Light Bulb RF Noise

[Bill, KD0HG]

The color spectrum of the currently-available white LEDs is awful.
Don't know why, but they need to fix it. They hurt my eyes.

[Tom WA3KLR]

White LEDs are made 2 ways.  One is they use a blue LED and phosphors.  The other way is to use 3 LEDs - blue, green and red.

Below is a table that was in the November 16, 2006 issue of Electronic Design. The table comes from Cree, who is one of the leaders in producing compound semiconductors in the U.S.  LEDs are compound semiconductors.

Cree says that up to half of the Lumens from a light source is lost in the fixture design.  Hence the "Usable Output" column.  Their LEDS are the only ones on the list too.  (Since LEDs radiate as a hemispherical source, I wonder if we will have LED panels on our ceilings in the future.) 

Light-source Type   Luminous efficiency  Usable Ouput   Lifetime
                                   (Lm/W)              (Lm/W)        (hours)
incandescent lamps            17                 10 - 17         3000
halogen lamp                     20                 12 - 20       10,000
Cree XLamp 7090 XR           47                    47         >50,000
 (color and white)
T12 fluorescent lamp          60                  45 - 50       20,000
metal-halide lamp               70                   <40      5,000 - 15,000
Cree XLamp 7090 XR-E        71                   71           >50,000
      (white)
T8 fluorescent lamp             74                55 - 60        20,000
high-pressure sodium lamp     91                 <50     20,000 - 24,000
T5 fluorescent lamp            100                   80           20,000
low-pressure sodium lamp    120                65 - 70       18,000

[WA1GFZ]

When it isn't too cold out I don't mind if the kids leave lights on because they will heat up a room keeping the heat off so I take advantage of inefficient glow lamps. Flat panel LEDs sound cool for the ceiling.

[W3RSW]

Didn't we cover a lot of this just a week or so ago?I
I think the bulb components and transmitter build were mentioned before.  Ooooh, a hissy. - Uh, sorry. Well they say that's the second thing to go....

Umm, back on topic.

I suppose someone else here has already pulled one of these lamps apart to see what's inside. I cut my finger breaking the spiral glass tube, of course, in getting the d**n thing apart.

Jeeze, enough parts to build a radio!  Get three or four of the earlier models and you can have a complete shack.

In one 13 watt Phillips I see one transformer, one 33uf/200v 105deg.C electrolytic, 3 ea. 1n4005 diodes (if I read em right), 2 more switching diodes in small glass pkg., 2 Motorola T5N25E 3pin regulators or some sort of push pull osc. transistors w/ the mounting hole tabs cut off, bunches of 1kv square package capacitors and resistors, e.g., MKP 6n8k@1kv, and of course the printed circuit board measuring about 1.9 in. dia. The transformer complete with iron core appears to be 3 pins (ct) and 2 pins out, and measures abt. 0.7" high, 0.6 wide and 0.4 thick. Well that'd be good for a class E kilowatt : )

Can you imagine the infrastructure necessary to manufacture these things by the billions, the heavy metals containment, disposal processes that would be necessary if done in the supposedly 'green' perfect world desired.   Reminds me of all the sci-fi, "Cover the World with photoelectric cells power generator" scenarios. Not mentioned, of course is the following mass world pollution from kilotons of seminconductor doping elements, arsnic, etc.  Shadowing of the ground in great swaths also will have some pretty interesting effects.

  Yes another, not thought out at all, political solution containing many levels of unintended consequences.  Of course, all will self collapse long before we get to the one billion mark. Well, I stand corrected...  look at McDonalds.

[WA1GFZ]

yes but Bud provided real test data for conducted emissions he measured coming off the little crap generator.

[WD8BIL]

Well... today was a busy day but tomorrow(friday) will be quite a bit slower.
I shud be able to complete the LF, HF and VHF radiated scans. I'll probably do the 10Khz to 500Khz Telchordia CE scans too.

Stay tuned.

[W3RSW]

I agree. The spectral scans are scary and contain very relevant information for us.

Could you please also run scans on light dimmers?
 My dining room light and kitchen are controlled and similar to what was previously mentioned, they're the dirtiest RF thing in the house, way dirtier than Phillips bulbs.

BTW just holding my cell phone, even when only receiving, next to any radio, from bat. portable am/fm to my 75a2 generates one good mysterious SciFi sounding howl. Wonder what several million of these do the the radiosphere?  Talk about uncontrolled oscillators.

[Bill, KD0HG]

I'm sure they *could* manufacture quieter lights and dimmers, but the manufacturers aren't about to spend a penny more than the minimum needed to meet FCC requirements. It's the FCC that's the problem here, and they're not exactly under major public pressure to tighten the Part 15 rules...

[WA1GFZ]

The best way to eliminate dimmer noise is to never install them.

[AF9J]

Yes, part of the issue is that.  The other part of the issue, is the fact, that unless representative units sent for  compliance, are basically off of the production floor, you can't even say that these items generally meet Part 15.  Oftentimes, items that are sent for compliance are pre-production, prototypes, or first run items, that are either built to a higher standard than the typical production item, or with greater attention to quality than the typical production item. in order to pass certification or compliance.  I've seen this happen with samples or early runs we've received from some suppliers we've had quality issues with.  We've caught these issues with regular inspection and testing of items.  The FCC doesn't do this, due to manpower and the cost of what doing periodic recertifcation would be for them.  So, the CFAs we buy, could have some differences from those that were certified to Part 15, and not even be compliant.

Ellen - AF9J

I'm sure they *could* manufacture quieter lights and dimmers, but the manufacturers aren't about to spend a penny more than the minimum needed to meet FCC requirements. It's the FCC that's the problem here, and they're not exactly under major public pressure to tighten the Part 15 rules...

[Bill, KD0HG]

Ellen, that's almost a different issue. If the item is designed and manufactured in full compliance with current regulations, and causes harmful interference, it's an FCC regulatory issue. If the same item is made without regard to quality control and violates Part 15, then the device is unlawful to use and the manufacturer should held accountable, and should be required to recall the product.

For the subject at hand, it's speculation as to whether the devices are or are not in violation of Part 15 standards.

Perhaps this is a role the ARRL or someone civic-minded should take on; running tests as Bud is doing and reporting violators to the Feds...Because if we don't do that, who else will?

[WD8BIL]

Let's look at the first scan a minute.(as other scans are running right now)

FCC Part 15 requirements for this product would be the Class B limits for residential usage. The FCC frequencies start at 450Khz. The freqs below 450 are EN55022(Europe) standards.

At @700Khz, 1.2 and 1.4Mhz the data says this sample is a pass. By the skin of it's teeth.... but a pass non the less.

BUT... as was pointed out, at 40dbuv we have an S9 signal source pumping crap onto the AC lines.

Now.... 99% of the public will never become concerned with this. We, as radio enthusiasts(sp), become concerned because it effects our passion.

These exercises are not meant to lead a campaign but simply to add info to a previous discussion.

[k4kyv]

What about touch lamps and touch type wall switches?  Those things wipe out wide swaths of spectrum over extended distances with garbage, and the ones I have seen for sale at places like Lowe's and Home Cheapo are Chinese imports that do not even have the required FCC Part 15 tag attached.

[Bill, KD0HG]

Bud, then perhaps it's time the League (or other interested person(s)) should petition the FCC to change the Part 15 regulations as the lights can interfere with licensed radio services. As I previously wrote, there's no question that they could be engineered to be cleaner.

One aspect that bears more scrutiny is why some users are receiving unacceptable levels of interference from CFLs, while others, myself included, do not. I can hear our light dimmers to varying degrees, but I can't hear any hash -at all- from the dozen or so CFLs in use in my house.
Is it variations between brands, designs, grounding of the ham station or AC mains, ground loops, maybe the use of metallic conduit or the proximity of the ham antenna? I dunno.

[Bill, KD0HG]

What about touch lamps and touch type wall switches?  Those things wipe out wide swaths of spectrum over extended distances with garbage, and the ones I have seen for sale at places like Lowe's and Home Cheapo are Chinese imports that do not even have the required FCC Part 15 tag attached.

Then report the things and their retailers to the FCC, Don!

[WA3VJB]

We've got a half dozen CFLs in the house and out back, and I don't hear them either, and I've walked around with a portable just to sniffum.

I do hear the old style fluorescent tubes, they ARE nasty, and also an unfiltered dimmer on a floor lamp.  Wall dimmers are filtered Lutrons, vy quiet, with a good aluminum heat sink besides.

I've had a switched power supply, an old thing that used to run an old laptop, dirty as hell all over HF if you were within 5 feet.

[WD8BIL]

Actually Bill, Paul, that isn't too surprising. Below is the radiated emissions on this sample taken with an active rod antenna at 3 meters. Telchordia standard GR1089.

It's pretty broadbanded and mostly below say... 4Mhz. The field intensity is  relatively low. Most likely due to the lack of radiating structure within the bulb. Conductors and componants are just too small to form any significant fields.

I do have more detailed CE (conducted emissions) scans I'll post later tonite. They, on the other hand, are pretty interesting. CE is where we're gonna see the worst.

Oh.... the purple scan is a reference. It is the chamber ambient with the power to the lamp turned on but the lamp has been unscrewed.

And before anyone says anything about how noisy the chamber is....
The scan starts at +20dbuv because of the correction factors needed to comp. for the active rod antenna.

[AF9J]

Good point Bill, 

I agree.  But I guess I wonder how well they really enforce Part 15 Compliance.  Nevertheless, I stand corrected.

Ellen - AF9J

Ellen, that's almost a different issue. If the item is designed and manufactured in full compliance with current regulations, and causes harmful interference, it's an FCC regulatory issue. If the same item is made without regard to quality control and violates Part 15, then the device is unlawful to use and the manufacturer should held accountable, and should be required to recall the product.

For the subject at hand, it's speculation as to whether the devices are or are not in violation of Part 15 standards.

Perhaps this is a role the ARRL or someone civic-minded should take on; running tests as Bud is doing and reporting violators to the Feds...Because if we don't do that, who else will?

[WD8BIL]

OK... here's some more pictures.
Low frequency CE and band specific CE for 160, 75/80 and 40 meters.
Some of the spikes look very familiar to my receiver. Hmmmmm

[WD8BIL]

And 40 !

The CE scans were done with an Agilent 7404A EMI analyzer, Solar 200amp 50uh LISNs and an HP 10db transient limiter. The 10db attenuation from the limiter is compensated for in the results.

The RE scans were done on an Electrometric EMC30 EMI Receiver using an EMCO Active Vertical Rod antenna and 30 meters (standard) 9913 mil spec. Distance was 3 meters.

[WA1GFZ]

Bud,
Was that just the bulb or bulb plus 2 meters of cable on the bench?
Never in my QTH. That is pretty strapping for 3 m distance. That active rod is a nice ant we have a couple of them.
Part 15 that is nothing Hams would prefer UK DEF STAN limits...or 461e USAF external fixed wing.

[WD8BIL]

Frank.... the RE scans were done as we discussed earlier. The socket mounting plate covered the access hole in the turntable so I got some 2" copper tape and sealed the wires under the turntable/GP. That's just the lightbulb, socket and about 8" of wire.

[WD8BIL]

But I guess I wonder how well they really enforce Part 15 Compliance.  Nevertheless, I stand corrected.

As in any interference issue Ellen, they usually instruct the offending party to contact the complaintant and mutually work it out. The FCC usually won't step in until an impass is reached or the offending party flat out refuses to rectify the problem. They then can order a shutdown of the offending system until repaired and/or level fines.

In the telco industry we cannot market a power system without FCC Part 15 certification. One sample PRODUCTION system must be certified by a third party FCC certified lab. My lab is actually a pre-compliance lab specifically for our companies products. It's good to know our systems are compliant before we drop $60 - $100K for third party certification.

[AF9J]

I figured as much, that you were working in an RF Lab environment Bud.  But, what about companies like GE for instance, where RF emissions for many of the items they make, are incidental/unintended in nature.  In other words, they are not Telecom in nature as an industry. Since many of their items are made overseas, how well, do we really guarantee that these items (past the initial samples that were used to check Part 15 compliance), are meeting Part 15. The company I work for purchases some of it's stuff from overseas.  I won't lie, some of the quality is pretty nasty.  I remember a few comments made by Don stating about lamp dimmers sold at Home Depot, made in China, that were not marked as being Part 15 compliant.  Since they aren't marked as such, how do we know any effort was made to meet compliance with them.

It's 11:30 PM.  I've just spent the last frustrating 45 minutes, trying to null out my nasty hum, that's either from powerlines, or CFLs.  Even with an active noise cancelling antenna, the best I can do, is drop my noise from 10 over 9, to about S8.  I'm at a loss.  I don't know if it's a case of me living in a noise saturated environment (I have 2 significant sources: tons of CFLs; and powerlines [both the typical residential service lines, and 250 kV transfer lines]), in which case I'm out of luck, or I'm approaching the noise source issue wrong.

73,
Ellen - AF9J

[WA1GFZ]

Wow Bud, so that is RE right off the bulb. Part 15 spec limits are pretty high when you look at a receiver noise floor. I guess I will never use these in my house.
I hope an S9 signal doesn't radiate well enough off my neighbors wiring for me to hear it.
It would be interesting to lay out a couple meters of romex on the bench and see how high it goes then add a common mode choke (snap on core) to see the effect.