Page 1 of 1

Radio Noise from PV System

Posted: Sat Sep 12, 2009 12:19 pm
by w0sd
My call is W0SD amateur radio and I have loud birdies every 24 khz on the amateur bands coming from the Flex 80. The are coming in over the antennas. As one would expect the direction of the beam and proximity of the antenna has an
effect on how loud they are but with an antenna about 100 feet away they are still very loud. Twisting the leads going
to the battery and ferrite has no helped at all. If I shut off the breaker for the panels the noise goes away on all the
frequencies except 80 meters. They are much weaker but still there. If I shut of the Flex 80 they of course go away.
They go away when the Flex 80 goes to sleep? It seems to be a classic case of the wiring radiating the "birdes" as sniffing with my two meter HT indicates the signal is not only radiated from the wiring going to the 12 volt battery bank
but also from the PV panels. I have four 205 watt panels and have around 60 volts input to the Flex 80.

Has anyone come up with a cure?


Re: Radio Noise from PV System

Posted: Sun Sep 13, 2009 2:49 am
by blackswan555
Not my field at all (Although I did have a sideband CB some 30 + yrs ago if that counts ? :grin: ) But you may find some help from this ? ... terference

Have a good one

Re: Radio Noise from PV System

Posted: Sun Sep 13, 2009 9:21 am
by tallgirl

You need to build a low pass filter into the PV conductors, preferably between the FM80 and the PV disconnect. Another possibility is to build a 24KHz filter and connect it directly to ground from the PV+ conductor. Be sure to fuse appropriately -- there shouldn't be much current in the 24KHz AC component, but you'd have to determine that experimentally (read: blow up a few parts first), then fuse.

Assuming you have your array conductors in metal conduit from the PV disconnect to the combiner box, I'd put an L-C circuit inside the combiner box as well. I trust you know where the "L" and the "C" are going to go.

I'm surprised you're hearing noise that far up the spectrum, but an FM80 going full tilt has a lot of power going through it and the harmonics probably just keep on coming. I was going to suggest the 24KHz signal might be getting into your receiver's IF stage, but if you are hearing them on a 2 meter handheld, and your 80 and 40 meter gear, I'm guessing that's not the case.

Re: Radio Noise from PV System

Posted: Sun Sep 13, 2009 3:58 pm
by Vic
Hi Ed,

OK, YES, the MX-60 -- the only OB CC that I have experience with does emit a lotta RFI. Believe that the FM is very much the same in this regard. Almost certainly, the noise is line-conducted. I would guess that mostly from the PV array, and on the PV conductors to the CC. Keep your antennas AWAY from the power room ( a bit late now, I bet). Use Ferrite common-mode chokes on the PV lines into and out of the CC. Run ALL conductors to/from the CC in pipe, grounded at the DC conduit box. A low-pass filter might help, it should be in a can with short leads for the capacitors. These capacitors should be stacked metal film, or other very low L units. Consider using several capacitors in parallel in each position of the filter.

Isolation of antennas from the power generation stuff, shielding/grounding of conductors, and filters are the best bet, BUT, really, the PV array will wind up being unshielded, and many arrays are fairly large, so, trying to isolate the antennas from the power facility is really important, if possible. Use Type 31 Ferrites for the common-mode filters, and work hardest trying to keep your low-band antennas away from the power stuff, as the largest emissions are at the lowest freqs, and any filters are., generally the least effective at the low freqs, and so on. And, the low-band antennas have the largest capture area.

When you do the work on the PV side of things (or at night), you will, almost certainly, hear switching birdies from the inverters, as well. So keeping the output of the inverters in pipe (or underground) should help attenuate these noises, as well. Forget if you noted weather you are grid-tied or not. Being off-grid probably worsens the situation. At my location have run the AC power lines underground, in pipe, and even used shielded cable for the AC runs in the dewlling (Coleman CorraClad). In the shack, AC is mostly in pipe, but some CorraClad there, too.

FYI, the Honda generators are great, but they, too, generate lotta hash on the lower bands, which can be dealt with as well.

All just MHO. Nice to see U here, Ed, and TU for SD in SS and other tests (think that you are the stalwart SD stn in SS and NS and so on). 73 GL Vic K6IC

Re: Radio Noise from PV System

Posted: Mon Sep 14, 2009 10:16 am
by w0sd
Thanks for the replies! They are very much appreciated. I have a lot of work and money in my amateur equipment and of course some in the PV system. I am not an electrical engineer or technician so I have some areas that are weak and some that are strong when it comes to electronics. First in SD we really don't have provisions for On Grid, there is no requirement at all for a power company to buy energy! So I am off grid and am providing 12 vdc power for a lot of my radio equipment and also we are using this as a test site before we buy things to power some FM repeaters at remote locations with solar. This radio noise is a "MAJOR ISSUE" for both! I am using no inverter so inverter noise is not a factor here. I may eventually have a small TSW inverter to power my computer, DSL modem and router so I can keep things operating when we lose power or have power glitches, ie short outages.

I have to get rid of most of the radio noise from the PV System as it is just unacceptable for my amateur radio operations and unacceptable for the remote repeater sites. Getting my antenna's farther away is not practical as I have 10 towers and they range from 70 feet to nearly 200 feet high and in most cases my antennas are 150 feet or more away already.

Information such as you have given me is extremely valuable. I sure would like to have a Class B emissions rated XW SCC or know what the Morningstar TS-MPPT 60 will be like to compare interference levels. Hopefully I don't have to buy one but can find someone to make the test for me. We choose the FM80 as we wanted to use four 24 v panels at 205 watts so we could run a higher voltage to the controller and also with the 12 volt battery bank we need the wattage rating of 1250 watts to provide sufficient energy for our remote FM repeater sites so the 60 AMP Xantrex was not adequate for our needs unless we bought two of them, etc.

Back to the FM80, IMHO- not an electronics expert, I need to contain, stop the interference coming out of the FM80 box. I think it is an aluminum box or at least non-ferrous that the FM80 is in. I note that the level of the interference changes, it switches fairly regular. By this I mean every few seconds. Right now in MPPT Bulk on 1.832.2 mhz I see the interference hetrodyne change from S-8, S-9+5 db and S-9+12 db. The interference is every 24 khz so I assume it is a harmonic of the 24 khz switching. This would be supported as it definitely strong at 1.832.3 and gets weaker as I go up in frequency but I can still hear it at two meters.

Interesting to me is I can take the lower cover off the FM80 and it really does not change the level so that would seem
to indicate it is going out on the wiring rather than leaking out of the box. So, as has been mentioned it seems logical to me I have to attenuate the 24 khz signal "AND" its harmonics going out on the wires. Putting everything in conduit will be a lot of work and may be a waste of time as the radiation may still be significant from the actual solar panels so again stopping it at the box seems to be what I have to do. I noted in previous tests that shutting off the PV panels got rid of the interference except on 80 meters (3.6 mhz). Today I am leaning a lot more by being on 1.832.3 mhz. Here I find shutting off the panels out at the array makes no difference (have to be quick before the controller goes to sleep). Also shutting off the PV panels right by the controller reduces the interference very little if any at 1.832.3 mhz. "SO" I went back and checked my previous testing above 3.6 mhz and it is still valid. At 7.0362 mhz I found out something very interesting. When I shut the panels off by the controller the interference goes away but also when I shut the panels off with the breakers "AT" the 4 panel array the noise goes away. It is about S-8 with everything running. I could see no difference between two panels or 4 panels being hooked up.

The bottom line here is I think this pretty well proves that putting everything in conduit is not going to guarantee a cure as the panels also radiate as indicated by the 7.0362 mhz test. The differences above 80 meters and at 80 meters and below are just a matter of signal strength of the harmonics. I likely could see improvement by shielding the wiring to the batteries and shielding the wiring to the PV panels but at the lower frequencies such as 80 meters, 160 meters and the broadcast band the PV panels are going to radiate and I can not shield them. What I want to try and do is stop it at the FM80 box.

This is a band new system and there is no corrosion and everything has good, new, tight connections so the harmonics just about have to be generated in the 24 khz switching circuitry. So the way I see it I have all these
harmonics coming out on the wires and broadcasting at 24 khz and every harmonic of 24 khz. These I am quite sure are not being generated by rectification outside the FM80 box. It seems to me that a 24 khz filter will indeed stop that but not all the harmonics and it is the harmonics that are the problem unless the 24 khz harmonics could be stopped but that would void my warranty and I don't have the technical expertise to design that. It seems to me I need to get rid of these harmonics that are on radiated on the DC carrying wire.

ENGINEERING QUESTIONS: Can I be successful with capacitors from the DC leads to ground right in the FM-80 box with different values to ground for low and high frequencies? It would seem that I would just have the voltage rating to be concerned about. Otherwise it would seem like I need a broadband filter/attenuation from the broadcast band through two meters (148 mhz). I am looking for design ideas?

FYI I come out of the FM80 to a breaker box, about 1 foot away. One breaker is for the panels and the other breaker is for the batteries. I also have breakers out at the panels which are about 50 feet away.

Unfortunately I need some engineering help!

Re: Radio Noise from PV System

Posted: Mon Sep 14, 2009 11:08 am
by Vic
Hi Ed,

Thanks for the reply and additional info.

I DO have a TriStar 60 CC, but it is a back-up unit, and have never used it. In my current steup, my Vin from the PV is 106 VDC, it will most probably tax the TS 60, as it is generally, a Pulse-Width-Modulated controller (or, alternately, a variable resistor type) controller. Am very busy now, so I will not be able to test any other CC at this moment. BUT, I am certain that the TS-60 in PWM mode will be quiter (probably much quiter) than ANY switch-mode controller. All of the major large MPPT CCs use FET transitors as switches to chop the PV in into a Pulse-Width-Modulated waveform. This approach uses a large inductor (or perhaps in some cases, a transformer) with filter capacitors as an integrator of this chopped signal. This approach needs to have very fast switching times to cut the losses to a very small value. These fast edges are very rich in harmonics, which make the emissions fairly large. The inductor/capacitor on the output reduces some of the harmonics, but there are limits. All of this is just MHO.

In addition to thi main switching of the MTTP function, there is at least one additional switcher, which provides the operating voltages for the the CC, and runs from the battery. THis switcher may also create some emissions when only the battery breaker is on, and the CC is idling (sleepein/snoozing etc).

During Bulk, the emitted signals should have fairly constant values, but during sweeps, in Asorb or in Float, you will notice the signals varying quite a bit, as the amount of current taken form the PV varies with the MPPT function.

To me, all of the noted measures will reduce the emissions, some more than others. Each will do something to help, and will sum into a fairly large benefit. Common-mode chokes using Type 31 Ferrites in a metal box, very close to the CC should help. A Low-Pass filter may be the largest single helpful item, in your situation, altho, I have not tried it. Think that if it is a Pi type filter -- capacitor on the input, an inductor plus a capacitor on the output, in a shielded box could help, and here, using low inductance caps, and perhaps a stacked-metal-film cap paralleled with lower value disc caps may be the ticket. But keep the capacitor leads SHORT, and place this LPF very closse to the CC.

Regarding the FM repeaters @ VHF/UHF, any of the filters should give a large benefit at these frequencies.

I am almost certain that the primary emissions are Line-Conducted, not directly radiated from the box. The FM60/80 should have (altho I've never seen one in the flesh) an alunimum (or perhaps zinc) casting, with a smaller steel wiring compartment, but this by itself should be perfectly adequate in shielding radiated emissions. There have been reports from others, that at remote repeater sites, CC noise has been an issue. Again, short conductors to/from the CC, sheilding of these conductors in EMT or Rigid pipe, and separation of antennas from the Power room will help. One additional item we use here: the power room is in a steel sea Cargo Container. This helps contain emissions which might be radiated by interconnections. If your remote repeater sites could use metal power rooms of some kind, and filtering on the conductors into/out of this metal room, you may benefit. Often repeaters use omni antennas, which can create problems vs directional antennas, and so on. At least for your repeater sites, you may have the benefit of starting from scratch, OR, do you ? OK, for now, gotta run. 73 GL VIc