.e. intentional radiators) increases exponentially... As an example radiated emission testing is around $5K a pop... intentional radiator testing... around $50K... a pop...

    5. Reliability
    Ethernet can be less reliable, as the physical layer interface (PHY) tends to consume a fair amount of power and the offshoot of this is heat dissipation. Statistically the cooler a semiconductor is operated, the longer it will live... In fact if you reduce a semi's temp by 30C its lifetime will increase 10 times (statistically). Since Ethernet interfaces are power intensive they tend to not live as long as interfaces that generate less heat (i.e. serial, etc). Again this is just statistics and so your mileage may vary.

    Another point is that with Ethernet there would be a need for more secondary components like switches, etc. Again statistics tell us that more components lowers the reliability of the system.

    Anyway, I am sure that others will comment, but these are few of the items I see.

    Cheers,

    Sam

  3. #3
    Join Date
    Oct 2014
    Location
    Columbus, Ohio
    Posts
    1,500
    Post Likes
    Thread Starter
    All good points. One of the reasons I didn't really think that the cost was much of a problem was because of the availability of things like the Raspberry Pi, which is a complete development board with Ethernet for a retail price of $35. If that was reduced to a piece of hardware just big enough to do Ethernet without all the other cruft, it really couldn't cost much more than adding a UART.

    Interesting note about wireless as well. I've never really seriously considered wireless anything for industrial processes except for between remote locations perhaps. I guess I'm a bit paranoid about a medium that I don't have total control over.

  4. #4
    Join Date
    Jan 2003
    Location
    USA
    Posts
    5,197
    Post Likes
    Quote Originally Posted by rjk_cmh View Post
    [*]Not deterministic, if your machine needs extremely tight control or timing.
    This is not much of an issue for BMS. Nothing moves all that fast.

    Quote Originally Posted by rjk_cmh View Post
    Daisy-chain topologies are not possible.
    Many of the OEMs that offer Ethernet based unitary controllers have two ports with a built in switch. So they basically are a huge daisy chain.

    Quote Originally Posted by rjk_cmh View Post
    Physical layer takes up more space because of additional wires needed (4 twisted pairs).
    Terminating all the ends are a hassle and a lot more time consuming compared to two wires. Bringing up the network takes a lot more time troubleshooting.

    The cost of the cable is more than RS-485 also. Typically we need plenum rated cable everywhere.

    I think this trend is a ways off. I just can’t see much benefit of having a high speed network. Most of these devices have less than 10-15 I/O on them. Even if you want to poll the heck out of them you’re only talking a handful of bytes to pull in these values. What exactly are you going to do with space temperatures every second?

    I recently saw Distech is coming out with unitary controllers that can operate as WIFI APs. At first I thought this was a cool use of running Ethernet to all the devices. Thinking about it more, if you’re only running a single 100mb link daisy chained through 50-100 devices operating as Aps, what kind of bandwidth are you really going to have? These devices are typically <100' of each other, is the RF spectrum going to even support that many APs so close? Seems to me if this was used in a school or commercial building with lots of wireless clients, the network would be a dog.

    HW brought up the idea of running IP cameras off the BMS network as another possibility. Again, on the surface sounds great. Thinking about the movement to megapixel IP cameras and how cheap they are becoming, there is no way this is going to fly on a single 100mb link. My 3MP cameras will suck down 10mb+ constantly when viewed live. Somehow we are going to hang these off a single 100mb link with 50 switches on the run…. epic fail. Even if they are just using FTP to shoot back full res images on motion detection, this can generate hundreds of megabytes of traffic in a day easily.

    Another point you forgot is the security of the network. Simply putting another 100+ devices on the network all with web servers or other services running on them seems counterproductive from a security standpoint. Get these (http://www.youtube.com/watch?v=8Q4JKMZN9LQ) guys near your embedded devices and 485 starts to sound good.

    And your $35 rasppi is coming from foundation that is a charity. I can assure you these OEMs are for profit and you can expect that same hardware will sell for a good deal more.
    Propagating the formula. http://www.noagendashow.com/

  5. #5
    Join Date
    Feb 2005
    Location
    Edmonton, AB Canada
    Posts
    838
    Post Likes
    Quote Originally Posted by rjk_cmh View Post
    All good points. One of the reasons I didn't really think that the cost was much of a problem was because of the availability of things like the Raspberry Pi, which is a complete development board with Ethernet for a retail price of $35. If that was reduced to a piece of hardware just big enough to do Ethernet without all the other cruft, it really couldn't cost much more than adding a UART.

    Interesting note about wireless as well. I've never really seriously considered wireless anything for industrial processes except for between remote locations perhaps. I guess I'm a bit paranoid about a medium that I don't have total control over.
    Good day rjk_cmh,

    The RPI is tailored and manufactured for a completely different market and so the components used, although similar, are actually different. For example:

    1. RPI's sales volumes are about 1M a year, whereas, Industrial controllers are produced from anywhere from 5K to 100K per year. RPI benefits from a much larger economy of scale and so they receive much better pricing on all levels (components and assembly)

    2. RPI is designed for a completely different market and so the necessary and expensive governmental testing is either not applicable or reduced dramatically.

    3. RPI's lifetime is much shorter than Industrial controllers... RPI a few years... Industrial controllers 5+ years and so the design and components components used/selected are different... Components with greater long term availability tend to cost more.

    4. Reliability... RPI's components are mostly consumer grade parts, whereas Industrial parts are "usually" a much higher quality. Take for example the Ethernet connector (with magnetics)... The inexpensive consumer part can be purchased for around $0.30 in volume... the industrial high quality version is several $.

    There are a number of other differences too, but I think you get the picture.

    That being said the economies of scale of Ethernet are trickling down to the field controllers, but there is still a cost associated with it. Remember to implement Ethernet you need four main components:

    a. processor with Ethernet resources (MAC, Memory, performance),
    b. Ethernet Physical Layer interface (PHY),
    c. Magnetics (connector with a transformer or an integrated connector/transformer
    d. A suitable Ethernet TCP/IP Stack (firmware)

    I have left out the additional support components that are needed for above, but the point is that there is cost and is certainly more than serial. Fundamentally almost all processors have serial communication built-in (except, of course, the physical layer interface) and so using a serial interface will be less expensive.

    Anyway, I have become sidetracked... my apologies!

    Cheers,

    Sam

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Contracting Business
HPAC Engineering
EC&M
CONTRACTOR