No announcement yet.

OT- long(ish) Term monitoring of the mains (230V)

  • Filter
  • Time
  • Show
Clear All
new posts

  • OT- long(ish) Term monitoring of the mains (230V)

    Hi all.

    Here's the deal.

    At work, we have reasonable suspicion that the power supply 230V S/P to one of our offices is dirty.

    We have had several items of IT kit fail in this one centre within the past few months. Much more than our other ten centres put together.

    Coincidence? Perhaps.


    We need a way of monitoring the mains supply.

    I was thinking of something along the lines of laptop, running storage scope and data logging software, using the sound card as an input and a few winds of insulated copper wire around the mains cable feeding the main comms cabinet.

    What i would like to know is:

    roughly how many turns of copper wire am I likely to need to get a reasonable reading on the scope.

    Any and all information, opinions and suggestions welcome.

    I am fairly sure this can be done. Just not too sure how.

    Thanks all.

    Just cause it works, don't mean you can't improve it

  • #2
    You need one of these for starters-
    I just need one more tool,just one!


    • #3
      Depends on WHAT you think is the issue.....

      Regular high-line conditions (presumably over 253/264 for 230/240)?

      Line transients? That may be caused by other industry, lightning (even if it does not strike the line) , etc.

      Either can be bad, but often different equipment is needed to check for the two.

      The equipment is available, and howling at the local power company MAY get them to put a monitor on the line for a while. But make sure it looks for ALL causes, and not just long term line voltage.

      Keep eye on ball.
      Hashim Khan


      • #4
        The gold standard for power monitoring is the Dranetz Power Analyzer. It's what the power company uses and if you want them to listen it's what you need to use. I used them for many years to verify poor power quality in offices. They are very expensive. A Dranetz will monitor all sags, surges, transients, droops, brownouts, frequency deviations etc. Some models provide a paper tape printout as a pemanent record as can be connected to a telephone line for remote monitoring via computer.

        Dranetz sells refurbished units from $2000 and up and new units go from about $7000 and up. Your best bet is to find a private company that provides the service using a Dranetz.

        I will emphasize that this is not a do-it-yourself job if you want results. If you don't use the right gear the power company will brush you off. They don't have time to deal with the vast quantity of uncalibrated measuring devices that people may use.
        Last edited by Evan; 06-23-2008, 09:57 AM.
        Free software for calculating bolt circles and similar: Click Here


        • #5
          In for a shock??

          Why not just put in a dedicated battery-driven inverter unit? At least you will have a stabilised supply as regards voltage, frequency and wave form as well as isolation from the suspect supply.

          Evan is dead right.

          The Power Company/Utility monitors its power supply quality pretty well and has a vested interest in keeping it within prescribed limits/parameters.

          They are very reluctant to commit resources to a "hunch" and if they do and it is their fault they will make it good. But if the "info" provided was wrong or the complaint was frivolous or vexatious they can take a very dim view of it and act (quickly) and charge accordingly (lots!!).

          If it turns out that the problem is on your side of the metered supply - it is your problem. If the "your" problem causes a problem with "their" side of things they may well charge/penalise you or give you "Notice to Repair/Desist" and/or disconnect you from supply until they are satisfied that there is no cause or risk of damage to their or their (other) customers assets.

          I don't know how it in the US or Canada of the UK, but that's how it is where I live in OZ.

          Best of luck - but hasten slowly!!!


          • #6
            Loose connections prevail thou.. The power company is not responsible for "meter on" connections.

            A arcing connection generates spike voltages akin to lightning strikes.

            I built a long term monitor from a old pc and opto22 equipment.. it could read and store waveforms for charting on the screen, give indications of over and under waves.

            I was working on a common supply, individual dc drive PID system back then.. It all got replaced by 3phase VFD's racked together thou. Mucho cheaper.
            Excuse me, I farted.


            • #7
              We use APC UPS's here and they will self-monitor for high and low voltage conditions (either of which is not a good thing). With their software, that information can be sent to a PC on an immediate basis and can also be logged. Of course, if you are using a device like a UPS which can both clamp high-voltage conditions, and which supplements in the event of a low-line voltage situation, your power is pretty well conditioned and your device is largely shielded from that sort of damage in the first place.

              Given the nature of lightening strikes and the current they induce even in ground as well as an increasing (at least here) incidence of power glitches from the utility, its pretty likely to see some damage without some sort of good power conditioning. I work for a university and we have a 300' microwave relay tower on campus. Its a giant lighting rod and as such, we have regularly had more than a reasonable number of issues.

              Deregulation of the power companies has been bad news. I am generally not for government meddling, but lets face it, the power company is for all intents and purposes a monopoly. If they cut costs to the point of providing lousy service, whatcha gonna do? Locally, our utility has been spending less and less on maintenance and providing more and more outages.

              Edit-- one other thing to look for in an environment where you are loosing lots of equipment, another thing to look for is sources of "tin whiskering" like galvanized (backside) raised floor panels. Google it. 2nd edit-- with the new CE standards requiring no-lead solder, this is going to become an increasing issue too as the tin content in the solder within a device is prone to this problem.

              Last edited by pcarpenter; 06-23-2008, 02:09 PM.
              Paul Carpenter
              Mapleton, IL


              • #8
                If you find noise or transients to be the problem...
                One of the best and most reliable solutions is a 'Sola' Transformer.
                Also called a Line Conditioner.

                It is just a special type Transformer running with a lightly Saturated Core. This clips Transients and removes most of the Noise and provides some Voltage regulation. Especially suited for the Industrial environment.

                They tend to be spendy but it's basic design makes them super reliable.
                Often available in surplus for good prices.

                Tom M.


                • #9
                  How about a used CT, what they use for remote metering, its just a donut looking thing with a pair of wires. Might be easier than wrapping copper wires, not sure how you would calibrate your scope to it gues it would be just a base line. But before you spend the big bucks for accurate monitoring equipment this might be an economical way to see if it actually is a power issue to start with.
                  Cheers, Bob


                  • #10
                    I dealt with suspected power problems for over two decades. In this area they are common as the out of town settlements are on very long single phase feeders, as long as 100 miles or more. Some, like the town of Bella Coola are on diesel generator power. However, it was much more common to find faults in the customers wiring. The most common fault is high neutral voltage. It's caused by a poor connection in the neutral somewhere in the circuit. Most commonly it will be in an octagon box someplace where the wire is spliced or in the main panel. It shows up as a high reading between neutral and ground and will vary depending on the load on the circuit.

                    This problem is the single most common cause of problems with computers and other digital devices. It will cause problems if the neutral to ground voltage exceeds about 5vac with a load on the circuit. It can be caused simply by a very long run to the panel with a heavy load at the other end. Wire resistance in the neutral alone will be sufficient to exceed the 5 volt level on a 14-2 wired 15 amp circuit with a 250 ft run and a 12.5 amp load, which is code maximum.

                    If this problem exists then it is time to check all the wiring connections and correct any problems. If the problem is caused by a long and highly loaded circuit then the affected equipment should be given a dedicated circuit with no other loads or, at the most, only similar sensitive loads.

                    In particular keep lighting loads off the data equipment circuits. Electronic ballast lighting can produce very high RFI levels. By placing them on separate circuits the wiring inductance to the main panel will tend to cancle most of the RFI on the lines.

                    Also, laser printers and photocopiers that rely on the xerographic process draw a lot of power in a cyclic manner, even when not being used, unless they are set to standby. These machines do not play well with data processing equipment and a laser printer should not be connected to the same circuit as the rest of the DP equipment.

                    High voltage can also be a customer wiring problem. On a split phase system if there is a poor connection in the main panel between neutral and ground it can cause a voltage difference between the phases making one lower than normal and raising the other phase.
                    Free software for calculating bolt circles and similar: Click Here


                    • #11
                      Check the grounds!

                      As Evan said, the most common troubles leading to poor power quality are bad in-building ground connections!
                      A friend is a PQ engineer at a large upstate New York, USA utility. He is the guy that comes out and troubleshoots PQ problems, etc. He's got all the tools, corona sensitive videocameras, infrared cameras, BMI-Dranetz power analyzers (well, he's using mine, now ).
                      He says that the trouble is almost invariably poor grounds, sometimes loose connections on neutral or line.
                      Occasionally the trouble is from out-of-building sources such as cap or LTC switching transients, auto-loop reconfigurations, etc. but an electronic power supply should be very tolerant of these sub-cycle events.


                      • #12
                        Evan, will send a PM. Maybe you could help with a problem here. Have a large multi complex facility spread over 88 acres but two of the 21 buildings seem to be accounting for 75% of the demand metering charge, but in my estimate account for less than 10% of use.
                        Cheers, Bob


                        • #13
                          Originally posted by Quetico Bob
                          but two of the 21 buildings seem to be accounting for 75% of the demand metering charge,
                          Probably Marijuana growing operations.

                          Tom M.


                          • #14
                            Not on my ship.


                            • #15
                              Quetico Bob,

                              What is in those buildings? If there is a high reactive component to the load such as inductive motors etc, you may need power factor correction often implemented as big capacitor banks.