@bit errorr rate etc: Digital TV/radio work above 50MHz (TV even abobe 400MHz), where these things do not emit any disturbances (used active components simple are not able to generate such noise), so the interferrence affect mostly long wave.
But their disturbance is still quite strong at short wave (bad for HAM's, as this is their "area of interest" and need the "air" to be very quiet)

Used to listen to AM radio when I was younger - Capital 604, which was only listenable after sunset, as I was outside the ground wave area, and they were not able to increase power for political reasons ( as they played that frowned on Rock and Roll, and had uncensored news reports).

Medved, that is what I thought as well. I heard before, and a previous post here confirmed, that those ballasts drive the lamps at 20 to 60KHz, so they should not transmit any noise more than a few harmonics above those low frequencies. However, the ballasts I tried create hash at any frequency I listened to. In fact, some of the worst noise was in the 2-meter ham band (144-148MHz).

I wish I had a spectrum analyzer so I could get a better picture of the noise. Right now I am just relying on anecdotal observations. Honestly, I don't have a good way to measure the noise.

My rep suggested installing a RFI filter at the ballast which I may try. Hopefully that will eliminate any noise being carried through the building wiring. I may also experiment with some homemade filters/chokes.

-Nelson

You might want to use some kind of grounded metal sheet to cover any wiring that is being used with the high frequency ballast.  That should shield the stray RF interference.

You can read more information at Electromagnetic shielding.

@ bit errorr rate etc: Digital TV/radio work above 50MHz

Actually my understanding of HD-Radio is that it operates within the normal AM/FM frequency range for each station.


---
I should also add that i leave the channel covers off the lights i use for plants (to allow more heat to excape)...so the ballasts & wires are visible

Quote from: Medved

@ bit errorr rate etc: Digital TV/radio work above 50MHz

Actually my understanding of HD-Radio is that it operates within the normal AM/FM frequency range for each station.

Maybe (technically) for FM, but for sure not AM - there is not enough "room", whole MW AM band is only 1MHz wide, not mentionning problems with reception of wide band signal on low central frequency...

AM digital radio, DRM, is carried in a subcarrier that is 12kHz offset from the centre frequency, and is around 6kHz wide. Thus, after demodulation ( a simple diode on the 455kHz IF) you can feed it into a PC soundcard ( using a point before any audio bandwidth limiting of course) as it is basically an audio signal from 6 to 18 kHz, and contains a heavily compressed and FEC corrected digital version of the audio signal.

Ballast noise is mostly harmonics of the switching frequency, and can extend up surprisingly high, as most of the switching is designed to be very rapid to reduce loss in the switch. You need common mode chokes on all leads, as well as ferrite beads on each wire, to absorb this high speed noise. Strangely this also includes the earthing lead of the ballast itself, as this will have a lot of this noise current in it. Ideally each lead needs a small feedthrough capacitor where it enters the case, with ferrite beads each side.

And the mentioned HD Radio is a US system that includes a mediumwave variant, pretty similar but incompatible to DRM. So digital radio below 30 MHz indeed exists.

There appear to be considerable differences in RF trash production between various ballast models. The CFL lamps I have are good in this regard, their noise turns out from the general noise floor on mediumwave only when bringing the radio within 20 centimetres from the lamp. In my experience TV sets and PC's plus their monitors are much more a problem.

Not to get too far off topic, but yes, I agree, PCs and monitors (LCD especially) can be horrible for interference. My LCD monitor is probably the second worst interference-causing thing I have. My Ethernet switch is the worst. However, I have never taken the time to combat this interference. I simply turn off my monitor and Ethernet switch when I am using the radio.

I haven't pursued the ballast interference problem any further yet. I have been swamped with work. I have a GE program start ballast and two lamps set up on the workbench right now. When I get a chance, I will start experimenting with noise filtering on the ballast lead wires. I am also going to be doing a lot of studying as I have little experience building noise chokes.

-Nelson

This is interesting discussion.  Thankfully, it does not apply to me, as I live in a rural area and do not have to worry with neighbors' interference and my almost totally magnetic, almost totally preheat array of fluorescents do not interfere with my AM or somewhat weak FM radio reception.  It will be interesting to see what solutions can be found!

Being an electronics collector and working on the stuff/always listening to music, this is a BIG problem for me too. Although, I can't really do anything about it because fluorescent lighting helps my winter blues, and I need the brightness. I usually try to just ignore it, although it is picked up even without the radio being on. It just plain "interferes"

Last December I had one of those small FM transmitters going.
One night when i was first getting it setup / tested I was getting a horrid high-pitched squeal in the transmission. Everything seemed fine with the wiring/etc I couldn't figure it out (even tried a different transmitter with no luck). ... then this thread crossed my mind, went in the room where the transmitter & unplugged all electronic lights.
The cause was a fairly new 4' T5 light with electronic ballast, somehow its HF was being picked up & transmitted this stuff is supposed to be the "latest & greatest" for lighting technology...it shouldn't cause issues like that LOL

@xmaslightguy: I would guess for poor a.f. cabling screening with very poor filtering on the transmitter input, picking up the ~38kHz from the ballast and beat with the stereo carrier in the stereo modulator. Try to disable the "stereo" on the receiver and i bet it disappear.

And only emission above 150kHz are regulated by standards/rules (ballasts operate at 20..60kHz, so mostly even 3'rd harmonic fit there...), emission below 150kHz is restricted only by consequent losses, so may be pretty strong.

But lot of mistakes are done with electronic ballasts installation:
- Usage of nonconductive or ungrounded fixture body
- HF wiring not placed correctly ("hidden" in the corner of the fixture body, so the metal body can not shield them)
- "live" HF wires are kept unnecessary long
- Swapping Line and Neutral on the input (more of a reliability/safety issue - the internal fuse is only on the "line" input)

The T8 HF ballasts I've tried all created hash well into VHF. It was sufficient to harmfully interfere with FM broadcast reception. As far as I can tell, everything is "right", fixtures are grounded, leads are short, etc. However, I have not tried any filtering methods. I just went back to magnetic!

On the other hand, I have been experimenting with T12 HF ballasts and found them to be RF quiet. First of all, I made sure to get the "b" ballasts, also known as residential ballasts. I forget the exact designation, but they are certified to meet the stricter standard under FCC part 18. I set up a T12 HF on the bench with no shielding. I just hooked it up to two tubes and plugged it in. Noise at HF and above was almost non-existent. I suspect that even the lower frequencies (AM broadcast) would go away if I installed the ballast in a proper, shielded, grounded fixture.

Sure, interference can often be filtered at the affected device, but I would prefer to use equipment that doesn't create interference in the first place. I've never had a problem with magnetic ballasts.

By the filtering i meant low pass filter on the audio input, what should block anything above 16kHz. This is in the standard for the stereo modulation and it has it's base in the Nyquist law (plus very little margin asking for very steep filter not affordable for consumer product; there the filter should start to attenuate at about 10kHz). Even without the ballast such transmitter would create very bad signal by mirroring all ultrasonic components from the incoming audio signal into audible range.
It is my experience, the these little FM transmitters are very poorly made on the signal front-end (having spurious noises even without any HF electronic around). Adding 4'th order low pass with the corner at 10kHz improved the quality very significantly - all noises disappear.
By the way the worst disturbance sources in these sub-150kHz frequency range are CRT based TV's (2'nd and 3'rd harmonics of the horizontal translate into about 8kHz) and mainly monitors (VGA has ~35kHz, what translate to ~3kHz => area of the highest human sensitivity range), by far worse then all HF ballasts i've seen (TV doesn't disturb itself, as it is synchronous with the picture signal, so corresponding effects are static on the screen and mostly "located" to the trace return blackening pulse)...