Oh My! FIRE!

Thanks, I'll give them a call soon. All 3 ballasts purchased from separate distributors do the same thing, so perhaps they made a bad batch. Appreciate the efforts!

 

Or perhaps, they don't know that it causes this effect because they have never experienced it or tested it...

It seems that the only thing these lighting companies are doing is building fixtures.. Some of them may not be putting a great amount of effort into researching the ballasts they use. Ballasts can be designed for different purposes, and ballasts that are designed for remote installation where they can be isolated far enough from any electronics should probably not be used in all but specifically designed aquarium system. It seems that ballasts for aquarium use have a specific set of design criteria that should be taken into account. These ballasts would probably cost the end user a bit more... and thus the cost of really good fixture designs that eliminate common issues is higher. In other words, you get what you pay for.

 

To take this a step further, I'd estimate that a single digit percentage of ballasts are used in our hobby. Unfortunately we have the unenviable position of being a minority of folks who rely on ballasts to not trip a GFI. Most ballasts which service halide and other HID lamps are in settings where a GFI doesn't figure into the equasion. Capitalism as it is, is not going to favor a surge in mass producing ballasts to fit our needs.

Perhaps down the line an electronics company will do some research with an oscilloscope--determine the offending stray interference for each particular ballast, and market a filter to mitigate the interference for our purposes. It's not hard to do. Filters are a relatively simple branch of electrical need.

 

This is exactly what I do for my company. Although, it's not done with an O-Scope, it's done with a Spectrum Analyzer. We have one rented right now (because purchasing them costs ~$10,000 for a cheap one, decent ones that certified testing labs use are around $100,000 or so).

When I have finalized a design, it costs us about $10,000 to have all of the tests confirmed by a certified lab. Once we confirm our results with the certified testing lab, we can issue an FCC label to our product, deeming compatibility.

I have antennas set up in my shop to capture radiated emissions and a balun to measure the conducted emissions. I design the EMI filters that are used with our amps that have potential to produce high frequency emissions. Normally the offending frequencies of HF switching power supplies or digital clocks are broadband and not attributed to a single frequency. This is because harmonics of the original offending frequency normally show up and when you get rid of one frequency, another will usually pop up (think about it like squeezing a tube of toothpaste). It takes quite a bit of work to get results when starting from scratch, not limited to jut EMI filtering, but also by shielding and encapsulating the RFI.

What is known as EMC (or Electromagnetic Compatibility) comprises the range of testing that can or will be done to products before they enter the marketplace. There are MANY different standards to which products are tested. All of these standards have a lot to do with the end use of the product. Whether it is used as a consumer product, commercial, or industrial use, can all determine the standard... sometimes the use can cause the standard to be very difficult to pass, and sometimes it can be very lax, it all depends... Hospitals, for example, generally have a very specific set of criteria for EMC that do not normally apply anywhere else.

Just as ReefSparky pointed out, aquarium use is a very specific use for a product. It is generally up to the manufacturer to determine EMC, however, in the case of an electronic ballast, where the manufacturer of the ballast itself does not dictate it's end use, it would be up to the maker of the lighting fixture to determine EMC for their use. My guess is that some companies are aware of this, and some are not. EMC compatibility should be taken into account the way power handling and requirement specs are, but that is probably not always the case when many companies do not even know what EMC is (as has been my experience when speaking with the lighting companies' techs).

 

I forgot to add... the reason I pointed out the costs associated with testing, correcting and certifying with EMC is because it's why a lot of companies don't do it. When they find out that the costs of NOT complying with EMC directives outweighs the upfront expense, is usually when they begin worrying about it.

 

Chiming in at an odd point in this thread....

I just installed a smoke detector above my tank.... to go with gfci outlet

 

I have a glass break sensor above mine!

Inadvertently, it just happens to be near the back door, which is all glass...

 

Very smart! I should do that. Because when my power strip was on fire, smoke was pouring out of the bottom of the stand, but my smoke alarm (which is located in the kitchen) didn't go off. Even though the tank is pretty close to the kitchen. So that's a good idea!

 

They make a gfci powerstrip. hd sells them I know as I have one of them they are ezpensive but 56 is cheap compared to a house.

 

Thank you. I'll look into those!