LED lights interferes with vhf receivers

[Simen]

I have a solar system with 310W of panels, 2x6V, 530Ah Rolls batteries and a Steca PR2020 regulator.

All of my lights are led-based with various makes and models; some are traditional 3 leds in a group in series with a resistor (6 to 16 groups), others are with 1 to 3 power-leds with builtin regulator, so they can run on 8-30V.

The leds with regulators builtin ( a switched regulator, i think) are the ones that blocks out both my fm-radio, and more important, my APT weather satellite receiver (137MHz wideband).

The interference are rather severe; the APT receiver are on the first floor, and the interfering lights are in the other end of the house at the ground floor. The receiver are running on on-grid power, while the lights are connected to the batteries.

Now; i have not grounded the solar-system in any way, but the receiver are not grounded in any way either... Would it help to ground both systems?

Living in Norway, we have, in what i can understand, a weird grounding system. With 230V, there is 115V between each phase and ground (2 'live' phases). According to my solar regulators manual, it has common positive, so if grounding should take place, one should ground either the positives, or ONLY the battery negative.

Or would a filter cap across the wires near the offending led lights do the trick?

[dnix71]

Can you give us the brand name of the LED string that is causing the interference or better yet, open the case and post a pic of the guts? I don't know why someone would use FM radio frequency to pulse LEDs unless maybe they just had some available cheap and it works.

[wooferhound]

When working with Sound Systems, a ground wire is also called a "Drain" wire because it's function is to drain the noise from the wire to ground. I would absolutely ground your Radio equipment to Earth Ground, and I would also ground the Negative of the RE system as it should be to help with lightning protection.

[dnix71]

Most radios in the US cannot be grounded. The outlet is polarized and the whole device is "double-insulated." I don't know how Norway does it. If the outlet in Norway is polarized, then flipping the plug over may be the answer (if that is possible).

Grounding the negative side of the batteries would eliminate noise if the LED controller is causing the battery "ground" to float high or oscillate for some reason.

Anything that caused that much radio interference would make me nervous. Over here, Uncle Charlie would be knocking at your door pronto if you were making noise like that.

http://retrotransistorradio.com/uncle-charlies-gonna-getcha.html

[independent]

LED lights with "ballasts" are high frequency switching devices and my 20-30v ones do exactly the same with radio. Shielding?

[Flux]

This is a problem with switching regulators. As we aim for higher efficiency they are used more and more.

The early ones were slow and only messed up long and medium wave radio. As the things are now switching very fast the interference get up into the higher frequency bands.

Perfect devices would be filtered to prevent such problems and they should be covered by many approval specifications but I suspect that if it needs a CE number they just put it on it, I doubt that every Chinese lamp has been tested for RF interference compatibility, some barely make basic safety.

I expect the makers are hoping that radio will go digital along with everything else so that it will never be noticed.

Grounding is important but I am not sure about your distribution system, I have enough trouble understanding the American split single phase that causes so much trouble with inverters. You may have to resort to filters and this is a black art and if you have to deal with many units it becomes a big issue.

Grounding the radio equipment and keeping antenna from power wiring may help a lot.

This problem was brought home to me a couple of years ago when  friend asked me to mend an old valve communications receiver. I got it working but it seemed almost useless until I thought of taking it out into a shed away from the computer. I never realised computers cause so much hash and they are carefully designed and regulated by many approval codes. What chance with a cheap string of lamps all over the house.

Flux

[Simen]

Thank you all for comments.

I do have some good pics of one of the cheapest and noisier one:

http://mwlmf.net/gallery/3W_Diodelamp (the pics have 3 sizes)

All of the led lights are 'made in China' i suppose, and most of them no-name; some of the better ones with low noise, are branded 'Soft Green' (of the regulated ones).

The weather satellite receiver gets its power from a double-insulated wallwart, but the whole receiver are encased in a metal box, and the electronics are 'grounded' to the case. I think i'll try to ground the whole case to earth...

I do not think that the led-lights are that noisy; it's the receiver that's just very sensitive; it receives signals from a 5-15W transmitter on the NOAA satellites about 820km out in space.

For those of you that are curious, here's the results of the receptions: http://mwlmf.net/sat/

'Grounding the solar system' would mean grounding either the positive, or only the battery, according to the manual for my solar regulator. i would be reluctant to ground the positive, since some of my current (and future) equipment connected to the system have ground on the negative. And grounding only the battery would seem futile in this case?

[DamonHD]

I have certainly noticed interference from a (5W) mains LED with an FM radio, but actually moving the LED light about another 2m over seems to have eliminated the problem for me.

And I haven't noticed interference from any of my 12V LED lamps which are much further away, though I haven't specifically being looking for it.

So, as you say, I suspect that having a very sensitive broadband receiver is the issue, and as Flux said, switch-mode power supplies and computer electronics issue a torrent of noise all over the radio spectrum well up into the GHz I suspect and right down to 10s of Hz...

Could you try partially screening your noisy LEDs with (say) metal foil or mesh behind them?  I doubt it would help, but it might be easy to try.

Rgds

Damon

[wooferhound]

Putting a Ferrite barrel over the power leads might help too

[12AX7]

Hello!

I think you have a heck of a problem to deal with.

What most people do not understand is that there are two types of ground.

The ground everyone thinks of day to day, the `earth' ground in your house that all the grounds in all the outlets are tied to.  Here in the states this ground it tied to the neutral at the main breaker/fuse box.

Other utilities servicing the house (cable, gas, telephone) all tie to this same ground.

The "OTHER" ground is an "RF" ground.   Yes, it is in fact the same ground.  However, with "special" qualifications.    An RF ground needs to be able to pass current at an RF rate.  Where the standard house grounds carry 50/60 Hz.   The RF ground is operating at several hundred hertz to gigahertz.  

Most house wiring fails miserable at passing RF frequencies (but do a reasonable job at transmitting them).

Over the years, I have been involved with locating/eliminating interference for short wave (back then who knew that short wave was really pretty long?)

HF (amateur radio) FM   VHF/UHF (both amateur and business band) Television and some microwave.   Some of these were at the transmission side, most on the receiving side.

I need to make the disclaimer that I do not have a "whole house" inverter or LED lights.   However, I believe that RF interference regardless of the source is still RFI and needs to be addressed the same.

RF noise is "radiated" thru the air and along power wires.  In addition, any attempt to fix the problems needs to address both.  

To add to your problem you probably have more than one source of noise.

Your inverter (I am guessing) is putting out one type of noise (thru the power wires and radiated from the wires) and the LED lamps (each one that is tied to the inverters `grid').   The LED's most likely are adding their noise to that same grid.  (Anyone remember BFOs or intermod?)

This mix of frequencies can mix with each other and create a completely different frequency of noises.   Add to all this are the "other" noise sources in most households such as your computers, wireless modems, radio receivers, microwaves, cd players and on and on.

As the others have already stated, ground ground ground.

However, be very careful how and where you ground things.

I have seen cases where grounding actually made things worse.

Be aware of ground loops,  

Do not be afraid of using too many ferrite  beads on the wires feeding into and out of your inverter as well as each of the LED lamps.  Make sure they are located as close to the inverter as possible (with in inches).  

Cap type filters can be effective but are frequency sensitive and can be very difficult to match to an "unknown" frequency.

I'm "Guessing"  that your inverter needs to be located as close to your grid/breaker box as possible, and the ground tie in should be made though a ground strap (a flat braided cable).

Once the noise has left the cables and becomes radiated the chances of fixing it at the receiver is just about zilch.    At this point, it is all about Signal to Noise ratio.   It is then too late to deal with the noise; all you can do is improve the Signal (a bigger better antenna and/or improved transmission cable between the antenna and receiver).   YES, you should have your radio and antenna grounded, but all that will do is improve your radios reception of both the signal AND the noise (not touching the debate RE: lighting protection).

Good Luck!

ax7

Mark

[Simen]

My house are grid-tied; it's only the lighting in the house that are solar-powered. i do have an inverter, but it's not connected to the grid at all. It only powers my workshop in the basement (which have no grid-power).

Being a older house, there's few grounded outlets, and none close to my receiver, so i'm planning to connect a separate ground to my receiver. The grid-ground are not connected to any of the phases, and get its ground through the copper-waterpipe that are buried in the ground.

I had the same noise problem where i lived earlier, but there i had a 12VDC cable (which also the led-lights was connected to) running a foot from the receiver, but not connected to the receiver. Thinking it was noise from the 12VDC cable, i connected the negative from the 12VDC to the receiver chassis, and that solved my problem there.

So it would seem that 'grounding' the receiver to the same negative as the lights use, would help... My ground floor have solar power; my first floor are powered by a separate battery which i charge once a week or so. I have none of the noisy leds on my first floor.

[DamonHD]

Careful of setting up 'ground loops' if you have multiple earthing points.  It may add to noise or even generate currents high enough to damage equipment.

Rgds

Damon

[Simen]

Being familiar to older guitar tube amplifiers, i do know about the joys of ground loops...

[12AX7]

Hello!

any luck?

I had another thought I'd like to mention.   I don't know what your water quality is like, here many have whole house water softeners.  If you have one I'd suggest that you check and make sure that there is a "ground jumper"  between the inlet and outlet.

Most have plastic bypass valves that can isolate the outlet and all other pipes from ground.  

ax7

Mark

[Simen]

Ah, i haven't tested or done anything more yet. Main reason being that i'm refurbishing part of the house, and i'm going to re-wire much of the 12VCD...

Regarding water purifiers/softeners; the springwater here in Norway are about the same as most countries bottlewater, so such purifiers are rare. They only appear among the fanatics...

That said, in newer houses, the standard here today are Pex-hoses, not copper, so point taken, but my house being from 1913, all waterpipes are copper. (In imperial dimensions even... )

I guess this will be a more frequent problem as led-lights get cheaper and more popular, so i'll keep you informed.