Announcement

Collapse
No announcement yet.

Yet, Another RF Noise Question...

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Yet, Another RF Noise Question...

    Hello,

    Is it worth enclosing ALL the wiring from my solar panels to the main disconnect switch in metal conduit for RF noise protection? Does it change anything if I am planning to install a Enphase system (micro inverters)?

    For the first time a few days ago I heard about the RF noise concerns from a friend of mine who is a big HAM radio operator. He sounded like the sky was falling as he commented on my solar panel array installation prewiring (for a 14 kW system with three 12-3 Romex wires run 100 feet to my second story attic). In researching the issue here and elsewhere, I see his concern, but also see the majority of professionals and people allot smarter than I who don't feel this issue is that big of a deal. I don't have any sensitivity to RF noise, nor do I plan to take up HAM radios as a hobby. Still, I am at a point in some remodeling that it wouldn't be too hard to re-route all of my solar panel wiring into metal conduit. It would basically just waste half a day of my time and the additional cost of the conduit.

    So, what are your comments and/or opinions on this?


    Peace,
    Dr. Z.
    Last edited by Zeigh; 06-29-2016, 07:50 PM.

  • #2
    I guess we would need to know a little more about the solar hardware you plan on installing. I am not sure if #12 wire would be big enough to handle the current or voltage drop on 100 foot run.

    So before you worry about RF I would first make sure your wiring is up to code and will do the job for a grid tie solar pv system.

    Comment


    • #3
      Regardless of the answer to the conduit question, have your friend come over when they test the system, and have him bring a receiver; if he notices RFI, you can have the installer add chokes or other filtering that might help.

      I didn't notice much RFI except right next to my inverter (SolarEdge), but I only looked with an old portable Sony shortwave radio.

      I have a funny feeling an AC microinverter system like Enphase has more chances to generate RFI than a DC optimizer system like Solaredge, but I don't have any data to back that up.
      17kw. I like science, but I'm no expert.

      Comment


      • #4
        Another reason not to do microinverters. While i'm sure they have the RF noise below any UL spec, you still have many little switching power supplies generating noise. I'm not sure it would be more or less total noise than a equivalent string inverter would generate, but at least on the DC side - a string inverter is absolutely quiet. All in all, our modern world is so full of power supplies with switching circuitry above 15kHz and harmonics that go way on up - that we are bombarded with RF already. It would be interesting to find out what you HAM guy can measure...
        BSEE, R11, NABCEP, >1200kW installed

        Comment


        • #5
          FWIW, at about 5pm in bright sunlight, I walked around the area with my shortwave receiver tuned to 5.000 MHz (WWV). The neighborhood is urban, lots of radio noise.
          Away from the house, I could clearly hear the ticks and tones, and depending on where I was, even clearly understand the announcer giving the time.
          In the house -- especially around the front porch, or right next to long wires or downspouts -- there was too much noise to hear any of that.
          One LED floodlight made lots of noise, so I unscrewed it. (Bad Phillips, bad!) A dimmer made some noise, so I turned it off.
          Killing the power to the entire house reduced the noise level enough so that I could even hear the ticks and tones in a few places inside the house.
          Turning off the SolarEdge and throwing the DC disconnect didn't seem to change much.
          Proximity to the panels didn't seem to do too much.
          Proximity to the inverter did show some noise, even when it was off, probably because it's near lots of long wires and downspouts.
          Being near one phone line was REALLY LOUD.

          Our solaredge inverter came from the factory with big chokes on the inside near the bottom that aren't in pictures I've seen online, so they may have gotten more careful recently.

          In short, found some noisy bits, but no smoking gun pointing to the solaredge inverters or optimizers, at least with my "drowns out WWV on a handheld shortwave reciever" test..
          17kw. I like science, but I'm no expert.

          Comment


          • #6
            I think RF chokes near the inverters would be likely to attenuate the data going to the Envoy.
            Steve Dold
            Say NO to useless over-quoting

            Comment


            • #7
              You should ask a HAM radio operator instead of listening to Dan. I am KF5LJW, 73's.

              The issue is not Radiated RF nose, or even generated on the DC side (panels) You are barking up the wrong tree. it is Induced and mitigated on the wiring of the output of the GT Inverter to utility that is the offender. No amount of shielding would be effective. The problem lies in the design of the Inverter end of story. So unless you are willing to spend the bucks on a Output Filter, don't bother with it or give it any thought.

              If it offends any Ham Operator let them file a complaint with the FCC. Then in turn you pass the buck to the Installer and Manufacture for resolution at their expense.
              MSEE, PE

              Comment


              • #8
                Here's a picture showing the chokes: IMG_20160212_080151.jpg
                17kw. I like science, but I'm no expert.

                Comment


                • #9
                  Some basics. First thing to understand is where the noise is generated. If you used a Pulse Width controller, the switching of the entire DC
                  output current would occur through the panels, the controller, and the battery. The whole system is an antenna. Not a lot can be done about
                  this, except try to control the rise and fall times of the switching.
                  With any kind of MPPT system, there is very noisy action in the controller/inverter, but a good product will shield and filter the input and
                  output leads to limit how much escapes or is conducted to be radiated by system wiring. For a very sensitive situation (HAM nearby), its
                  not outrageous to build a (well ventilated) metal cage (hardware cloth) around the inverter and add another level of filtering to power leads
                  to the inverter. My Fronius inverters had good internal filters which did not require this, even right next to my radio tower. Those doughnut
                  filters might get a product just under a gov reg, but they do so little as to useless to a HAM.
                  Finally if you use micro inverters, you have moved the radiation generator to a place where you can't treat radiation, and there are lots
                  of them instead of one. My expectation is that in order to keep the size and cost of individual micros down, they would not be shielded
                  and filtered as well as one big inverter. I recommend against them in a sensitive situation; putting the final cable in conduit isn't going
                  to buy you many DB (if any), and not much else is possible. Bruce Roe K9MQG since 1958

                  Comment


                  • #10
                    Going back to the OP original post I see that he is using Enphase micro inverters so it would be the AC lines that he is worried about being affected by RF.

                    I am not sure how to install chokes on the micro inverters ouputs but maybe it can be done.

                    I still wouldn't worry about the RF since my concern is that #12 wire runs of 100 foot length which might cause even an AC voltage drop and might not have a high enough amp rating depending on the number of micro inverters per string..

                    Comment


                    • SunEagle
                      SunEagle commented
                      Editing a comment
                      Do those micros generate 240v or 120v? Also the OP indicated a 14kw system so that would mean close to 5kw per string with 3 strings. I am not sure what the micros are rated at in wattage but I felt the #12 at 100 feet might be pushing the current rating and possibly the VD if it was at 120v.

                    • Engineer
                      Engineer commented
                      Editing a comment
                      All of them are 240VAC. Enphase allows 14 280's per string, I have a 12kW system with 3 strings (14-14-10), the smaller micros like the M250 are 16/string I believe. I went through the math when I did the system, I believe each S280 maxes out at 1A, hold on let me check the spec sheet ... yes 1.13A for the 280, 14 max gives 15.82A which means a 20A breaker/string with the 80% rule (which maxes at 16A).

                      Yes checking the M250 spec sheet nominal output current is 1A with a maximum of 16 (single phase) per 20A branch circuit. Also don't forget that what matters isn't the nameplate size but the micro size. My nameplate is 12kW (38 panels * 315W) but the micros (38 micros * 280), so at maximum I have a 10.6kW system. Peak I've measured so far is just over 10kW.
                      Last edited by Engineer; 06-30-2016, 12:01 PM. Reason: spelling

                    • Engineer
                      Engineer commented
                      Editing a comment
                      Actually string inverters would be more likely to have losses than micros, given they're a low voltage DC versus the high voltage AC (remember the War of the Currents 100 years ago). I've wondered what gauge wire is used for string inverters? Presumably they did it right but if I had a string I'd have checked.

                  • #11
                    I work in RF test and have a 100k SA/NA in my house (I work from home part time) I could take some measurements with. Deciding outright that micros are a problem as solarix did above is stupid, RF signals and interference is extremely tricky - i.e. there's NO way to decide a-priori whether there is an issue or not and doing so is being ignorant, you have to measure a particular system. And recognize that another 'identical' system may not have the same interference characteristics (all it takes is for one 'rusty bolt' to cause some intermod distortion).

                    I definitely wouldn't enclose in conduit as, supposing you have some interference, that wouldn't necessarily fix the problem (and probably wouldn't anyhow). RF has a way of sneaking around simple brute force approaches like that.

                    Anyhow I can talk more about it later. I'm on vacation so don't feel like playing the RF engineer, but maybe I'll do some tests.

                    Comment


                    • #12
                      Here's a story for you. At work we have a large 1MW tracking solar array which is a beauty to see. We also have electric car stations, and having a lot of engineers we have a number of electric cars including a bunch of home brews. Anyhow these engineers noticed difficulties charging their cars, and being engineers went about finding why with using lots of really expensive test equipment we have there at work. Turns out RF interference wasn't an issue (we are a RF facility with a large state of the art anechoic chamber) but it was the noisy power coming off the inverters. You see because they couldn't figure out what department to charge for the power for the cars, they just put the car charging on the inverter side of the system (AC directly from the inverters), not the load side which for reasons unclear to me would have resulted in cleaner AC. Anyhow the dirty power coming off the inverters was causing trouble with the cars charging systems. Then we got a new CEO who is into electric cars (he has a BMW i8), so the problem got fixed by putting them on the other side of the circuit (and dealing with the costs somehow).

                      What kind of system do we have? Big old fatty string inverters.

                      Comment


                      • #13
                        Originally posted by Engineer View Post
                        Here's a story for you. At work we have a large 1MW tracking solar array which is a beauty to see. We also have electric car stations, and having a lot of engineers we have a number of electric cars including a bunch of home brews. Anyhow these engineers noticed difficulties charging their cars, and being engineers went about finding why with using lots of really expensive test equipment we have there at work. Turns out RF interference wasn't an issue (we are a RF facility with a large state of the art anechoic chamber) but it was the noisy power coming off the inverters. You see because they couldn't figure out what department to charge for the power for the cars, they just put the car charging on the inverter side of the system (AC directly from the inverters), not the load side which for reasons unclear to me would have resulted in cleaner AC. Anyhow the dirty power coming off the inverters was causing trouble with the cars charging systems. Then we got a new CEO who is into electric cars (he has a BMW i8), so the problem got fixed by putting them on the other side of the circuit (and dealing with the costs somehow).

                        What kind of system do we have? Big old fatty string inverters.
                        They probably should have installed HMT transformers to reduce the harmonic distortion and clean up the power quality. Those HMT's would also be more efficient and probably reduce electric costs.

                        Comment


                        • Engineer
                          Engineer commented
                          Editing a comment
                          Don't know, it's obviously a large commercial system and was a big engineering project. Big tracking array over a parking lot, the caissons are crazy, they go down to bedrock 30 feet down. Anyhow I assume they knew what they were doing but who knows.

                      • #14
                        Originally posted by SunEagle View Post
                        Going back to the OP original post I see that he is using Enphase micro inverters so it would be the AC lines that he is worried about being affected by RF.

                        I am not sure how to install chokes on the micro inverters ouputs but maybe it can be done.
                        I thought he said "I am planning"; I suggest that going that route means you are stuck with whatever radiation happens. Bruce Roe

                        Comment


                        • #15
                          Enclosing the Wires will do nothing. The noise is generated in Differential Mode, not Common Mode. Any Engineer would know that.

                          It requires an In-Line Filter that cost you money. One for each Micro Inverter.
                          Last edited by Sunking; 06-30-2016, 08:38 PM.
                          MSEE, PE

                          Comment


                          • Engineer
                            Engineer commented
                            Editing a comment
                            People are talking about two different things as far as I can tell, AC power line noise and radiated noise
                        Working...
                        X