[RFI] LED Floodlight warning!!

Mon Jan 24 00:14:06 PST 2011

Hi Dale

The filter components were all placed inside the aluminum  housing. The light is very well shielded,  thats why I was so surprised at the very high levels of QRM. 

I placed  the IEC EMC mains filter, #31 common mode chokes(input and output) inside the shielded housing.

The ground conductor was initially grounded to the housing. I lifted the ground so that I could solder it directly to the EMC filter ground at the 240 volt AC input. I will try a choked earth conductor as well. I might also try grounding the EMC filter ground to the housing. Since the 240AC mains cable is not bypassed to the housing  at the entrance gland, it may well be contributing to the incidental radiation.

When I examined the switch mode power supply module it had a diagram indicating  that a diode is placed directly across the output of the power supply. I will investigate if the RF hash is being created  by the diode junction being switched at high speed by the switcher.  I will also try capacitor by-passing of all the leads.   I generally dont like doing this because of the high  leakage currents which can cause problems with the earth leakage switch.

Australia like the rest of the world's regulation  bodies is turning a blind eye to the  junk from China. I dont know why these laws are being allowed to be flouted in the name of free trade and appeasing the rubbish import merchants. 

The problem is that even if you are a discerning consumer the markings and certifications on  the boxes mean absolutely nothing,  we all know  that the Chinese factories are acting in a fraudulent manner when it comes to their EMC compliance.

As for this 50 watt LED floodlight there is no doubt that performance is excellent. They are very bright for such a low power consumption light. 
Its a darn shame that there is no quality manufacturers like Philips for these high power LED lights.

73
Craig
VK3HE

--- On Sun, 1/23/11, Dale Svetanoff <svetanoff at earthlink.net> wrote:

> From: Dale Svetanoff <svetanoff at earthlink.net>
> Subject: RE: [RFI] LED Floodlight warning!!
> To: "Craig" <vk3he at yahoo.com>, rfi at contesting.com
> Date: Sunday, January 23, 2011, 6:38 PM
> Craig,
> 
> Thank you for documenting your work on these errant
> lamps.  I am not
> surprised by the severity of the problem, however, since
> the switcher
> powering those LEDs is most likely operating in a constant
> current mode
> (see my posting of 1/21 on the topic).  I doubt if the
> use of a modular
> array is the issue - more likely is the amount of current
> beign delivered,
> since LEDs designed for lighting applications are capable
> of much higher
> current per diode than those intended as mere indicators
> (those little
> colored light dots we've known for many years now). 
> As noted in my post,
> some of these power supply designs have no filtering caps
> on the output end
> of the supply.  One question:  where did you
> physically locate the filter
> components with which you have been experimenting?  If
> they are not within
> the shield of the lamp assembly itself, or of the fixture,
> then the leads
> may be long enough to make for good radiators and/or the
> filters themselves
> radiate due to inadequate shielding.  
> 
> Keep up the good work and let us know how it goes.  I
> might add that I am
> very interested in how well (or not) LEDs work in the
> floodlamp
> application, as I have already discovered that CFLs will
> not fit into my
> outdoor flood holders.  (The CFL floods are too fat in
> the area just above
> the screw base - that causes the lamp to not be able to set
> deeply enough
> into the socket to make contact.)   I have a
> pair of floods that are
> controlled by a photoeye and run each and every night form
> many hours
> (their main function is for security purposes).  I've
> gone through a lot of
> incandescent flood lamps in the 10 years I've lived at this
> QTH, even
> though I use lamps rated for 130 volt operation (on my 123
> volt mains).
> 
> 73, Dale
> WA9ENA   
> 
> 
> > [Original Message]
> > From: Craig <vk3he at yahoo.com>
> > To: <rfi at contesting.com>
> > Date: 1/23/2011 5:46:25
> > Subject: [RFI] LED Floodlight warning!!
> >
> > Hi All
> >
> > I  recently purchased this   new
> LED floodlight which has turned into a
> RFI nightmare. I just want to warn other hams about the
> huge amount of
> interference that these lights cause.
> >
> > My warning might also help those looking for
> interference sources around
> their neighborhood. LED floodlights are a disaster as big
> as bad plasma TV
> sets because the QRM level is so intense.
> >
> > These light are becoming popular because they can save
> you  a lot of
> money when compared  to running a normal 500 watt
> floodlight.
> >
> > I purchased several of these LED lights from various
> Ebay sellers. They
> all seem to originate from the same factory. They
> have   ratings from 10 to
> 50 watts. These lights are the ones that use a huge 
> large LED CHIP  rather
> than discreet individual LEDs
> >
> > Here is a link for one on Ebay, if it looks like this
> light avoid it.
> >
> >
> http://cgi.ebay.com/Lemansind-50W-White-LED-FloodLight-Wall-WashLight-/27069
> 4205505?pt=US_Flashlights&hash=item3f06a1c841
> >
> > A generic search for "LED floodlight" will show the
> many seller listings
> for these lights.
> >
> >
> > These light cause S9+20db hash right across the HF
> spectrum. Its the
> worst case of switch mode power supply noise that I have
> ever encountered.
> What also amazes me is the strength of this hash that could
> be detected
> from a long distance away, upto 1km away!
> >
> > I emailed the supplier,  and  he 
> advised me that he has paperwork which
> certifies the lights with  a C tick compliance. The
> paperwork is obviously
> fraudulent or he is just telling lies. Most of the 
> goods from China have 
> fake EMC compliance documents and certification ..
> >
> > When I opened the light, it had a small switch mode
> power supply module.
> This module does not appear to have any mains filter or
> other output
> filtering. The light is well shielded since the whole
> assembly is cast 
> aluminum. The power supply module is well shielded, hence
> the surprise at
> the rather high  noise levels.
> >
> >  I added a IEC filter which did nothing to reduce
> the noise level. I
> added 2 X Fair-rite FT240-31 ring cores on the power supply
> lead and this
> did not help either. I added 2  X #31  material
> cores onto the DC output
> and this did help, however the noise was still shocking.
> >
> >  I am going to try some capacitor bypassing and
> perhaps a professional 
> DC filter. So far the signs are not encouraging. The noise
> is still very
> intense. If I dont have luck I will just throw these lights
> in the bin.
> >
> > I have these lights in 10 20 and 50 watts and they all
> have the same
> interference levels. Its a real shock to turn on my 
> SDR receiver and see
> the waterfall go red(5/9+) from 1mhz to 30mhz! 
> >
> > Anyway avoid these lights like the  plague! 
> >
> > 73
> > Craig
> > VK3HE
> >
> >
> >
> >
> >       
> > _______________________________________________
> > RFI mailing list
> > RFI at contesting.com
> > http://lists.contesting.com/mailman/listinfo/rfi
> 
> 
> 