If you are a radio listener-electronic ballasts may cause SEVERE wideband rf interference.I quit using mine because of this.

If you are a radio listener-electronic ballasts may cause SEVERE wideband rf interference.I quit using mine because of this.

Which frequencies? I don't know what wideband means.

Medved, this ballast's only recommended application is remote, and the manufacturer also lists ballast-to-lamp distance of 20'. Electronic ballasts start the lamp a little differently than magnetic pulse start, im not sure of the specifics on how it works, but ive read its equivalent to a 6.5Kv ignition pulse, but without that high of a voltage, I think the HF may play a role.

Also Ive seen large installations of these Lumatek ballasts where the cable distance between lamp was greater than 20' and pulse start lamps still fired up reliably. I understand the HF operation and long distance of cable between ballast and lamp isn't a good idea, but these are clearly designed for remote operation.

If it is really specifically rated for up to 20', then it should be indeed OK.
My remark was assuming the ballasts, which state nothing about maximum wire length, those really do not expect more than few feet.
But definitely I wold not exceed the rated limit, whatever that is.

Regarding the starting: The HF helps, indeed (the pulse is a HF anyway), but still I would expect few kV there. Don't forget if the HF helps the wanted discharge within the lamp, it may help as well the unwanted discharge in some insulation fault. So the wiring should really be designed with the full pulse voltage rating.
With HF you may sustain the high voltage at high capacitance (= longer cable) by means of a resonance (it becomes part of the voltage boosting series resonance circuit).
The classic pulsers for magnetic ballasts operate aperiodically, so there the capacitance only kills the high voltage.

But the high frequency with long wiring is really asking for RFI troubles, there is no rocket science in it...

Quote from: BlueHalide on June 04, 2016, 01:07:58 PM

Medved, this ballast's only recommended application is remote, and the manufacturer also lists ballast-to-lamp distance of 20'. Electronic ballasts start the lamp a little differently than magnetic pulse start, im not sure of the specifics on how it works, but ive read its equivalent to a 6.5Kv ignition pulse, but without that high of a voltage, I think the HF may play a role.

Also Ive seen large installations of these Lumatek ballasts where the cable distance between lamp was greater than 20' and pulse start lamps still fired up reliably. I understand the HF operation and long distance of cable between ballast and lamp isn't a good idea, but these are clearly designed for remote operation.

If it is really specifically rated for up to 20', then it should be indeed OK.
My remark was assuming the ballasts, which state nothing about maximum wire length, those really do not expect more than few feet.
But definitely I wold not exceed the rated limit, whatever that is.

Regarding the starting: The HF helps, indeed (the pulse is a HF anyway), but still I would expect few kV there. Don't forget if the HF helps the wanted discharge within the lamp, it may help as well the unwanted discharge in some insulation fault. So the wiring should really be designed with the full pulse voltage rating.
With HF you may sustain the high voltage at high capacitance (= longer cable) by means of a resonance (it becomes part of the voltage boosting series resonance circuit).
The classic pulsers for magnetic ballasts operate aperiodically, so there the capacitance only kills the high voltage.

But the high frequency with long wiring is really asking for RFI troubles, there is no rocket science in it...

...And then the FCC won't be happy.
But I think they have kind of forgotten about that and instead is dealing with the internet instead.

When I mentioned "wideband" interference-its just that-the interference can cover standard AM,HF and even into VHF FM and TV.With the electronic ballast you have made a "spark transmitter" of sorts.Radio hams have been complaining about the interference on the HF bands from electronic flourescent and HID ballasts.At the worksite I am at they subscribe to several "Ham" magazines all of them discuss the interference problems.The ballasts generate a "hash" on these bands interfering with reception.I would think the makers of these ballasts could incorporate RFI suppresion devices in the ballasts.Don't think it would be that difficult.And also provide filters in the ballasts so the RF generated within them cannot "ride out" on the powerline feeding the ballast primary circuit.The switching circuits in those ballasts generate RFI as a byproduct of their operation.Some ballast makers now recognize this and have the ballast run at lower frequencies to prevent the RFI.Look in the websites that provide electronic ballasts for plant lights and you will see the LF ballasts for HPS and MH plant grow lamps.

When I mentioned "wideband" interference-its just that-the interference can cover standard AM,HF and even into VHF FM and TV.With the electronic ballast you have made a "spark transmitter" of sorts.Radio hams have been complaining about the interference on the HF bands from electronic flourescent and HID ballasts.At the worksite I am at they subscribe to several "Ham" magazines all of them discuss the interference problems.The ballasts generate a "hash" on these bands interfering with reception.I would think the makers of these ballasts could incorporate RFI suppresion devices in the ballasts.Don't think it would be that difficult.And also provide filters in the ballasts so the RF generated within them cannot "ride out" on the powerline feeding the ballast primary circuit.The switching circuits in those ballasts generate RFI as a byproduct of their operation.Some ballast makers now recognize this and have the ballast run at lower frequencies to prevent the RFI.Look in the websites that provide electronic ballasts for plant lights and you will see the LF ballasts for HPS and MH plant grow lamps.

I forgot Little Ham radio was a thing. (Little Ham is also a story for later)

The lower frequency is not motivated that much from the RFI reason, but from the pure efficiency reason. The thing is, the magnetic components can not be made much smaller (with less turns or so) on higher frequencies, because the magnetic losses will explode. So if they have to use certain amount of turns, lower frequency means lower losses mainly in the predriver part and the skin effect in the winding (mainly important for higher current ballasts).
The EMI is a problem on it's own. When the ballast is designed decently, it does not radiate at all above few 100's kHz even without any RFI suppression filter (that isn't effective below 1MHz anyway and above that the suppression is not that great). Quite important part of the design is frequency dithering. What is does, is mainly making sure the energy isn't concentrated within narrow spikes in the spectrum (which then slowly wander up and down with the changing temperature or so, so in any way they once a time "meet" the band of interest). Instead the energy is spread.
At a first glance, it means instead of disturbing just few frequencies a lot and keeping the rest then quiet it disturbs everything a bit, but the point is, the frequency is not that stable, nor well define to not make one of the spikes to just fall into your reception and cause a loud whistle there. With the dithering you get only slightly elevated noise floor level, so for most application way less disturbing than the same energy (that falls into the receiver bandwidth) within one tone (this is the case not only for voice transmission, but for most digital modulations as well - it ensures at least some gaps, where the data may be received, so the resulting BER is lower). These dependencies are then quite well covered by the CISPR25 or similar (quasi-peak detector specification,...).

The thing is, the "normal" ballasts are designed to work in "normal" environment, with "normal" immunity against parasitic radiations.
But the HAM shack belongs to rather a "special low noise requirement" environment, so you need to design all the equipment accordingly (many commercial equipment will exceed those requirements). I think comparable spec's may be the marine ballasts (which are supposed to not disturb the communication and navigation equipment working in LW and SW bands; again these requirements are way stricter to what normal general service devices offer)