Page 2 of 2 FirstFirst 12
Results 14 to 20 of 20
  1. #14
    Join Date
    Feb 2005
    Location
    Edmonton, AB Canada
    Posts
    599
    Quote Originally Posted by TheControlsFreak View Post
    A-silly-scope?! Oh yeah I hang one of those off my hammer holding ring.

    I have never seen any controls guy ever have one of these, though I have seen some IT guys have one of those fancy Flukes that has the huge screen and the ability to perform oscilloscope functions.

    It just seems a bit overkill to have on the shelf at the office such a high dollar tool that will only rarely be needed. Most of my checks on RS-485 are easily done with a Multimeter on DC volts. When working with new equipment/controllers:
    • I find a good working controller and start measuring
    • I take note of the voltage level(s)
    • I take note if there is a rhythmic fluctuation
    • If possible I get the voltage level on short runs as well as long runs to note how much drop there is


    After that, if I come across a comm problem I check my voltage on the RS-485 and if they don't look like the good readings. I start dividing and conquering (splitting the cable run in half over and over) and re-taking voltage measurements until I find the one or few bad controllers.
    Good day TheControlsFreak,

    I have never seen any controls guy ever have one of these, though I have seen some IT guys have one of those fancy Flukes that has the huge screen and the ability to perform oscilloscope functions.
    You are quite right... however, times are a changing. When data rates were low one could mostly ignore transmission line effects, but now not so much. Secondly, devices are switching (edge rates) much faster than in the past and so this also creates the need for transmission line considerations (i.e. switching harmonics). Will an Oscilloscope be the magic bullet and solve your problems? Nope, but if you know when and how to use one, it can greatly assist the diagnoses and understanding of what is going on with a problem communication bus.

    It just seems a bit overkill to have on the shelf at the office such a high dollar tool that will only rarely be needed.
    Indeed, scopes can be very expensive (over $50K), but also relatively inexpensive ... a few hundred dollars. Does one need the big high end scope to troubleshoot RS485 issues? Nope... even the least expensive ones would be suitable, as the bandwidth of the RS485 is relatively modest.

    As for its need... Would one use it every time? No, but it wouldn't hurt... If you are proficient with its use it is as simple to use as multimeter in RS485 applications. Why would I use it on fresh installs? For me it only takes a few moments and I can then see what the waveforms look like as well as ensure that minimal AC coupled noise is present. Where the device shines are in cases of weird/abnormal issues with a bus. For example I was called to a site where all was working fine up to one day... no devices were added or removed and no config/software changes where made. The problem was that one controller's (out of a total of 70) step down power transformer partially failed where one of the secondaries shorted to Earth ground. Because of how this device was designed this created an electrical path to the RS485 bus and the result was that the entire bus was acting like a huge 60 Hz antenna...and a good one at that. The result the communication signals were being heavily distorted and so retries, dropped packets, etc were causing havoc.

    Most of my checks on RS-485 are easily done with a Multimeter on DC volts. When working with new equipment/controllers:
    • I find a good working controller and start measuring
    • I take note of the voltage level(s)
    • I take note if there is a rhythmic fluctuation
    • If possible I get the voltage level on short runs as well as long runs to note how much drop there is
    Your testing a approach is a good one, however, the signals are actually time varying and your DC multimeter will not provide too much information on this important content. Measuring the static voltage levels (i.e. no RS485 communication) gives you only a bit of information. Measuring the DC level on an active RS485 bus will give you an average DC level which is not very helpful. Measuring an AC voltage is also not very helpful, as this is also an average. One must remember that these signals are not strictly DC, but are time varying. It is the time varying aspect that creates issues, as electrical characteristics that could be ignored at DC can be quite influential outside of DC. An example of this is that I can connect one of my test instruments to a RS485 cable that is unconnected at the other each end (wires are open circuited). My test device will tell me within 1-3 feet where the end of the wire... whether the wire being on a spool or installed. It can also tell me where I have kinked the wire. Remember all of this is done with the opposite end unconnected or terminated... How can this be the case when the DC current flow is zero? The reason is that I am looking at the time varying content of the signal that I am injecting (via the test instrument) and the transmission line characteristics of the signal create responses (reflections) that my test instrument analyzes. If I go one step further, and connect the test device to to a bus that has RS485 devices connected, but are un-powered (i.e. no communication) I can tell you within 1-3 feet where each of the devices sit on that RS485 bus.

    Anyway my point is that there is a lot of information in the non DC content of RS-485 and it is these issues that I see causing people the most grief.

    Cheers,

    Sam

  2. #15
    Join Date
    Dec 2008
    Location
    Seattle WA
    Posts
    107
    On multi-vendor networks it is imperative to understand how each vendor does stuff. Some have built in terminating resistors, others don't - but you must find out and disable all but the end of line ones. I prefer to disable all the built-in stuff, and use physical resistors.
    For the record, we have (3) Fluke Oscilloscopes at my office; I don't use them regularly, but I always do if I am having comm issues, or if I am starting up a multi-vendor trunk.

    Be aware that some VFD's have 'dynamic' end of line resistors. If you are using one of their eol resistors, and you power off the drive - the resistor effectively disappears. Now you only have a resistor on the other end of the trunk...

  3. #16
    Join Date
    Jul 2010
    Location
    Denver, CO
    Posts
    435
    Thanks for all the information guys. I definitely learned a lot.
    Last question. How do I size EOL resistors?

  4. #17
    Join Date
    Feb 2005
    Location
    Edmonton, AB Canada
    Posts
    599
    Quote Originally Posted by HuNGRYTeCH View Post
    Thanks for all the information guys. I definitely learned a lot.
    Last question. How do I size EOL resistors?
    Good day HuNGRYTeCH,

    Ideally and in Theory the EOL should be the same as the Characteristic Impedance (Zo) of your communication wiring. This value is usually anywhere from 100 to 200 or so ohms. 120 ohms is considered standard, as a lot of communications have a Zo of 120 ohms. If you do not know or cannot find the Zo spec of the wire, then either ask the equipment vendor or just try the 120 ohm standard and observe your communication signals, stats, etc...

    Cheers,

    Sam

  5. #18
    Join Date
    Jul 2010
    Location
    Denver, CO
    Posts
    435
    Thanks Sam, the wire I use is 120 ohms. I wasn't sure if different equipment required different resistors. 120 sounds good to me though.

  6. #19
    Join Date
    Aug 2009
    Location
    Jurupa Valley, CA
    Posts
    1,777
    The equalization is to cancel out the effects of the cable, and SHOULDN'T matter what hardware it's connected to (as long as it doesn't have an obscenely low impedance itself).

  7. #20
    Join Date
    Feb 2005
    Location
    Edmonton, AB Canada
    Posts
    599
    Quote Originally Posted by HuNGRYTeCH View Post
    Thanks Sam, the wire I use is 120 ohms. I wasn't sure if different equipment required different resistors. 120 sounds good to me though.
    Good day HuNGRYTeCH,

    In theory it should not matter, as CraziFuzzy mentioned... however... Some equipment vendors embed or have a self terminating resistor permanently installed... and so by adding another termination resistor in parallel to this type of equipment can also cause issues. Hardwiring a terminating resistor (i.e. not jumpered, etc), etc is frowned upon but does happen from time to time with vendors. Ideally contact the equipment vendor to see if they have embedded a termination resistor and if so find out its value. Then use this value and your target of 120 ohms to determine what your termination resistor should be (i.e parallel combination of your external termination resistor with equipment embedded termination resistor should equal to Zo or in your case 120 ohms) ... In most cases this should not be the case and so using the Characteristic Impedance (Zo) of the wire is where one should start.

    Cheers,

    Sam

Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Comfortech Show Promo Image

Related Forums

Plumbing Talks | Contractor Magazine
Forums | Electrical Construction & Maintenance (EC&M) Magazine
Comfortech365 Virtual Event